Background

[0001] Electronic displays which are implemented on a computer screen or other display device are rarely suitable for presenting and selecting represented patient information for homogeneous selection of potential patients in a clinical trial. Generation of the electronic displays used in presenting patient pool information may be improved by including various factors, including: adjusting presented axes on which the graphical representation of patients are presented; indicating homogeneous patients for ready selection by associating displayed and represented patients with patient matching index information in combination with integrating patient co-factor information; and, allowing the user to select on the electronic display, via the user interface or client device display, a homogeneous clinical trial patient population. Such integration of relevant patient information within the generated electronic display however is typically not provided nor included in past systems.

[0002] Such systems however may be beneficial for use in clinical trials. Clinical trials are typically performed to evaluate new medical treatments. In order to conduct a clinical trial for a particular health problem, medical researchers must select participants for the trial that have that particular health problem. In selecting the trial participants, the researchers analyze medical records of numerous potential participants that are often scattered across disparate health systems and IT platforms. In many cases, this is a time-consuming manual procedure.

[0003] In some clinical trials, automated systems can initially screen potential participants. When an automated system is employed, the criteria for selection is initially based on a primary diagnostic code assigned to the potential participants, such as an International Classification of Diseases (ICD) code. Participants selected based on a primary diagnostic code may be subsequently filtered based on one or more binary exclusion criteria. For example, certain participants may be filtered out if they fail to possess certain disease stages, specific cell markers, and/or specific gene sequences. However, the final participants may still represent an overly heterogeneous population that may introduce confounding variables that may skew the trial conclusions and lead to false acceptance or rejection of a new treatment. False acceptance or rejection of a new treatment may be costly to a creator of the new treatment and/or may stifle good public health.

Summary

[0004] This specification is directed generally to improving a patient presentation system working in combination with a client device having a display by incorporating with the client device and display relevant and adjusted patient information to ensure the display appropriately sets forth patient information for homogeneous patient population selection. The patient presentation system may generate an electronic display based upon determined axes whose limits are dependent upon determined patient matching indices and patient co- factor indices thereby grouping graphical representations of patients for more ready determination and selection of homogeneous clinical trial patient pools. The patient presentation system may present a large pool of patient information to the user on a display on the client device wherein the pool of patient information is presented to provide more ready grouping and homogeneous selection of patients for use in clinical trials. The represented patient information incorporates patient matching indices and patient co-factor indices on the client device display or graphical user interface. Generating and presenting the electronic display on the client device to incorporate such patient information improves the display for selection of homogeneous boundary conditions. The display is generated so as to transform graphical representations of patient information for use in patient pool selection of a clinical trial. The electronic display is generated based on the transformed patient information for homogenous patient population through incorporation of patent matching indices and patient co-factor indices. The transformed patient data may include at least a co-factor index for each of one or more patients. The co-factor index for a patient for a clinical trial generally represents a degree of impact of one or more secondary factors of the patient on the clinical trial. In some implementations, the co-factor index for a patient for a clinical trial may be calculated based at least in part on values of one or more selected medical test results of the patient, wherein the selected medical test results exclude at least those test results that define a primary diagnosis for the clinical trial. In some implementations, the transformed patient data may also include a patient matching index for each of one or more patients. The patient matching index for a patient for a clinical trial generally represents at least one of an extent of the patient's primary diagnosis for the clinical trial and the degree of the patient's conformance to one or more additional selection criteria for the clinical trial. By creating the electronic display to include such factors, prompt verification and selection of homogeneous populations is available.

[0005] The patient presentation system in various embodiments presents a generated electronic display on a client device graphical user interface or display and give rise to the technical advantage of generating and electronically providing an electronic display to a user that eases the cognitive burden on the user in selecting a clinical trial set that includes a satisfactorily homogenous population of patients. The electronic display which is presented on the client device for review and interaction by the user positions graphical

representations of candidates for the clinical trial set based on a patient matching index and/or a patient co-factor index determined for those candidates. Such indices are based on the detailed techniques set forth herein. The positioning of particular patient information on the generated electronic display based on the patient matching index and/or the co- factor index enables the user to efficiently ascertain candidates that are homogenous and/or ascertain candidates that are non-homogenous. Moreover, implementations of the patient presentation technology set forth herein give rise to the technical advantage of user interaction (via one or more user interface input devices ) with the electronic display to select a clinical trial subset while easing the cognitive burden on the user. The positioning of patient information in the electronic display as presented based on the patient matching index and/or co-factor index clusters graphical representations of candidates for a clinical trial based on the patient matching index and/or co-factor index. Accordingly, a single bounded selection area such as circle, rectangle, etc. will encompass graphical

representations for those candidates desired to be selected for a trial without

simultaneously encompassing graphical representations for undesired candidates.

[0006] The present disclosure is therefore directed towards technical features for generation of an electronical display which defines positional presentation of graphical representations of candidates. The technical problems associated with prior generation of representations of candidates prevented the efficient recognition and selection of homogenous clinical trial candidates. Such technical problems and limitations associated with past systems centered on a lack of recognition for initial patient diagnosis and conformance of such diagnosis and their relevancy to the clinical trial. These technical problems as well prevented recognition of any increased likelihood that other health issues would impact individual outcomes within the clinical trial. Such technical problems in past systems therefore prevented the ready presentation and selection of homogenous patient populations for clinical trial selection of patient pools and inhibited generation and interaction of useful electronic displays.

[0007] The various embodiments of the presently described patient presentation system integrated with the display of the user device provides the technical solution to the above stated problems. Namely, generated electronic display on a client device graphical user interface eases the users' ability in selecting homogeneous patient populations for clinical trials by incorporating in the generated and presented electronic display on a user device, relevant patient matching index and co-factor index of the patient pool. The adjusted presentation of candidate representation takes into account both relevancy of initial diagnosis, by use of the patient matching index, as well as co-morbidity issues by use of the co-factor index, thus presenting grouping of patients which may be more preferable for the clinical trial set. Placement and position of candidate representation on the various defined axes are based on the detailed techniques set forth herein. The further technical solution of integrating within the electronic display the position of particular patient information enables the user to efficiently ascertain candidates that are homogenous and/or ascertain candidates that are non-homogenous and provide the technical advantage of user interaction with the display to select an accurate and effective clinical trial subset. Such technical advantage further allows the user to more readily recognize and select such homogeneous pool while taking into account such other factors set out in the various embodiments.

[0008] In various embodiments, other advantages of the patient presentation system and associated display includes significant reduction in utilization of network resources, memory and storage demands. Potential patient pool information may incorporate substantially large amounts of disparate patient information. The various embodiments of the patient presentation system may include implementation on a remote server thereby allowing for the definition and placement of the graphical representation of candidate patients on a background server. Such determination for placement of the graphical representation of candidates may thereby be generated while taking into account these la rge amounts of disparate patient information so as to include related co-factor index and patient matching index information built within the patient representations. Thus, in various embodiments, the patient presentation system may distill large amounts of patient information into, for example, a patient identifier, a co-factor index and a patient matching index or associated data points for rendering on the electronic display. It is then only necessary to send to the client device and/or render at the client device such simplified and distilled information. As a result, less data could be sent from server to client to enable health professionals to make informed decisions on a tria l set, less com puter resources may be used in storing such data, and/or less resources may be used in rendering the display based on such patient information.

[0009] In some implementations, an electronic display is generated based at least on co- factor indexes for potential patients for a clinical trial and ena bles improved selection of a subset of those patients for inclusion in the clinical trial. I n some of those implementations, the electronic display may increase the likelihood of selection of a satisfactorily

homogenous population of patients for a clinical trial and may result in selection of a population that is more homogenous tha n populations selected utilizing other techniques. The more homogenous population may increase the accuracy of clinical trial results, thereby providing less false acceptances and/or rejections of a new treatment and/or promoting good public health.

[0010] In some implementations, an electronic display is generated based at least on multiple co-factor indexes for one or more given patients, with each of the co-factor indexes being based on secondary factors of a patient for a distinct time period. In some of those im plementations, the electronic display may provide a visual representation of changes in the patients' health (e.g., changes during a clinical trial) that may assist a medical professional's a nd/or the patients' understanding of those changes.

[0011] In some implementations, an electronic notification is generated based at least on calculated co-factor indexes for one or more patient identifiers for multiple time periods satisfying a threshold value. The electronic notification may be provided in various electronic forms and may be provided to the one or more patients and/or one or more health professionals via one or more client devices. Generally, the electronic notification may inform the patients and/or health professionals that one or more of the calculated co- factor indexes are at levels that indicate further action may be needed.

[0012] Other implementations may include one or more non-transitory computer readable storage media storing instructions executable by a processor to perform a method such as one or more of the methods described above. Yet another implementation may include a system including memory and one or more processors operable to execute instructions, stored in the memory, to perform a method such as one or more of the methods described above.

[0013] It should be appreciated that all combinations of the foregoing concepts and additional concepts described in greater detail herein are contemplated as being part of the subject matter disclosed herein. For example, all combinations of claimed subject matter appearing at the end of this disclosure are contemplated as being part of the subject matter disclosed herein.

Brief Description of the Drawings

[0014] FIG. 1 illustrates an example environment in which patient data may be transformed and an electronic display generated based on the transformed patient data.

[0015] FIG. 2 illustrates an example of selecting a set of patient identifiers based on clinical trial criteria, calculating patient matching indexes and co-factor indexes for the patient identifiers of the set, generating an electronic display of graphical representations of the patient identifiers based on the calculated values, and selecting a clinical trial set based on an electronic selection received responsive to the electronic display.

[0016] FIG. 3A illustrates an example of an electronic display of graphical representations of patient identifiers that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifiers.

[0017] FIG. 3B illustrates the example electronic display of FIG. 3A, with additional information being displayed for a patient identifier associated with one of the graphical representations.

[0018] FIG. 3C illustrates an example selection of the graphical representations of the electronic display of FIG. 3A.

[0019] FIG. 4 is a flow chart illustrating an example method of transforming patient data, generating an electronic display based on the transformed data, and selecting a trial set based on an electronic selection received responsive to the electronic display.

[0020] FIG. 5 is a flow chart illustrating an example method of calculating a regression correlation coefficient for a medical test result for a primary diagnosis.

[0021] FIG. 6 is a flow chart illustrating an example method of calculating a contribution to a co-factor index for a patient identifier based on a value of a medical test result for the patient identifier.

[0022] FIG. 7 illustrates an example of calculating patient matching indexes and co-factor indexes for one or more patient identifiers for multiple time periods, and generating an electronic display of a graphical representation of the patient identifiers for the multiple time periods based on the calculated values.

[0023] FIG. 8A illustrates an example of an electronic display of a graphical

representation of a patient identifier that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifier for multiple time periods.

[0024] FIG. 8B illustrates an example of an electronic display of graphical representations of multiple patient identifiers that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifiers for multiple time periods.

[0025] FIG. 8C illustrates another example of an electronic display of graphical representations of multiple patient identifiers that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifiers for multiple time periods.

[0026] FIG. 9 is a flow chart illustrating an example method of generating an electronic display of a graphical representation of one or more patient identifiers based on calculated patient matching indexes and co-factor indexes for the patient identifiers for multiple time periods.

[0027] FIG. 10 illustrates an example architecture of a computer system.

Detailed Description

[0028] FIG. 1 illustrates an example of a patient presentation system for patient selection and presentation in which patient information and representations may be transformed and an electronic display generated based on the transformed patient data. The example environment of FIG. 1 includes a patient presentation system 120, a client device 106, a patient data processing system 130, medical center systems 103a-n, and a patient data database 154. The patient presentation system 120 and/or other components of the example environment may be implemented in one or more computers that communicate, for example, through one or more networks.

[0029] The patient presentation system 120 is an example system in which the systems, components, and techniques described herein may be implemented and/or with which systems, components, and techniques described herein may interface. One or more components of the patient presentation system 120 and/or the patient data processing system 130 may be incorporated in a single system in some implementations. Also, in some implementations one or more components of the patient presentation system 120 may be incorporated on the client device 106. For example, all or aspects of display generation engine 124 and/or trial set selection engine 125 may be incorporated on the client device 106.

[0030] Generally, the patient data processing system 130 collects electronic patient data from a plurality of medical center systems 103a-n and/or other sources, optionally normalizes and/or otherwise alters one or more aspects of the patient data, and stores the optionally altered patient data in a patient data database 154. The patient data processing system 130 may collect electronic patient data from various sources such as medical center systems 103a-n that may be associated with hospitals, doctor offices, universities, and/or health facilities focused primarily on identifying candidates for clinical trials. In some implementations, the patient data processing system 130 may receive patient data via one or more standard transfer and/or data protocols such as Health Level-7 (HL7) or Admit Discharge Transfer (ADT). Patient data received by patient data processing system 130 may include, for each a plurality of patients, an optionally anonymized patient identifier, demographic information, health provider(s) information, information for one or more patient diagnoses, medical test results information, and/or other data. Additional description of some of the data that may be included in patient data for one or more patients is provided herein.

[0031] In some implementations, the patient data processing system 130 stores collected patient data in patient data database 154 and assigns a unique patient identifier to each patient's patient data. In some of those implementations, the patient data processing system 130 may generate a unique patient identifier for patient data of a patient based on one or more of a date associated with the patient data of the patient (e.g., date of initial collection of the patient data by patient data processing system 130, date of initial diagnosis), a value associated with a doctor or other health professional indicated by the patient data of the patient, and/or a value associated with the medical center system(s) 103a-n that provided the patient data. The patient data processing system 130 may also encrypt the patient data of patient data database 154. For example, the patient data processing system 130 may assign a cryptographic key to a patient's patient data, and provide the patient identifier and electronically provide the key to the medical center system(s) 103a-n that provided the patient data and/or to the patient. I n some

im plementations, the patient data processing system 130 may validate the identity of each patient for which patient data is collected, and determine whether that patient already exists in the patient data database 154 prior to creating a new entry in the patient data database 154.

[0032] In this specification, the term "database" will be used broadly to refer to any electronic collection of data. The data of the database does not need to be structured in any particular way, or structured at all, and it can be stored on storage devices in one or more locations. Thus, for example, the patient data database 154 may include multiple collections of data, each of which may be organized and accessed differently. Also, in this specification, the term "entry" will be used broadly to refer to any mapping of a plurality of associated information items. A single entry need not be present in a single storage device and may include pointers or other indications of information items that may be present in unique segments of a storage device and/or on other storage devices. For example, an entry that identifies a patient identifier and patient data for the patient identifier in patient data database 154 may include multiple nodes mapped to one another, with one or more nodes including a pointer to another information item that may be present in another data structure and/or another storage medium.

[0033] In some implementations, the patient data processing system 130 may normalize one or more values associated with medical test results information and/or other information of the patient data. Normalization of values associated with medical test results information may, inter alia, ameliorate problems associated with different laboratories utilizing different test equipment and/or having different calibrations of test equipment. For example, for a first laboratory, "normal" values for a particular medical test result may be from 100 to 120, whereas for a second laboratory, "normal" values for that particular medical test result may be from 108 to 125. In some implementations, the patient data processing system 130 may normalize values associated with a medical test result by calculating a z-score for each of those values. In some of those implementations, the z-score for a patient's value for a medical test result may be calculated based on a formula such as: z-score = (patient's value for medical test result) - (mean of the value for the medical test result for a population of patients / standard deviation of the value for the medical test result for the population of patients).

[0034] In some implementations, the patient data processing system 130 may not normalize values for one or more medical test results. In some of those implementations, the patient presentation system 120 may optionally normalize values for one or more of those medical test results. Also, it is noted that some medical test results may be associated with values that will not be normalized, such as values that identify presence or absence of specific cell markers and/or gene sequences for a patient that have been identified utilizing one or more medical tests.

[0035] In some implementations, the patient data processing system 130 may index the patient data of patient data database 154 to enable more efficient searching of the patient data in identifying patient identifiers that match primary diagnosis criteria and/or other criteria for a clinical trial. For example, patient data database 154 may include a plurality of entries, with each entry being associated with a patient identifier and including patient data for that patient identifier. The patient data processing system 130 may generate an index of the entries based on one or more properties of the entries. For instance, an index may include one or more values associated with each entry, wherein the values each indicate at least one diagnosis associated with a respective entry.

[0036] Generally, the patient presentation system 120 transforms patient data of one or more patient identifiers and generates an electronic display based on the transformed patient data. In some implementations, the patient presentation system 120 selects a set of patient identifiers from the patient data database 154 based on mandatory clinical trial criteria, calculates patient matching indexes and co-factor indexes for the patient identifiers of the set, and generates an electronic display of graphical representations of the patient identifiers of the set based on the calculated va lues. The patient presentation system 120 may provide the electronic display to client device 106 for presentation to a user. The user may utilize the electronic display to ascertain which of the patient identifiers of the set are most appropriate for inclusion in a clinical tria l set. The user may further utilize the client device 106 to provide a selection of those patient identifiers to the patient presentation system 120 and the patient presentation system 120 may select those patient identifiers for inclusion in the clinical tria l set.

[0037] In some implementations, the patient presentation system 120 additionally and/or a lternatively generates an electronic display based at least on multiple co-factor indexes for one or more patients, with each of the co-factor indexes being based on secondary factors of the patient for a distinct time period. I n some of those

im plementations, the patient presentation system 120 may generate the electronic display to provide a visual representation of changes in the patients' health (e.g., changes during a clinical trial) that may assist a medical professional's and/or the patients' understanding of those changes.

[0038] In various implementations patient presentation system 120 may include a trial criteria matching engine 121, a patient matching index engine 122, a co-factor index engine 123, a display generation engine 124, and/or a trial set selection engine 125. I n some im plementations, all or aspects of engines 122, 123, 124, and/or 125 may be omitted. In some implementations, all or aspects of engines 122, 123, 124, and/or 125 may be combined. In some implementations, all or aspects of engines 122, 123, 124, and/or 125 may be implemented in a com ponent that is separate from patient presentation system 120, such as client device 106.

[0039] Generally, tria l criteria matching engine 121 identifies one or more mandatory clinical trial criteria and selects a set of patient identifiers from patient data database 154 that have associated patient data that indicate that mandatory clinical trial criteria.

Accordingly, the trial criteria matching engine 121 may identify an initial set of patient identifiers that are potential candidates for a clinical trial. The mandatory clinical trial criteria identify at least a primary diagnosis defined for the clinical tria l. A primary diagnosis may be, for example, "cancer", "colon cancer", "thyroid cancer", "chronic kidney disease" (CKD), or "diabetes". In some implementations, the mandatory clinical trial criteria may further identify an extent of a primary diagnosis such as one or more specific stages of a prima ry diagnosis cancer (e.g., a stage defined by tumor size, lymph node involvement, and/or metastasis). As another example, a primary diagnosis may be, for example, "CKD", and additional mandatory clinical trial criteria may further define a specific stage of the CKD such as a subset of stages 0 to V. In some implementations, the mandatory clinical trial criteria may also include one or more criteria that are not directly tied to the primary diagnosis. For example, other mandatory clinical criteria may include gender criteria, age criteria (e.g., a particular age range), additional diagnosis criteria (e.g., an additional diagnosis that is distinct from the primary diagnosis), etc.

[0040] In some implementations, the mandatory clinical trial criteria may be identified based on input from client device 106 and/or from another source. In some

im plementations, the mandatory clinical trial criteria may be provided as input that is in a form to enable direct com parison to patient data of patient data database 154. For example, the input may define a primary diagnosis based on an I nternational Classification of Diseases (ICD) code and the patient data may a lso define diagnoses based on the ICD code.

[0041] In some other implementations, the mandatory clinical trial criteria may be provided as input that is in a form that does not enable direct comparison to patient data of patient data database 154. I n some of those implementations, the trial criteria matching engine 121 may convert the input and/or the patient data to enable a determination as to whether patient data indicates the mandatory clinica l trial criteria. For example, the input may include one or more words that define the mandatory clinical trial criteria and the trial criteria matching engine 121 may access a data structure that maps each of one or more ICD codes to one or more words that describe the ICD codes. Based on the mapping, the trial criteria matching engine 121 may determine an ICD code that matches the word of the input and compare that ICD code to patient data of patient data database 154.

[0042] Generally, patient matching index engine 122 calculates patient matching indexes for patient identifiers. For example, patient matching index engine 122 may calculate a patient matching index for each of the patient identifiers of the set identified by trial criteria matching engine 121. The patient matching index engine 122 calculates a patient matching index for a given patient identifier based on at least one of: an extent of the primary diagnosis associated with the given patient identifier; and conformance of the patient data associated with the given patient identifier to one or more additional selection criteria defined for the clinical trial. The one or more additional selection criteria may be

"mandatory" selection criteria and/or other desired (but non-mandatory) selection criteria. For example, additional selection criteria may include one or more of: age criteria, ethnicity criteria, height criteria, weight criteria, location criteria, compound allergy criteria, education level criteria, and involvement in another clinical trial criteria.

[0043] As one example of calculating a patient matching index, the primary diagnosis for a clinical trial may be "thyroid cancer" and the patient matching index for a patient identifier may be calculated based on an extent of the thyroid cancer as indicated by patient data for that patient identifier. For example, in some implementations the patient matching index may be influenced a first degree for stage III thyroid cancer and influenced a second degree (unique from the first degree) for stage II thyroid cancer.

[0044] As another example, the primary diagnosis for a clinical trial may be "thyroid cancer" and additional selection criteria defined for the clinical trial may include the following criteria: an age range of 30-45, a gender of male, presence of a particular first tissue marker associated with the cancer, absence of a particular second tissue marker associated with the cancer, a weight of greater than 120 pounds, and a height of greater than 150 centimeters. The patient matching index for a patient identifier may be calculated based on an extent the selection criteria are satisfied as indicated by patient data for that patient identifier. For example, in some implementations the patient matching index may be influenced by a greater degree if five of the six aforementioned selection criteria are satisfied than if only three of the six aforementioned selection criteria are satisfied.

[0045] Generally, co-factor index engine 123 calculates co-factor indexes for patient identifiers. For example, co-factor index engine 123 may calculate a co-factor index for each of the patient identifiers of the set identified by trial criteria matching engine 121. The co- factor index engine 123 calculates a co-factor index for a given patient identifier based at least in pa rt on one or more test values, of the patient data associated with the given patient identifier, for one or more selected medical test results. Those selected medical test results (whose values on which the co-factor index may be based at least in part) for a clinical trial exclude medical test results that define the primary diagnosis for the clinical trial and optionally exclude medical test results that define any additional ma ndatory criteria for the clinical trial. For example, where the primary diagnosis is thyroid cancer and no other additional mandatory criteria are present, the values may be based on medical test results that indicate one or more of: complete blood counts with white cell differential, serum electrolytes, liver profile enzymes, meta bolic study values, estimated glomerular filtration rate, selected tumor markers, NCI (National Cancer I nstitute) identified genetic tumor ma rkers, , lipid and tri-glyceride values, total cholesterol and ratio high-density lipoprotein (HDL) & low-density lipoprotein (LDL), blood pressure, body mass index, waist circumference, and/or patient health questionnaire-9 (PHQ.-9) results. In some

im plementations, and as described herein, the test values may be normalized in the patient data of patient data database 154 and/or may be normalized by the patient presentation system 120.

[0046] The co-factor index engine 123 may, for each of the test values on which the co- factor index is based, identify a regression correlation co-efficient that indicates a statistically calculated historical impact of the test value on a medical condition indicated by the primary diagnosis. For example, for Medical Condition 1, regression analysis of a corpus of patients' values associated with Medical Test Result 1 may determine a regression correlation coefficient of 0.01; whereas regression analysis of a corpus of patients values associated with Medica l Test Result 2 may determine a regression correlation coefficient of 0.12. The correlation co-efficients may be utilized in calculating the co-factor index for a patient. For exa mple, assume values for Medical Test Results 1 and 2 are utilized in calculating a co-factor index and that a patient identifier is associated with a normalized value of 1 for Medical Test Result 1 and a normalized value of 2 for Medical Test Result 2. Further assume the regression correlation coefficients of the previous example (0.01 for Medical Test Result 1 and 0.12 for medical Test Result 2). I n some of those

im plementations, the co-factor index for the patient identifier may be determined based on weighting the value "1" by "0.01" and weighting the value "2" by "0.12". For instance, the co-factor index for the patient identifier may be calculated based on the following equation:

Co-factor index= (1)*(0.01) + (2)*(.12) + (Medical Test Result 3 value for patient

identifier)*(regression correlation co-efficient for Medical Test Result 3 for Medical

Condition 1) + ... (Medical Test Result n value for patient identifier)*(regression correlation co-efficient for Medical Test Result n for Medical Condition 1).

[0047] In some implementations, an identified regression correlation co-efficient may indicate a statistically calculated historical impact of the test value on a medical condition indicated by the primary diagnosis, given an extent of the primary diagnosis and/or other factors. For example, for Medical Condition 1 Stage I, regression analysis of a corpus of patients' values associated with Medical Test Result 1 may determine a regression correlation coefficient of 0.02; whereas regression analysis for Medical Condition 1 Stage I I (a more advanced stage) of a corpus of patients values associated with Medical Test Result 1 may determine a regression correlation coefficient of 0.15. Such regression correlation coefficients may be utilized in calculating the co-factor indexes for patients.

[0048] Generally, the display generation engine 124 generates a display of one or more graphical representations, wherein each of the graphical representations is associated with one of the patient identifiers. The display generation engine 124 positions the graphical representations in the display based on their associated patient matching indexes and/or co-factor indexes. For example, the display generation engine 124 may position a graphical representation of a patient identifier along a first axis based on the patient matching index for the patient identifier and along a second axis based on the co-factor index for the patient identifier. For instance, the first axis may be a horizontal axis and the second axis may be a vertical axis.

[0049] As described herein, in some implementations the patient presentation system 120 is utilized to enable improved selection of a subset of potentia l patients for inclusion in a clinica l trial. In some of those implementations, the display generation engine 124 may generate a display of graphical representations of patient identifiers associated with the potential patients and may position each of the graphical representations based on associated calculated patient matching indexes and co-factor indexes.

[0050] As also described herein, in some implementations patient presentation system 120 is additionally and/or alternatively utilized to provide a visual representation of changes in one or more patients' health to assist a medical professional's and/or the patient's understanding of those changes. I n some of those implementations, the display generation engine 124 may generate an electronic display of graphical representations of patient identifiers associated with those patients, and may position each of the graphical representations based on associated calculated patient matching indexes and co-factor indexes associated with multiple time periods. For example, a graphical representation of a pa rticular patient identifier may be at a first position based on patient matching and co- factor indexes for a first time period and the graphical representation of the particular patient identifier may be at a second position based on patient matching and co-factor indexes for a second time period. The graphical representation may be displayed at the first position and the second position simultaneously and/or non-simultaneously. Also, the graphical representation may be exactly the same at the first and second position in some im plementations, or may graphically vary in one or more respects in other implementations.

[0051] Generally, in implementations where the patient presentation system 120 is utilized to enable improved selection of a subset of potential patients for inclusion in a clinical trial set, the trial set selection engine 125 receives input responsive to a display generated by the display generation engine 124 and utilizes the input to select the subset of potential patients to include in the clinical tria l set. For example, the display generation engine 124 may generate a display of graphica l representations of patient identifiers associated with a set of potential patients a nd may position each of the graphical representations based on associated calculated patient matching indexes and co-factor indexes. Responsive to viewing the display, a user may utilize client device 106 to provide electronic input to the trial set selection engine 125 that indicates a subset of one or more of those patient identifiers— and the trial set selection engine 125 may utilize that input to select the clinical trial set. For example, a user may select one or more graphical

representations to include and/or exclude from a clinical trial set via interaction with the electronic display utilizing one or more input devices (e.g., a mouse, keyboard, touch screen, or audio input). Based on the selection, the trial set selection engine 125 may determine one or more patient identifiers to include in a clinical trial set and/or to exclude from a clinical trial set. An indication of the selected clinical trial set may be stored in one or more storage mediums {e.g., patient data database 154) and/or presented to one or more users {e.g., via client device 106). The selected clinical trial set may be subsequently utilized in conducting the clinical trial.

[0052] A user may interact with the patient presentation system 120 via the client device 106. While the user may operate a plurality of computing devices, for the sake of brevity, examples described in this disclosure will focus on the user operating client device 106. Moreover, while multiple users may interact with the patient presentation system 120 via multiple client devices, for the sake of brevity, examples described in this disclosure will focus on a single user operating the client device 106. The client device 106 may be a computer coupled to the patient presentation system 120 through one or more networks 101 such as a local area network (LAN) or wide area network (WAN) {e.g., the Internet). The client device 106 may be, for example, a desktop computing device, a laptop computing device, a tablet computing device, and/or a mobile phone computing device. Additional and/or alternative client devices may be provided.

[0053] The client device 106 includes one or more applications to facilitate the sending and receiving of data over a network, to enable presentation {e.g., display) of data received from the patient presentation system 120, and/or enable data to be sent to the patient presentation system 120 {e.g., the submission of criteria for a clinical trial, selections of patient identifiers, feedback related to an electronic display). For example, the client device 106 may execute one or more applications, such as a browser or stand-alone application, that may render one or more of the electronic displays described herein and/or that may receive input from one or more user interface input devices of the client device 106 and provide data to the patient presentation system 120 based on such input. Other computer devices may submit data to the patient presentation system 120 such as additional client devices and/or one or more servers that may provide primary criteria for a clinical trial to the patient presentation system 120. For brevity, however, certain examples are described in the context of the client device 106.

[0054] The components of the example environment of FIG. 1 may each include memory for storage of data and software applications, a processor for accessing data and executing applications, and components that facilitate communication over a network. In some implementations, such components may include hardware that shares one or more characteristics with the example computer system that is illustrated in FIG. 10. The operations performed by one or more components of the example environment may optionally be distributed across multiple com puter systems. For example, the steps performed by the patient presentation system 120 may be performed via one or more computer programs running on one or more servers in one or more locations that are coupled to each other through a network.

[0055] FIG. 2 illustrates an example of selecting a set of patient identifiers based on clinical trial criteria, calculating patient matching indexes and co-factor indexes for the patient identifiers of the set, generating an electronic display of graphica l representations of the patient identifiers based on the calculated values, and selecting a trial set based on an electronic selection received responsive to the electronic display.

[0056] To aid in explaining the example of FIG. 2, it will be described in the context of an example clinical trial focused on Stage I I and Stage II I thyroid cancer. Reference will also be made to FIGS. 3A-3C to further aid in explaining the example of FIG. 2.

[0057] Trial criteria matching engine 121 receives clinical trial criteria 102. In some im plementations, the clinical trial criteria 102 may be defined based on input from a user via a client device 106. In some implementations, the clinical trial criteria 102 may be criteria predefined by one or more organizations such as the National Cancer Institute. The clinical trial criteria 102 define at least one or more mandatory criteria that includes at least a prima ry diagnosis for the clinical trial. For example, the clinical trial criteria 102 may include, as a primary diagnosis criteria, an ICD value that indicates "thyroid cancer". I n some implementations, additional mandatory clinical criteria may be included such as criteria that define one or more extents of the "thyroid cancer" primary diagnosis (e.g., Stage I I or Stage II I) and/or other criteria (e.g., minim um age, minimum weight). In some im plementations, the clinical trial criteria also include other desired (but non-mandatory) selection criteria.

[0058] The trial criteria matching engine 121 uses the mandatory criteria to select a set of patient identifiers from patient data database 154 that have associated patient data matching the mandatory criteria. For example, the mandatory criteria may be an ICD value that indicates thyroid cancer and the trial criteria matching engine 121 may select a set of patient identifiers based on those patient identifiers being associated with current patient data indicating the ICD value that indicates thyroid cancer.

[0059] The trial criteria matching engine 121 provides the set of patient identifiers to the patient matching index engine 122. The patient matching index engine 122 calculates patient matching indexes for the patient identifiers of the set. Each patient matching index may be based on at least one of: a most recent extent of the primary diagnosis as indicated by patient data of a respective patient identifier; and conformance of one or more additional selection criteria defined for the clinical trial to patient data of a respective patient identifier.

[0060] For example, assume the clinical trial criteria 102 defines a primary diagnosis of thyroid cancer and additional mandatory criteria that indicates an extent of the thyroid cancer is either Stage II or Stage III. Further assume the clinical trial criteria 102 defines the following six additional non-mandatory selection criteria of: an age range of 30-45, residence within 50 miles of a testing facility, a particular first tissue marker associated with the cancer, lack of a particular second tissue marker associated with the cancer, a weight of greater than 120pounds, and a height of greater than 150 centimeters. In some of those implementations, the patient matching index engine 122 may calculate the patient matching indexes by assigning a first value for an extent of Stage II, a second value for an extent of Stage III, and a third value for each of the six additional non-mandatory selection criteria; and summing, for each patient identifier, the values for each criteria indicated by the associated patient data.

[0061] As one example, assume Stage II has a value of "20", Stage III has a value of "30", and each of the six additional non-mandatory selection criteria has a value of "1.5". For such an example, a patient who has Stage III thyroid cancer and meets all six additional non- mandatory selection criteria would have a patient matching index of 39. A patient who has Stage III thyroid cancer but fails to meet any of the six additional non-mandatory selection criteria would have a patient matching index of 30.

[0062] The patient matching index engine 122 provides the set of patient identifiers and patient matching indexes to the co-factor index engine 123. The co-factor index engine 123 calculates co-factor indexes for the patient identifiers of the set identified by trial criteria matching engine 121. The co-factor index engine 123 calculates a co-factor index for a given patient identifier based at least in part on one or more test values, of the patient data associated with the given patient identifier, for one or more selected medical tests. Those selected medical tests (whose values on which the co-factor index is based at least in part) for a clinical trial exclude medical tests that define the primary diagnosis for the clinical trial and optionally exclude medical tests that define any additional mandatory criteria for the clinical trial and/or any additional selection criteria for the clinical trial.

[0063] As one example, assume the primary diagnosis is thyroid cancer and that the only other mandatory trial criteria is presence of either a Stage I or Stage II extent of the thyroid cancer. The co-factor index engine 123 may calculate the co-factor index based on values of medical test results that indicate: complete blood counts with white cell differential, serum electrolytes, liver profile enzymes, metabolic study values, estimated glomerular filtration rate, selected tumor markers, genetic markers, body mass index, and waist circumference. Where multiple values are present for a particular medical test result for a particular patient identifier, the most recent in time value may optionally be utilized for that patient identifier. As one example, for Patient Identifier 1 assume the following vector of values for the aforementioned medical test results: {0, 0, 1, 7, 0, 0, 0, 1, 1}. It is noted in the preceding list that for selected tumor markers and genetic markers "0" indicates lack of presence of those markers and "1" indicates presence. The other values may be normalized values. For example, the values may be normalized by patient data processing system 130 as described herein. Further assume that Patient Identifier 1 has patient data indicating Stage II thyroid cancer and that the following vector of regression correlation coefficients have been determined for Stage II thyroid cancer for the aforementioned medical test results: {0.01, 0.9, 0.15, 0.81, 0.03, 0.2, 0.01, 0.5, 0.5}. As described herein, the regression correlation coefficients are statistically calculated based on historical data and each generally indicate the calculated impact, for a particular primary diagnosis (and optionally a particular stage of the primary diagnosis and/or other factor(s)), of values associated with a respective medical test result. The co-factor index engine 123 may calculate the co-factor index for Patient Identifier 1 based on the dot product of the two vectors. For example, for the immediately preceding vectors, the co-factor index for Patient Identifier 1 may be 6.82.

[0064] The co-factor index engine 123 provides the set of patient identifiers and associated patient matching indexes and co-factor indexes to the display generation engine 124. In some implementations, the patient matching index engine 122 may provide the patient matching indexes to the display generation engine 124 directly. The display generation engine 124 may generate a display of graphical representations of patient identifiers associated with the subset of potential patients and may position each of the graphical representations in the display based on associated calculated patient matching indexes and co-factor indexes.

[0065] In some implementations, the display generation engine 124 may position a graphical representation of a patient identifier along a first axis based on the patient matching index for the patient identifier and along a second axis based on the co-factor index for the patient identifier. For example, with reference to the electronic display 180A of FIG. 3A, each of the circles (e.g., PI, P2, P3) is a gra phica l representation of a patient identifier. The circles are positioned along the vertical axis based on their respective patient matching index and are positioned along the horizontal axis based on their respective co- factor index. For example, circle PI corresponds to Patient Identifier 1 described in the examples above and is positioned at a vertical axis value of approximately 39 and a horizontal axis value of approximately 6.82. Although circles are illustrated in the example electronic display 180A of FIG. 3A, additional and/or alternative graphical representations may be used such as, for example, alternative shapes, alphanumeric characters, a nd/or symbols.

[0066] The electronic display 180A includes major gridline markings a long the vertica l axis at values 10, 20, 30, 40, 50, and 60 and also includes labels for some of the major gridline markings that indicate a stage of the primary diagnosis of thyroid cancer. For example, the label for value 20 is "Stage II", indicating that the circles from vertical axis values of 20 to 29 correspond to patient identifiers having Stage I I thyroid cancer. For example, as described above, the patient matching index may be calculated based on stage II having a value of "20", stage I II having a value of "30", and each of six additional non- mandatory selection criteria having values of "1.5". Thus, in such an example, patient matching indexes for patient identifiers having Stage II thyroid cancer would have a minimum value of 20 (no additional non-mandatory selection criteria met) and a maximum value of 29 (all six additional non-mandatory selection criteria met). Likewise, in such an example, patient matching indexes for patient identifiers having Stage II thyroid cancer would have a minimum value of 30 (no additional non-mandatory selection criteria met) and a maximum value of 39 (all six additional non-mandatory selection criteria met).

[0067] The electronic display 180A also includes major gridline markings along the horizontal axis at values 10, 20, 30, 40, 50, 60, and 70 to indicate co-factor indexes. In the illustrated electronic display 180, as the co-factor index increases from left to right along the horizontal axis, it indicates an increasing likelihood that other health issues will impact the clinical trial. For example, circle P3 has a co-factor index of approximately 68, while circle P2 has a co-factor index of approximately 6.82. As described above, the co-factor indexes may be calculated based on normalized test values associated with patient identifiers associated with circles P2 and P3 and based on regression correlation coefficients associated with those test values for the primary diagnosis (and optionally an extent of the primary diagnosis). Accordingly, the co-factor index for each patient identifier may reflect likelihood that other health issues associated with that patient identifier will impact the clinical trial - and are graphically represented in electronic display 180A by the horizontal displacement of the circle associated with that patient identifier. Thus, from view of the display 180A it can be determined that there is a greater likelihood that that other health issues will impact the clinical trial for a patient associated with circle P3 than for a patient associated with circle P2.

[0068] Although not illustrated in FIG. 3A, in some implementations the display generation engine 124 may provide a view range interface to enable user adjustment of the view range of display 180A. For example, the view range interface may enable adjustment of the vertical axis to only display from 25 to 50 instead of from 0 to 60 as illustrated in FIG. 3A. Also, for example, the view range interface may enable adjustment of the horizontal axis to only display from 0 to 60 instead of from 0 to 80 as illustrated in FIG. 3A.

[0069] The display generation engine 124 provides data to client device 106 to enable the client device to visually present the electronic display 180A to a user via one or more user interface output devices such as a screen. As described herein, in some

implementations the display generation engine 124 may itself be implemented, in whole or in part, on the client device 106. In some implementations, a user of the client device 106 may provide display feedback to the display generation engine 124 via one or more user interface input devices such as a mouse, touch-screen, and/or keyboard. The display feedback may indicate one or more requested modifications to be made to the electronic display 180A. For example, the display feedback may be responsive to user manipulation of the view range interface described in the preceding paragraph and the display generation engine 124 may modify the display 180A (e.g., generate a new display and/or dynamically update the existing display) to adjust the view range.

[0070] With reference to FIG. 3B, another example of display feedback is provided via client device 106 and a resulting modification to the display 180A is provided to generate modified display 180B. FIG. 3B illustrates the example electronic display of FIG. 3A, with an additional information graphical element 182 being displayed for a patient identifier associated with one of the graphical representations P3. The graphical element 182 may be displayed responsive to display feedback that indicates additional information related to circle P3 is desired, such as hovering of cursor 181 over circle P3 and/or clicking on circle P3. Additional and/or alternative display feedback may be utilized such as selection of P3 via a touch-screen, voice commands, etc. The graphical representation 182 displays some of the z-scores and regression correlation coefficients utilized to calculate the co-factor index for the patient identifier associated with circle P3, along with an option to "load more" of the utilized z-scores and regression correlation coefficients. I n other implementations, additional and/or alternative information may be displayed in graphical element 182 such as underlying data related to the patient matching index for circle P3.

[0071] A user may select, via one or more user interface input devices of client device 106, one or more graphica l representations of the provided electronic display that the user desires to include and/or exclude from a clinical tria l set. The selection may be provided to the trial set selection engine 125. The trial set selection engine 125 utilizes the input to select the subset of potential patients to include in a clinical trial set. Displaying the graphical representations positioned based on the patient matching indexes and/or the patient matching indexes may enable the user to take into account the conformance of patient identifiers to clinical trial criteria (e.g., as indicated by the patient matching indexes positioning) and/or the potential effect of one or more secondary factors of the patient identifiers on the clinical trial (e.g., as indicated by the co-factor indexes positioning).

[0072] For example, with reference to FIG. 3C, a user may utilize cursor 181 to define a selection area 183 and an indication of the selection area 183 and/or the circles (or associated patient identifiers) included in the selection area 183 may be provided to the trial set selection engine 125. FIG. 3C illustrates the example electronic display of FIG. 3A, that has been modified by the display generation engine 124 to generate a modified display 180C that illustrates the selection area 183. The trial set selection engine 125 may utilize that input to select the patient identifiers to include in a clinical trial set. The trial set selection engine 125 may store an indication of the selected clinical trial set in one or more storage mediums (e.g., patient data database 154) and/or send the indication in a graphical format for viewing by one or more users (e.g., to client device 106). The selected clinical trial set may be subsequently utilized in conducting the clinical trial.

[0073] Although a particula r selection is illustrated in FIG. 3C, additional and/or alternative selections via the client device 106 may be utilized to provide an indication of patient identifiers to include a nd/or exclude from a clinical trial set. For example, in some im plementations a user interface input device of client device 106 may be utilized to select one or more graphical representations to exclude from a clinical trial set. For instance, one or more graphical representations could be "swiped away" to exclude them and/or an exclusion selection area may be "drawn" around the graphical representations to exclude them. As another example, in some implementations a user interface input device of client device 106 may be utilized to select one or more graphical representations by inputting specific ranges of patient matching indexes and/or co-factor selection indexes that should be included and/or excluded from a clinical trial data set. For instance, a patient matching index range of 20-39 and a co-factor index range of 1-59 may be inputted a nd provided as the selection. Also, for instance, a patient matching index range of 23-29 and 33-39 and a co-factor index range of 1-59 may be inputted and provided as the selection. Also, for insta nce, a co-factor index range of 1-59 may be inputted and provided as the selection.

[0074] Also, although the example electronic displays of FIGS. 3A-3C illustrate the graphical representations of the patient identifiers being positioned based on both the patient matching index and the co-factor index, in some other implementations only one of the two indexes may be utilized in positioning the graphical representations. For example, in some implementations the patient presentation system 120 may calculate co-factor indexes as described herein, but not calculate patient matching indexes. In some of those im plementations, the display generation engine 124 may position graphical representations associated with patient identifiers along a first axis based on the co-factor indexes. I n some versions of those implementations, the display generation engine 124 may optionally position the graphical representations along a second axis based on one or more other factors besides the patient matching index. For example, the display generation engine 124 may position all of the graphical representations at the same distance along the second axis. Also, for example, the display generation engine 124 may position all of the graphical representations at random and/or predefined distances along the second axis to "de- clutter" the graphical representations along the second axis.

[0075] FIG. 4 is a flow chart illustrating an example method of transforming patient data, generating an electronic display based on the transformed data, and selecting a trial set based on an electronic selection received responsive to the electronic display. Other implementations may perform the steps in a different order, omit certain steps, and/or perform different and/or additional steps than those illustrated in FIG. 4. For convenience, aspects of FIG. 4 will be described with reference to a system of one or more computers that perform the process. The system may include, for example, one or more of the engines 121-125 of patient presentation system 120.

[0076] At step 400, a set of patient identifiers is selected from an electronic database based on one or more mandatory criteria for a clinical trial. For example, trial criteria matching engine 121 may select a set of patient identifiers based at least on patient data of those patient identifiers being associated with a primary diagnosis for the clinical trial. For instance, the set of patient identifiers {PI, P2, Pn} may be selected.

[0077] At step 405, patient matching indexes are calculated for the patient identifiers of the set. For example, patient matching index engine 122 may calculate a patient matching index for each patient identifier PI, P2, Pn of the set of patient identifiers {PI, P2, Pn}. Each patient matching index may be calculated based on at least one of: a most recent extent of the primary diagnosis as indicated by patient data of a respective patient identifier; and conformance of one or more additional selection criteria defined for the clinical trial to patient data of a respective patient identifier.

[0078] At step 410, co-factor indexes are calculated for the patient identifiers of the set. For example, co-factor index engine 123 may calculate a co-factor index for each patient identifier PI, P2, Pn of the set of patient identifiers {PI, P2, Pn}. Each co-factor index may be calculated based at least in part on one or more test values, of the patient data associated with a respective patient identifier, for one or more selected medical tests. As one example, the co-factor index engine 123 may calculate the co-factor index based on values of one or more medical test results that indicate: complete blood counts with white cell differential, serum electrolytes, liver profile enzymes, metabolic study values, estimated glomerular filtration rate, selected tumor markers, genetic markers, body mass index, and waist circumference.

[0079] As one example, for a patient identifier assume the following vector of norma lized test result values on which a co-factor index is calculated: {0, 0, 1, 10, 20, 0, 0, 1, 1}. Further assume that the following vector of regression correlation coefficients have been determined for the primary diagnosis (and optionally a particular stage of the diagnosis and/or other factor) for the aforementioned medical test results: {0.01, 0.9, 0.15, 0.81, 0.03, 0.2, 0.01, 0.5, 0.5}. As described herein, the regression correlation coefficients are statistically calculated based on historical data and each generally indicate the calculated im pact, for a particular prima ry diagnosis (and optionally a particular stage of the prima ry diagnosis and/or other factor(s)), of values associated with a respective medical test result. The co-factor index engine 123 may calculate the co-factor index for patient identifier PI based on the dot product of the two vectors. For example, for the immediately preceding vectors, the co-factor index for patient identifier PI may be 13.15.

[0080] At step 415, an electronic display of graphical representations of the patient identifiers is generated based on the patient matching and co-factor indexes. For example, the display generation engine 124 may position a graphical representation of each of the patient identifiers PI, P2, Pn of the set of patient identifiers {PI, P2, Pn} along a first axis based on the patient matching index for the respective patient identifier and along a second axis based on the co-factor index for the respective patient identifier. For example, with reference to the electronic display 180A of FIG. 3A, each of the circles (e.g., PI, P2, P3) is a graphical representation of a patient identifier. The circles are positioned along the vertical axis based on their respective patient matching index and are positioned along the horizontal axis based on their respective co-factor index.

[0081] At step 420, an electronic selection of one or more of the graphical

representations is received. For example, the electronic selection may indicate a subset of one or more of the patient identifiers of the set of patient identifiers {PI, P2, Pn}— and the trial set selection engine 125 may utilize that input to select the subset. For example, a user may select one or more graphical representations to include and/or exclude from a clinical trial set via interaction with the electronic display utilizing one or more user interface input devices (e.g., a mouse, keyboard, touch screen, or audio input).

[0082] At step 425, a clinical trial set is selected based on the electronic selection. For example, based on the selection, the trial set selection engine 125 may determine one or more patient identifiers of the set of patient identifiers {PI, P2, Pn} to include in a clinical trial set and/or to exclude from the clinical trial set. An indication of the selected clinical trial set may be stored in one or more storage mediums (e.g., patient data database 154) and/or presented to one or more users (e.g., via client device 106). The selected clinical trial set may be subsequently utilized in conducting the clinical trial.

[0083] In some implementations, steps 415, 420, and/or 425 may be omitted. For example, in some implementations the system may additionally and/or alternatively select a clinical trial set based on the patient matching indexes and/or co-factor indexes, without necessarily generating an electronic display and/or receiving an electronic selection responsive to the electronic display. For instance, in some implementations, a range of acceptable patient matching indexes and/or co-factor indexes for a clinical trial set may be defined and the system may select a clinical trial set by including in the clinical trial set patient identifiers that have associated patient matching indexes and/or co-factor indexes in the defined range. The range and/or other clinical trial set selection criteria may be defined by user input via client device 106 or otherwise defined or determined (e.g., calculated based on the patient matching indexes and co-factor indexes of all of the patient identifiers of the set identified at step 400), without necessarily generating an electronic display. The patient identifiers of the selected clinical trial set may be assigned as pa rticipants in the clinical trial and patients corresponding to the patient identifiers subsequently utilized in conducting the clinical trial.

[0084] FIG. 5 is a flow chart illustrating an example method of calculating a regression correlation coefficient for a medical test result for a primary diagnosis. Other

im plementations may perform the steps in a different order, omit certain steps, and/or perform different and/or additional steps than those illustrated in FIG. 5. For convenience, aspects of FIG. 5 will be described with reference to a system of one or more computers that perform the process. The system may include, for example, co-factor index engine 123 of patient presentation system 120.

[0085] At step 500, a primary diagnosis and a medical test result a re identified. For example, the system may identify a primary diagnosis of thyroid cancer and a medical test result of thyroid hormone levels (T3/T4) . I n some implementations, multiple medical test results may be identified (e.g., thyroid stimulating hormone, body mass index, and waist circumference) and/or the prima ry diagnosis may be defined more particularly, such as a prima ry diagnosis of Stage I I thyroid cancer. Generally the selected medical test results will include medical test results that define the primary diagnosis.

[0086] At step 505, a set of patient identifiers each associated with a value for the prima ry diagnosis and a value for the medical test result is identified from an electronic database such as patient data database 154 and/or another database with historical patient data . For example, where the primary diagnosis is thyroid cancer and the medical test result is thyroid hormone levels (T3/T4) , the system may identify a random set of patient identifiers that are associated with a "true" or "false" value for the primary diagnosis and a value for Thyroid hormone levels that optionally substantially corresponds in time with the value for the prima ry diagnosis (e.g., the time of the primary diagnosis and the time that the medical test are sufficiently close).

[0087] At step 510, one or more dependent variable values are determined for an analysis set based on the values of the primary diagnosis. For example, where the system identifies a primary diagnosis of thyroid cancer, actual values of "true" or "false" may be determined for the set of patient identifiers identified at step 505.

[0088] At step 515, one or more independent variable values are determined for the analysis set based on the values for the medical test result. For example, where the system identifies a test result of thyroid hormone levels serum electrolytes, actual values associated with test results may be determined for the set of patient identifiers identified at step 505, and paired with respective actual values of the dependent variables determined at step 510. In some implementations, the independent variables may be normalized by the system and/or may already be normalized in a database from which the system retrieves the independent varia bles. [0089] At step 520, a regression correlation coefficient is calculated based on the analysis set. The regression correlation coefficient is associated with the independent variable and provides an indication of the relationship of the medical test result to the primary diagnosis. The system may utilize various statistical regression techniques in calculating the regression correlation coefficient. For example, linear regression, logistic regression, and/or machine learning techniques may be utilized. In some implementations the analysis set may be provided to a statistical software package and the statistical software package may return the regression correlation coefficient.

[0090] At step 525, the regression correlation coefficient is assigned to the medical test result and the primary diagnosis. For example, the system may create a database entry that defines a triple that includes the regression correlation coefficient, the medical test result, and the primary diagnosis. In some implementations, the system may assign the regression correlation coefficient to the medical test result for the primary diagnosis in one or more databases and/or in code executable by the co-factor index engine 123 in determining co- factor indexes as described herein.

[0091] FIG. 6 is a flow chart illustrating an example method of calculating a contribution to a co-factor index for a patient identifier based on a value of a medical test result for the patient identifier. For convenience, aspects of FIG. 6 will be described with reference to a system of one or more computers that perform the process. The system may include, for example, patient data processing system 130 and/or co-factor index engine 123 of patient presentation system 120.

[0092] At step 600, a medical test result is identified for a co-factor index for a primary diagnosis. For example, the co-factor index engine 123 may identify one of multiple medical test results that contribute to the co-factor index for the primary diagnosis. As one example, assume the primary diagnosis is thyroid cancer and the medical test results on which the co-factor index is based include a medical test result of complete blood counts with white cell differential, serum electrolytes, metabolic function studies, e-GFR, total cholesterol, lipid profile, blood pressure, and PHQ.-9.

[0093] At step 605, a value for the medical test results is identified for a patient identifier. For example, the co-factor index engine 123 may access patient data database 154 to identify a value for the medical test result for a patient identifier that has been identified as having the primary diagnosis.

[0094] At step 610, a normalized value is calculated for the value. For example, the patient data processing system 130 or the co-factor index engine 123 may calculate a z- score for the value based on a formula such as: z-score = (patient's value for medical test result) - (mean of the value for the medical test result for a population of patients / standard deviation of the value for the medical test result for the population of patients).

In some implementations step 610 may be omitted. For example, in some implementations the value identified at step 605 may already be normalized.

[0095] At step 615, a calculated regression correlation coefficient assigned to the medical test result for the primary diagnosis is identified. For exam ple, the co-factor index engine 123 may access a data structure that contains, for each of a plurality of primary diagnoses, an indication of test results and regression correlation coefficients for those test results. For instance, the co-factor index engine 123 may identify that for the primary diagnosis of thyroid cancer, a regression co-efficient of 0.15 is associated with the medical test result of complete blood counts with white cell differential. In some implementations, the regression correlation coefficient identified at step 615 may be calculated based on the method of FIG. 5.

[0096] At step 620, a contribution to the co-factor index is calculated based on the standardized value for each test result and the regression correlation coefficient. For example, the co-factor index engine 123 may multiply the regression correlation coefficient and the standardized value for each test result and then sum this value with the other calculated regression correlation coefficients multiplied with their respective standa rdized test result values. The sum product result is the basis for the co-factor index. For instance, the result may itself be the contribution and/or it may be multiplied and/or divided by a factor (e.g., 10) to scale the resulting co-factor index up or down.

[0097] At step 625, the contribution is incorporated in the co-factor index. For example, the co-factor index engine 123 may use the contribution as one of multiple contributions to the co-factor index for the patient identifier. For instance, the co-factor index engine 123 may repeat the method of FIG. 6 for each of a plurality of medica l test results that contribute to the co-factor index and may sum a ll of the contributions calculated at step 620 to determine the co-factor index for the patient identifier.

[0098] FIG. 7 illustrates an example of calculating patient matching indexes and co-factor indexes for one or more patient identifiers for multiple time periods, and generating a n electronic display of a graphical representation of the patient identifiers for the multiple time periods based on the calculated values. Reference will also be made to FIGS. 8A-8C to further aid in explaining the example of FIG. 7.

[0099] One or more patient identifiers 105 are identified by patient matching index engine 122 and co-factor index engine 123. In some implementations, the patient identifiers 105 may be identified based on input from a user via client device 106. For example, the user may be the patient and the patient identifiers 105 may be identified based on matching a patient identifier to login information or other verifying credentials provided by the patient to the patient presentation system 120. Also, for example, the user may be a medica l professional, and the patient identifiers 105 may be identified based on matching a patient identifier to demographic a nd/or other input provide by the medical professional to the patient presentation system 120.

[00100] The patient matching index engine 122 calculates patient matching indexes for the identified patient identifiers 105 for each of a plurality of distinct time periods. For example, the patient matching index engine 122 may calculate a first patient matching index for a single patient identifier based on patient data of patient data database 154 that is associated with that patient identifier and associated with a first time period. The patient matching index engine 122 may further calculate additional patient matching indexes for the single patient identifier, with each being calculated based on patient data of patient data database 154 that is associated with that patient identifier and associated with a distinct time period (e.g., a second time period, a third time period, ... an n ^th time period). A time period associated with patient data generally defines a chronological period associated with that patient data and may be defined in various manners in the patient data. For example, the first time period may be a certain date (e.g., December 1, 2014) associated with input from the patient and/or medical professional that resulted in the patient data, a certain date that indicates the date of input from one of the medical center systems 103a-n that resulted in the patient data, and/or a certain date that the patient data was collected by the patient data processing system 130 and stored in patient data database 154. Also, for example, the first time period may indicate a chronologica l stage of a clinical trial such as "start of the trial", "one month into the trial", "three months into the trial", etc.

[00101] When the example of FIG. 7 is performed in the context of a clinical trial with a prima ry diagnosis, the patient matching index may be calculated by patient matching index engine 122 based on at least one of: an extent of the primary diagnosis associated with the given patient identifier; and conformance of the patient data associated with the given patient identifier to one or more additional selection criteria defined for the clinical trial. When the example of FIG. 7 is performed outside of the context of a clinical trial, at least a prima ry diagnosis may still be defined, and the patient matching index engine 122 may calculate the patient matching index based at least in part on the extent of the primary diagnosis.

[00102] The co-factor index engine 123 calculates co-factor indexes for the identified patient identifiers 105 for each of the plurality of distinct time periods. For exa mple, the co- factor index engine 123 may calculate a first co-factor index for a single patient identifier based on patient data of patient data database 154 that is associated with that patient identifier and associated with a first time period. The co-factor index engine 123 may further calculate additional co-factor indexes for the single patient identifier, with each being calculated based on patient data of patient data database 154 that is associated with that patient identifier and associated with a distinct time period (e.g., a second time period, a third time period, ... an n ^th time period). The co-factor index engine 123 may calculate a co-factor index as described herein, ta king into account regression correlation coefficients for one or more test values that contribute to the co-factor index, wherein the regression correlation coefficients are particularized to the primary diagnosis (and optionally a stage of the primary diagnosis at the respective time period).

[00103] The calculated patient matching indexes and co-factor indexes are provided to the display generation engine 124. The display generation engine 124 generates an electronic display of graphical representations of the one or more patient identifiers, and may position each of the graphica l representations based on associated calculated patient matching indexes and co-factor indexes associated with multiple time periods.

[00104] For example, for a single patient identifier, patient matching indexes and co- factor indexes for a first time period, a second time period, and through an n ^th time period may be provided to the display generation engine 124. The display generation engine 124 may generate a display that positions a graphical representation of the single patient identifier at a first position based on patient matching and co-factor indexes for a first time period, positions the graphical representation of the single patient identifier at a second position based on patient matching and co-factor indexes for a second time period, and so forth.

[00105] The display generation engine 124 may generate a display that causes display of the graphical representation at the first position, the second position, and/or other positions simultaneously and/or non-simultaneously. For example, the display generation engine 124 may generate a display that displays the graphical representation n times simultaneously, with each occurrence of the graphical representation being positioned based on patient matching and co-factor indexes for a respective time period. Also, for example, the display generation engine 124 may generate a display that animates the graphical representation by altering the position of the graphical representation as it is displayed, with each of the alterations being based on patient matching and co-factor indexes for a respective time period. In some implementations, the graphical

representation may be exactly the same at one or more of the multiple positions. In some im plementations, the graphical representations may graphically vary in one or more respects at one or more of the multiple positions. For instance, the graphical representation may include an annotation that provides an indication of the time period corresponding to a position of the graphical representation, and the annotation for each position may reflect its corresponding time period.

[00106] The display generation engine 124 provides data to client device 106 to enable the client device 106 to visually present the generated electronic display to a user via one or more user interface output devices such as a screen. As described herein, in some im plementations the display generation engine 124 may itself be implemented, in whole or in part, on the client device 106. I n some implementations, a user of the client device 106 may provide display feedback to the display generation engine 124 via one or more user interface input devices such as a mouse, touch-screen, a nd/or keyboard. The display feedback may indicate one or more requested modifications to be made to the generated electronic display.

[00107] For example, in implementations where the display generation engine 124 generates a display that animates a graphical representation by altering the position of the graphical representation as it is displayed (with each of the alterations being based on patient matching and co-factor indexes for a respective time period), the display feedback may indicate a desire for the position of the graphical representation to be updated to the next time period— and the display updated based on such display feedback. Also, for example, display feedback may indicate that additional information related to the patient data for patient identifier for one of the time periods is desired, and additional information may be provided by display generation engine 124 for display via the client device 106. For example, a user may select a graphical representation displayed at a particular position and z-scores and regression correlation coefficients utilized to calculate the co-factor index for that position may be displayed in a graphical element such as one similar to graphical element 182 of FIG. 3B.

[00108] FIG. 8A illustrates an example of an electronic display 185A of a graphical representation P4 of a patient identifier that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifier for multiple time periods. For example, with reference to FIG. 7, the patient identifiers 105 may be a single patient identifier corresponding to the graphical representation P4 of FIG. 8A. Patient matching index engine 122 may calculate patient matching values for that patient identifier for time periods Tl, T2, T3, and T4. Co-factor index engine 123 may calculate co-factor indexes for that patient identifier for time periods Tl, T2, T3, and T4.

[00109] The display generation engine 124 may generate the display 185A of FIG. 8A that illustrates the graphical representation P4 at each of the time periods T1-T4. The display generation engine 124 positions the graphical representation at each time period based on a patient matching index and co-factor index for the time period. For example, the position of graphical representation P4 at time period Tl (indicated as P4, Tl in FIG. 8A) indicates the patient identifier had stage III of the primary diagnosis and a co-factor index of

approximately 13. The positional progression of the graphical representation P4 to the right along the horizontal axis for time periods T1-T4 illustrates that the co-factor index for the patient identifier increased as time progressed. In the example of FIG. 8A, as the co-factor index increases, it indicates a n increased likelihood that other health issues associated with that patient identifier impact the primary diagnosis/ or that the experimental treatment im pacted other non-primary organ systems. The positional change of the graphical representation P4 from 30 to 40 along the vertical axis at time period T4 illustrates that the patient matching index increased at time period T4. I n the example of FIG. 8A, the progression from 30 to 40 along the vertical axis indicates progression of the primary diagnosis from stage I II to stage IV. The electronic display 185A may provide a visual representation of changes in the patient's health that may assist a medical professional's and/or the patient's understanding of those changes.

[00110] Although FIG. 8A illustrates the graphical representation P4 for each time period T1-T4 being displayed simultaneously, in some other implementations the graphical representation P4 for one or more time periods T1-T4 may be displayed non-simultaneously with the graphical representation P4 for one or more other of the time periods T1-T4. For example, the graphical representation P4 for time period Tl may be displayed a lone until display feedback is received (e.g., selection of an interface element via cursor 181 or "clicking" while cursor 181 is in the display area). In response to the display feedback, the display generation engine 124 may modify the display (e.g., generate a new display and/or dynamically update the existing display) to adjust the position of the graphical

representation P4 based on patient matching and co-factor indexes for the time period T2 alone. The position of the graphical representation P4 may be further progressed to correspond to future time periods in response to further display feedback.

[00111] FIG. 8B illustrates an example of an electronic display 185B of a graphical representation P5 of a first patient identifier a graphical representation P6 of a second patient identifier that may be generated based on calculated patient matching indexes and co-factor indexes for those patient identifiers for multiple time periods.

[00112] For example, with reference to FIG. 7, the patient identifiers 105 may be the patient identifiers corresponding to the graphical representations P5 and P6 of FIG. 8B. Patient matching index engine 122 may calculate patient matching indexes for those patient identifiers for time periods Tl, T2, T3, and T4. Co-factor index engine 123 may calculate co- factor indexes for those patient identifiers for time periods Tl, T2, T3, and T4. The time periods T1-T4 for the first patient identifier P5 may represent the same and/or similar time periods T1-T4 of the second patient identifier. For example, each of the time periods T1-T4 for the first patient identifier may correspond to dates that are distinct from dates of the time periods T1-T4 for the second patient identifier, but that represent similar chronological progressions. For instance, the time periods T1-T4 for the first patient identifier may be associated with January, February, March, and April dates of a particular year, whereas the time periods T1-T4 for the second patient identifier may be associated with July, August, September, and October dates of the particular year.

[00113] The display generation engine 124 may generate the display 185B of FIG. 8B that illustrates the graphical representations P5 and P6 at each of the time periods T1-T4. The display generation engine 124 positions the graphical representations at each time period based on a respective patient matching index and co-factor index for the time period. For example, the position of graphical representation P5 at time period Tl (indicated as P5, Tl in FIG. 8B) indicates the patient identifier had stage II of the primary diagnosis and a co- factor index of approximately 15. The positional progression of the graphical representation P4 to the right along the horizontal axis for time periods T1-T4 illustrates that the co-factor index for the patient identifier increased as time progressed. In the example of FIG. 8B, as the co-factor index increases, it indicates an increased likelihood that other health issues associated with that patient identifier impact the primary diagnosis. The positional change of the graphical representation P5 from 25 to 35 to 45 along the vertical axis illustrates that the patient matching index increased at time period T3 and increased again at time period T4. In the example of FIG. 8B, the progression from 25 to 35 along the vertical axis indicates progression of the primary diagnosis from stage II to stage III and the progression from 35 to 45 indicates progression of the primary diagnosis from stage III to stage IV.

[00114] Also, for example, the position of graphical representation P6 at time period Tl (indicated as P6, Tl in FIG. 8B) indicates the patient identifier had stage III of the primary diagnosis and a co-factor index of approximately 53. It is noted that the size of the graphical representation P6 is larger than the size of the graphical representation P5 to reflect that cancer staging is determined at the time of initial primary diagnosis (stage III vs. stage II). It is further noted that the sizes of the graphical representations P5 and P6 are maintained throughout the various time periods to reflect the initial extent even though the actual extent at a particular later time period may be different than the initial extent. [00115] The positional progression of the graphical representation P6 to the left along the horizontal axis for time periods T1-T4 illustrates that the co-factor index for the patient identifier decreased as time progressed. I n the example of FIG. 8B, as the co-factor index decreases, it indicates a decreased likelihood that other health issues associated with that patient identifier impact the primary diagnosis/ or it may reflect that the experimental drug or treatment did not produce unintended side-effects in other non-primary organ systems. The positional change of the graphical representation P6 from 38 to 28 to 18 along the vertical axis illustrates that the patient matching index decreased at time period T2 and decreased again at time period T4. I n the example of FIG. 8B, the regression from 38 to 28 along the vertical axis indicates regression of the primary diagnosis from stage I II to stage I I and the regression from 28 to 18 indicates regression of the primary diagnosis from stage II to stage I.

[00116] The electronic display 185B may provide a visual representation of comparison in changes in the health of two patients that may assist a medical professional's and/or one or more of the patients' understanding of those changes. For example, the electronic display 185B may be presented to a patient corresponding to the graphical representation P5 to illustrate that secondary factors may have impacted the patient's primary diagnosis as there was a correlation between progression of that patient's co-factor index and progression of that patient's primary diagnosis, whereas there was a correlation between regression of the other patient's co-factor index and regression of that patient's primary diagnosis.

[00117] Although FIG. 8B illustrates two graphical representations P5 a nd P6, in some other implementations gra phical representations corresponding to more than two patients may be provided in a generated display. Also, the graphical representations P5 and/or P6 for one or more time periods T1-T4 may be displayed non-simultaneously with the graphical representation P5 and/or P6 for one or more other time periods T1-T4 in a similar manner as described above with respect to FIG. 8A.

[00118] FIG. 8C illustrates another example of an electronic display 185C of graphical representations P7 and Gl of multiple patient identifiers that may be generated based on calculated patient matching indexes and co-factor indexes for the patient identifiers for multiple time periods.

[00119] For example, with reference to FIG. 7, the patient identifiers 105 may be a single patient identifier corresponding to the graphical representation P7 of FIG. 8C a nd a set of patient identifiers corresponding to the graphical representation Gl of FIG. 8C. Patient matching index engine 122 may calculate patient matching indexes for the patient identifier corresponding to graphical representation P7 for time periods Tl, T2, T3, and T4. Co-factor index engine 123 may calculate co-factor indexes for the patient identifier corresponding to graphical representation P7 for time periods Tl, T2, T3, and T4.

[00120] Patient matching index engine 122 may calculate patient matching indexes for the set of patient identifier corresponding to graphical representation Gl for time periods Tl, T2, T3, and T4. Patient matching index engine 122 may further calculate an average value and/or other statistica l measure (e.g., a median value) for the patient matching indexes for each of the time periods T1-T4. For example, the average value for time period Tl may be the average of the patient matching indexes for the set of patient identifiers corresponding to the graphical representation Gl. Co-factor index engine 123 may calculate co-factor indexes for the time period Tl for the patient identifier corresponding to graphical representation Gl for time periods Tl, T2, T3, and T4. Co-factor index engine 123 may further calculate an average value and/or other statistical measure (e.g., a median value) for the co-factor indexes for each of the time periods T1-T4. For exam ple, the average value for time period Tl may be the average of the co-factor indexes for time period Tl for the set of patient identifiers corresponding to the graphical representation Gl. The time periods Tl- T4 for each of the patient identifiers of the set may represent the same and/or similar time periods T1-T4 of other patient identifiers of the set. Also, the time periods T1-T4 for each of the patient identifiers of the set may represent the same and/or similar time periods T1-T4 of the patient identifier corresponding to graphical representation P7.

[00121] The set of patient identifiers corresponding to the graphical representation Gl may be defined in various manners to provide a desired comparison set. The comparison set enables a visual comparison of the health changes of a patient identifier corresponding to graphical representation P7 to the average (or other statistical measure) health changes associated with the set of patient identifiers corresponding to gra phical representation Gl. For example, the patient identifier corresponding to graphical representation P7 may be a pa rticipant in a clinical trial and the set of patient identifiers may be a randomly selected set that includes other patient identifiers from the clinical tria l. In some implementations, the selected set of patient identifiers may be restricted to those that have one or more pa rticular initia l extents of a primary diagnosis (e.g., the same as the patient identifier corresponding to graphical representation P7) and/or one or more particular initial co-factor indexes (e.g., within +/- 10% of the initial co-factor index of the patient identifier

corresponding to graphical representation P7).

[00122] The display generation engine 124 may generate the display 185C of FIG. 8C that illustrates the graphical representations P7 and Gl at each of the time periods T1-T4. The display generation engine 124 positions the graphical representation P7 at each time period based on a respective patient matching index and co-factor index for the time period. For example, the position of graphical representation P7 at time period Tl (indicated as P7, Tl in FIG. 8C) indicates the patient identifier had stage II of the primary diagnosis and a co- factor index of approximately 33. The positional progression of the graphical representation P4 to the right along the horizontal axis for time periods T1-T4 illustrates that the co-factor index for the patient identifier increased as time progressed. I n the example of FIG. 8C, as the co-factor index increases, it indicates an increased likelihood that other health issues associated with that patient identifier are impacting the primary diagnosis. The positional change of the graphical representation P7 from 23 to 33 to 43 along the vertical axis illustrates that the patient matching index increased at time period T3 and increased again at time period T4. In the example of FIG. 8C, the progression from 23 to 33 along the vertical axis indicates progression of the primary diagnosis from stage II to stage I II and the progression from 33 to 43 indicates progression of the primary diagnosis from stage I II to stage IV.

[00123] The display generation engine 124 positions the graphical representation Gl at each time period based on a respective patient matching index average va lue and co-factor index average value for the time period. The positional changes of the graphical

representation Gl along the horizontal axis for time periods T1-T4 illustrates that the co- factor index average va lue for the set of patient identifiers increased from time period Tl to time period T2, then decreased from time period T3 to time period T4. The positional change of the graphical representation Gl from 28 to 19 to 15 along the vertical axis illustrates that the patient matching index average value decreased at time period T3 and decreased again at time period T4. I n the example of FIG. 8C, the regressions along the vertical axis indicates regression of the extent of primary diagnosis among the set of patient identifiers.

[00124] The electronic display 185C may provide a visual representation of comparison in changes in the health of a single patient to a larger group of patients that may assist a medical professional's and/or the patient's understanding of those changes. For example, the electronic display 185C may be presented to a patient corresponding to the graphical representation P7 to illustrate how that patient's health changed throughout a clinical trial as compared to other participants in the clinical trial.

[00125] Although FIG. 8C illustrates comparing a graphical representation P7 that corresponds to a single patient identifier to a graphical representation Gl that corresponds to a set of patient identifiers, in some other implementations two or more sets of patient identifiers may be compared. For example, a first set of patient identifiers that correspond to graphical representation Gl may defined and a unique second set of patient identifiers may be defined. Average patient matching and co-factor indexes for each may be calculated for each of a plurality of time periods and graphical representations provided in a generated display to enable comparison of the average values of the two sets. The two sets may be defined to enable a desired comparison. For example, the first set may be patient identifiers from a first clinical trial for a primary diagnosis and the second set may be patient identifiers from a second clinical trial for the primary diagnosis to enable comparisons between the two clinical trials. Also, for example, the first set may be patient identifiers for a clinical trial that are associated with a first range of initial co-factor indexes and the second set may be patient identifiers for the clinical trial that are associated with a second range of initial co-factor indexes.

[00126] FIG. 9 is a flow chart illustrating an example method of generating an electronic display of a graphical representation of one or more patient identifiers based on calculated patient matching indexes and co-factor indexes for the patient identifiers for multiple time periods. For convenience, aspects of FIG. 9 will be described with reference to a system of one or more computers that perform the process. The system may include, for example, one or more of engines 122-124 of patient presentation system 120.

[00127] At step 900, a patient identifier and associated first and second patient data associated with respective first and second time periods is identified from an electronic database. For example, the patient identifiers 105 may be identified based on input from a user via client device 106 and the first and second patient data may be identified from patient data database 154. For example, the patient presentation system 120 may identify the first patient data based on it being associated with that patient identifier and associated with the first time period and may identify the second patient data based on it being associated with that patient identifier and associated with the second time period.

[00128] At step 905, first and second patient matching indexes are calculated based on the first and second patient data. For example, the patient matching index engine 122 may calculate the first patient matching index for the patient identifier based on the first patient data of patient data database 154 that is associated with that patient identifier and associated with the first time period. The patient matching index engine 122 may further calculate the second patient matching index based on the second patient data of patient data database 154 that is associated with that patient identifier and associated with the second time period.

[00129] At step 910, first and second co-factor indexes are calculated based on the first and second patient data. For example, the co-factor index engine 123 may calculate a first co-factor index for the patient identifier based on the first patient data of patient data database 154 that is associated with that patient identifier and associated with the first time period. The co-factor index engine 123 may further calculate the second co-factor index for the patient identifier based on the second patient data of patient data database 154 that is associated with that patient identifier and associated with the second time period.

[00130] At step 915, an electronic display of the patient identifier is generated for the first and second time periods based on the patient matching and co-factor indexes. For example, the display generation engine 124 may generate a display that positions a graphical representation of the patient identifier at a first position based on patient matching and co- factor indexes for a first time period and positions the graphical representation of the patient identifier at a second position based on patient matching and co-factor indexes for the second time period. The display generation engine 124 may generate a display that causes display of the graphical representation at the first position and the second position simultaneously and/or non-simultaneously.

[00131] Although the example method of FIG. 9 is described with respect to a single patient identifier and only two time periods, in some implementations multiple patient identifiers and/or more than two time periods may be identified. In those implementations, patient matching and co-factor indexes may be calculated for each of the patient identifiers for each of the time periods, and gra phica l representations of the patient identifiers positioned in a graphical representation based on such calculated indexes. For example, those implementations may include implementations such as those described above with respect to FIG. 8B and FIG. 8C.

[00132] In some implementations, patient presentation system 120 may monitor calculated patient matching indexes and/or co-factor indexes for one or more patient identifiers for multiple time periods and generate an electronic notification when one or more of the calculated indexes satisfy a threshold value. The threshold may be, for example, a fixed user defined value or a fixed historically calculated value. The electronic notification may be provided in various electronic forms (e.g., an email, a text, an interface element that surfaces on one or more of the displays illustrated in Figures herein) and may be provided to the one or more patients and/or one or more health professionals via one or more client devices. Genera lly, the electronic notification may inform the patients and/or health professionals that one or more of the ca lculated patient matching indexes and/or co- factor indexes are at levels that indicate further action may be needed. For example, the electronic notification may be generated when a delta change between a more recent in time co-factor index of at least one patient and a less recent in time co-factor index of the at least one patient exceeds a threshold value during a clinical trial. Such a delta change may indicate a change in health (e.g., a change in the health of one or more organs that a re unrelated to the primary diagnosis) that may indicate a side-effect of a test drug of the clinical trial a nd/or other hea lth issues.

[00133] As one example, a set of patient identifiers may be identified and, for each of the patient identifiers, associated first and second patient data associated with respective first and second time periods may be identified from the patient data database 154. For insta nce, the first time period for each patient identifier may be associated with most recent test values for the patient identifier and the second time period may be associated with less recent test values for the patient identifier (e.g., the initia l test values for the patient identifier, or the immediately preceding test values for the patient identifier). The co-factor index engine 123 may calculate a first co-factor index for each of the patient identifiers based on the first patient data and a second co-factor index for each of the patient identifiers based on the second patient data.

[00134] The patient presentation system 120 may generate an electronic notification when one or more of the calculated indexes satisfy a threshold value. For example, the patient presentation system 120 may generate the electronic notification when a delta change between one or more of the calculated first co-factor indexes and respective second co-factor indexes satisfies a threshold value. For instance, if at least five patient identifiers have a delta change that satisfies the threshold value, the electronic notification may be generated. Also, for instance, if the average delta change for all of the patient identifiers satisfies the threshold value, the electronic notification may be generated. As another example, the patient presentation system 120 may generate the electronic notification when one or more of the first co-factor indexes (associated with the more recent in time test values) satisfy a first threshold value and one or more respective delta changes satisfy a second threshold value. As one example of an electronic notification, an email may be generated that provides an indication of the changes in co-factor indexes and optionally provides graphical representations for one or more of the patient identifiers to enable visual analysis of the changes (e.g., as illustrated in FIGS. 8A-8C).

[00135] FIG. 10 is a block diagram of an example computer system 1010. Computer system 1010 typically includes at least one processor 1014 which communicates with a number of periphera l devices via bus subsystem 1012. These peripheral devices may include a storage subsystem 1024, including, for example, a memory subsystem 1025 and a file storage subsystem 1026, user interface input devices 1022, user interface output devices 1020, and a network interface subsystem 1016. The input and output devices allow user interaction with computer system 1010. Network interface subsystem 1016 provides an interface to outside networks and is coupled to corresponding interface devices in other computer systems.

[00136] User interface input devices 1022 may include a keyboard, pointing devices such as a mouse, trackball, touchpad, or graphics tablet, a scanner, a touchscreen incorporated into the display, audio input devices such as voice recognition systems, microphones, and/or other types of input devices. In general, use of the term "input device" is intended to include all possible types of devices and ways to input information into computer system 1010 or onto a communication network.

[00137] User interface output devices 1020 may include a display subsystem, a printer, a fax machine, or non-visual displays such as audio output devices. The display subsystem may include a cathode ray tube (CRT), a flat-panel device such as a liquid crystal display (LCD), a projection device, or some other mechanism for creating a visible image. The display subsystem may also provide non-visual display such as via audio output devices. In general, use of the term "output device" is intended to include all possible types of devices and ways to output information from computer system 1010 to the user or to another machine or computer system.

[00138] Storage subsystem 1024 stores programming and data constructs that provide the functionality of some or all of the modules described herein. For example, the storage subsystem 1024 may include the logic to perform one or more of the methods described herein such as, for example, the methods of FIGS. 4, 5, 6, and/or 9.

[00139] These software modules are generally executed by processor 1014 alone or in combination with other processors. Memory 1025 used in the storage subsystem can include a number of memories including a main random access memory (RAM) 1030 for storage of instructions and data during program execution and a read only memory (ROM) 1032 in which fixed instructions are stored. A file storage subsystem 1026 can provide persistent storage for program and data files, and may include a hard disk drive, a floppy disk drive along with associated removable media, a CD-ROM drive, an optical drive, or removable media cartridges. The modules implementing the functionality of certain implementations may be stored by file storage subsystem 1026 in the storage subsystem 1024, or in other machines accessible by the processor(s) 1014.

[00140] Bus subsystem 1012 provides a mechanism for letting the various components and subsystems of computer system 1010 communicate with each other as intended. Although bus subsystem 1012 is shown schematically as a single bus, alternative implementations of the bus subsystem may use multiple busses.

[00141] Computer system 1010 can be of varying types including a workstation, server, computing cluster, blade server, server farm, or any other data processing system or computing device. Due to the ever-changing nature of computers and networks, the description of computer system 1010 depicted in FIG. 10 is intended only as a specific example for purposes of illustrating some implementations. Many other configurations of computer system 1010 are possible having more or fewer components than the computer system depicted in FIG. 10.

[00142] While several implementations have been described and illustrated herein, a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein may be utilized, and each of such variations and/or modifications is deemed to be within the scope of the

implementations described herein. More generally, all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the teachings is/are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific implementations described herein. It is, therefore, to be understood that the foregoing implementations are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, implementations may be practiced otherwise than as specifically described and claimed. Implementations of the present disclosure are directed to each individual feature, system, article, material, kit, and/or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the scope of the present disclosure.