FIELD OF THE INVENTION

[0001] The present disclosure generally relates to systems and methods for using image analysis to generate surgical schedules based on available resources. In particular, the present disclosure relates to using computed tomography (CT) images and automatic image processing in combination with a database of available resources to more efficiently allocate resources.

BACKGROUND

[0002] Hospitals and other medical centers have limited resources, and as such, efficient allocation of such resources is desirable.

[0003] As an example, certain types of medical operating rooms, such as cardiac catheterization laboratories (cathlabs) are a limited resource, and certain types of medical procedures may require such a resource. Cathlab procedure scheduling is, therefore, a challenging task that typically requires a large amount of domain knowledge of a highly regulated and constrained environment.

[0004] Hospitals must typically account for various characteristics of the procedures to be performed and the resources available for performing the procedures. For example, resources, including medical teams available to perform procedures, must conform to requirements with respect to radiation safety, working hours, or specialty staff required to be present for a procedure. Ideally, such scheduling further accounts for the skills and preferences of particular staff members as well.

[0005] Such scheduling is typically done manually, via a human operator, or based on simple rules of thumb.

[0006] Imaging, such as computed tomography (CT) imaging, is increasingly used as a first-line test for various issues requiring procedures. For example, a cardiac CT is often a first- line test for coronary artery disease, and more patients with a prior CT exam are therefore scheduled for diagnostic and interventional cathlab procedures.

[0007] There is a need for systems and methods that can better make best use of available resources, such as available rooms, staff, etc., by improving scheduling of procedures. SUMMARY

[0008] Computer-implemented methods are provided for generating a medical procedure schedule. Such a schedule may be for scheduling cardiac procedures typically performed in a cathlab.

[0009] The method includes retrieving a plurality of images at a complexity analysis module. Each image is associated with a corresponding medical procedure of a plurality of medical procedures, where the plurality of medical procedures are to be scheduled.

[0010] The method proceeds with processing each image at the complexity analysis module to extract at least one value for at least one corresponding complexity metric. The method then generates a medical procedure characterization for each medical procedure of the plurality of medical procedures based at least partially on the at least one corresponding complexity metric.

[0011] The method then generates a time estimation for each medical procedure based on the medical procedure characterization.

[0012] The method separately determines, at an available resource module, a set of resources available for performing the plurality of medical procedures. The resources include at least one operating room and at least one medical team.

[0013] The method then receives, at a scheduling module, the medical procedure characterization associated with each of the plurality of medical procedures and the set of resources, and allocates at least one operating room and at least one medical team to each medical procedure.

[0014] The method then generates a schedule based on the time estimation associated with each medical procedure and the allocated operating room and medical team.

[0015] In some embodiments, the processing of each image comprises identifying plaque in the corresponding image and determining a composition of the plaque. The composition may then define a value for the corresponding complexity metric. In some such embodiments, the plaque is classified as calcified, soft, or mixed, and that classification may further define an urgency metric. The schedule may then be generated at least partially based on the urgency metric. [0016] In some embodiments, at least one lesion is identified in an image of the plurality of images. A classification value for each lesion is then based at least partially on a centerline, lumen, and attenuation of the corresponding lesion. The complexity metric extracted from the image is then based at least partially on the number of lesions identified and the corresponding classification value for each lesion.

[0017] In some such embodiments, the method further includes defining a confidence metric for the time estimation associated with each medical procedure. The confidence metric may be based at least partially on the classification values for lesions associated with the medical procedure. The medical procedure characterization is then at least partially based on or updated to reflect the confidence metric.

[0018] In some such embodiments, the medical procedure characterization further comprises an urgency metric. The urgency metric may be at least partially based on a classification of a lesion as a short-term risk, an evaluation of plaque stability, or how focused the location of the lesion is.

[0019] In some embodiments, the classification value for any lesion determined to be a chronic total occlusion (CTO) is further based on at least one of a determination that an entry for the CTO is blunt, as opposed to tapered, a determination that plaque comprising the occlusion is calcified, and an evaluation of bending in the CTO segment, as well as a length of the occlusion.

[0020] In some embodiments, the classification value for any lesion may further be based on a determination of a location of the lesion, a percentage of stenosis, and a number of vessels impacted.

[0021] In some embodiments, the set of resources determined at the available resource module comprises a plurality of medical teams, wherein each medical team is associated with a set of skills.

[0022] In some such embodiments, the set of skills associated with each medical team includes a measure of experience for each medical team. The allocation of at least one medical team to each medical procedure is then based at least partially on the measure of experience.

[0023] In some such embodiments, upon determining that a medical procedure has a value for a complexity metric above a defined threshold, the scheduling module allocates the at least one medical team to the corresponding medical procedure only if the medical team has a measure of experience above a defined threshold.

[0024] In some embodiments, the set of resources determined at the available resource module further comprises a plurality of operating rooms, and each operating room is associated with defined capabilities. The allocation of an operating room to a medical procedure by the scheduling module is based on a comparison between a corresponding medical procedure characterization and the capabilities associated with the operating room. Further, the allocation of a medical team to the corresponding medical procedure by the scheduling module is based at least partially on the corresponding complexity metric.

[0025] In some such embodiments, the method further includes defining an urgency metric for each medical procedure. The schedule is then generated based at least partially on the urgency metric and an availability of the allocated medical team for the corresponding medical procedure.

[0026] In some embodiments, the method further includes defining a confidence metric for the time estimation associated with each medical procedure, and the schedule is generated at least partially based on the confidence metric.

[0027] In some embodiments, the time estimation is a dimensionless work unit and each medical team is associated with at least one time modifier, such that the time estimation may be converted to units of time by modifying the work unit for a particular medical procedure based on the time modifier associated with a particular medical team.

[0028] In some such embodiments, the at least one time modifier comprises a plurality of distinct time modifiers, each time modifier associated with a specified type or group of medical procedures. Accordingly, when the scheduling module determines which of the plurality of distinct time modifiers is to be applied for a specific medical procedure, such a determination may be based on the corresponding medical procedure characterization.

[0029] In some embodiments, at least one complexity metric extracted from each image is based on a predictive model.

[0030] In some such embodiments, the method further includes training the predictive model based on previous iterations of the method of generating a medical procedure schedule, such that the at least one complexity metric generates more accurate time estimations. [0031] In some embodiments, the set of resources determined at the available resource module further comprises a quantity of each of a plurality of consumables available for use, such consumables including at least one of guidewires, balloons, stents, and contrast agent.

[0032] In some embodiments, the allocation by the scheduling module of the at least one operating room and at least one medical team is at the same time as the generation of the schedule based on an overall optimization metric. The scheduling module may then attempt to minimize or maximize the overall optimization metric.

[0033] In some such embodiments, the overall optimization metric is based on weighted measures of skill level and preferences associated with individual medical teams of the at least one medical team.

[0034] Also provided is a system for generating a medical procedure schedule. Such a system may include a memory for storing a plurality of instructions as well as several modules for supporting the generation of the schedule.

[0035] The system may include a complexity analysis module for evaluating a plurality of medical procedures and for generating a complexity metric for each medical procedure of the plurality of medical procedures.

[0036] The system may further include a case database for storing a medical procedure characterization for each medical procedure of the plurality of medical procedures.

[0037] The system may further include an available resource module for determining a set of resources available for performing the plurality of medical procedures. The resources include at least one operating room and at least one medical team.

[0038] The system may further include a scheduling module for generating a schedule.

[0039] The system then includes processor circuitry that couples with the memory and is configured to execute instructions to perform implementations of the method described above. Accordingly, the processor circuitry retrieves a plurality of images at the complexity analysis module, wherein each image is associated with a corresponding medical procedure of the plurality of medical procedures.

[0040] The processor circuitry then processes each image at the complexity analysis module to extract at least one value for the corresponding complexity metric for the corresponding medical procedure. [0041] The processor circuitry then generates, at the complexity analysis module, a medical procedure characterization for each medical procedure of the plurality of medical procedures based at least partially on the corresponding complexity metric, and stores the medical procedure characterization in the case database.

[0042] The processor circuitry then generates a time estimate for each medical procedure based on the medical procedure characterization and stores the time estimate in the case database with the corresponding medical procedure characterization.

[0043] The processor circuitry then allocates at least one operating room and at least one medical team from the available resource module to each medical procedure in the case database and generates a schedule at the scheduling module based on the time estimation associated with each medical procedure and the allocated operating room and medical team.

[0044] In some embodiments, the system further comprises a computed tomography imaging device, and the processor circuitry retrieves the plurality of images at the complexity analysis module from the imaging device.

BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a schematic diagram of a system according to one embodiment of the present disclosure.

[0046] Figure 2 illustrates modules utilized in a method in accordance with the present disclosure in the context of the system of FIG. 1.

[0047] Figure 3 illustrates a method for generating a medical procedure schedule in accordance with this disclosure.

[0048] Figure 4 illustrates a scoresheet for use in embodiments of the method of FIG. 3.

[0049] Figure 5 illustrates an example of scoring that could be used in the context of the method of FIG. 3.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0050] The description of illustrative embodiments according to principles of the present invention is intended to be read in connection with the accompanying drawings, which are to be considered part of the entire written description. In the description of embodiments of the invention disclosed herein, any reference to direction or orientation is merely intended for convenience of description and is not intended in any way to limit the scope of the present invention. Relative terms such as “lower,” “upper,” “horizontal,” “vertical,” “above,” “below,” “up,” “down,” “top” and “bottom” as well as derivative thereof (e.g., “horizontally,” “downwardly,” “upwardly,” etc.) should be construed to refer to the orientation as then described or as shown in the drawing under discussion. These relative terms are for convenience of description only and do not require that the apparatus be constructed or operated in a particular orientation unless explicitly indicated as such. Terms such as “attached,” “affixed,” “connected,” “coupled,” “interconnected,” and similar refer to a relationship wherein structures are secured or attached to one another either directly or indirectly through intervening structures, as well as both movable or rigid attachments or relationships, unless expressly described otherwise. Moreover, the features and benefits of the invention are illustrated by reference to the exemplified embodiments. Accordingly, the invention expressly should not be limited to such exemplary embodiments illustrating some possible non-limiting combination of features that may exist alone or in other combinations of features; the scope of the invention being defined by the claims appended hereto.

[0051] This disclosure describes the best mode or modes of practicing the invention as presently contemplated. This description is not intended to be understood in a limiting sense, but provides an example of the invention presented solely for illustrative purposes by reference to the accompanying drawings to advise one of ordinary skill in the art of the advantages and construction of the invention. In the various views of the drawings, like reference characters designate like or similar parts.

[0052] It is important to note that the embodiments disclosed are only examples of the many advantageous uses of the innovative teachings herein. In general, statements made in the specification of the present application do not necessarily limit any of the various claimed disclosures. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise indicated, singular elements may be in plural and vice versa with no loss of generality.

[0053] Imaging, such as computed tomography (CT) imaging, is increasingly used as a first-line test for various issues requiring procedures. For example, a cardiac CT is often a first- line test for coronary artery disease, and more patients with a prior CT exam are, therefore, scheduled for diagnostic and interventional cathlab procedures. There is a need for systems and methods that can better make best use of available resources, such as available rooms, staff, etc., by improving scheduling of procedures.

[0054] While such scheduling is currently performed manually by a human operator or based on simple rules of thumb, an automated approach using an objective measure of procedure complexity and duration estimation allows for improved utilization of resources. Accordingly, it is beneficial to preemptively estimate the complexity of any given procedure and project a time estimate for each such procedure and to schedule procedures accordingly. Such an approach avoids room overbooking, staff work overload, or idle time.

[0055] Because CT imaging has typically already been performed, such an estimate of complexity and time may be based on the prior CT exam analysis, thereby allowing for better use of available human staff and hospital facilities. Such an automated approach may further account for legal constraints, such as maximum allowable working hours and maximum exposure to radiation for particular operators while ensuring that appropriate staff is available for each procedure.

[0056] Figure 1 is a schematic diagram of a system 100 according to one embodiment of the present disclosure. As shown, the system 100 typically includes a processing device 110 and an imaging device 120.

[0057] The processing device 110 may apply processing routines to images or measured data, such as projection data, received from the imaging device 120. The processing device 110 may include a memory 113 and processor circuitry 111. The memory 113 may store a plurality of instructions. The processor circuitry 111 may couple to the memory 113 and may be configured to execute the instructions. The instructions stored in the memory 113 may comprise processing routines, as well as data associated with processing routines, such as machine learning algorithms, and various filters for processing images. While all data is described as being stored in the memory 113, it will be understood that in some embodiments, some data may be stored in a database, which may itself either be stored in the memory or stored in a remote discrete system.

[0058] The processing device 110 may further include an input 115 and an output 117. The input 115 may receive information, such as images or measured data, from the imaging device 120. The output 117 may output information, such as processed images, to a user or a user interface device. The output 117 similarly may output determinations generated by the method described below, such as recommendations. The output may include a monitor or display.

[0059] In some embodiments, the processing device 110 may relate to the imaging device 120 directly. In alternate embodiments, the processing device 110 may be distinct from the imaging device 120, such that it receives images or measured data for processing by way of a network or other interface at the input 115.

[0060] In some embodiments, the imaging device 120 may include an image data processing device, and a spectral or conventional CT scanning unit for generating the CT projection data when scanning an object (e.g., a patient).

[0061] While a system is shown including an imaging device 120 and a processing device 110, it will be understood that the method may be implemented directly on a processing device, as in the context of an image received by way of a network at the input 115. The methods described herein involve processing an image as a component of generating a schedule for various medical procedures. As noted above, prior to such a procedure, imaging is performed. As such, previously generated imaging may be retrieved by way of the input 115 and evaluated prior to or in place of obtaining a new image.

[0062] Figure 2 illustrates modules utilized in a method in accordance with the present disclosure in the context of the system 100 of FIG. 1. As noted above, the modules are typically stored in the memory 113 and the method is then implemented by way of the processor circuitry 111 of the processing device 110. Accordingly, during use, the memory 113 may contain instantiations of the modules described herein, as well as the databases and underlying data relied on.

[0063] The system 100 may then include a complexity analysis module 200 which evaluates a plurality of medical procedures 110a, 110b, 110c to be scheduled. The complexity analysis module 200 may then generate a complexity metric for each medical procedure 110a, 110b, 110c. As described herein, each medical procedure 110a, 110b, 110c may have a corresponding fde which may include imaging, such as CT scans, along with other data to be utilized by the complexity analysis module 200.

[0064] The system 100 may further include a case database 220 for storing a medical procedure characterization for each medical procedure of the plurality of medical procedures 110a, 110b, 110c. The medical procedure characterizations may be generated at the complexity analysis module 200, and may be based at least partially on the complexity metric generated.

[0065] The system 100 further includes an available resource module 230 which determines a set of resources available for performing the plurality of medical procedures 110a, 110b, 110c. The resources include at least one operating room 240, such as a cathlab, and at least one medical team 250. As will become apparent below, the resources will typically include multiple operating rooms 240 and/or medical teams 250, and in many cases, different operating rooms and medical teams will have distinct characteristics. As such, the allocation of specific rooms 240 and medical teams 250 for specific medical procedures 110a, 110b, 110c will become more important for efficient usage of resources.

[0066] In some embodiments, the available resource module 230 may be a database, and may catalog available resources. In other embodiments, the module may be connected to other systems, such as hospital databases, and may thereby obtain information so as to populate the resources 240, 250 in the context of the module 230. The available resource module 230 may further include information related to other resources required for various medical procedures 110a, 110b, 110c, such that resources can be properly allocated.

[0067] The system 100 further includes a scheduling module 260 which may then generate a schedule, or a worklist 270, based on information obtained from each of the case database 220 and the available resource module 230.

[0068] During use, and as discussed in more detail below with respect to FIG. 3, the processor circuitry 111 couples with the memory 113 and is configured to execute instructions to retrieve a plurality of images at the complexity analysis module 200. Each image is associated with a corresponding medical procedure 110a of the plurality of medical procedures 110a, 110b, 110c.

[0069] The processor circuitry 111 is then used to process each image at the complexity analysis module 200 to extract at least one value for the corresponding complexity metric for the corresponding medical procedure 110a, 110b, 110c. The processor circuitry 111 then generates, at the complexity analysis module 200, a medical procedure characterization for each medical procedure 110a, 110b, 110c based on least partially on the corresponding complexity metric, and stores the characterizations in the case database 220. [0070] The processor circuitry 111 further generates a time estimate for each medical procedure 110a, 110b, 110c based on the medical procedure characterization and stores the time estimate in the case database 220 with the corresponding characterization.

[0071] The processor circuitry 111 then allocates, at the scheduling module 260, at least one operating room 240 and at least one medical team 250 from the available resource module 230 to each medical procedure 110a, 110b, 110c in the case database 220 and generates a schedule based on the time estimation associated with each medical procedure and the allocated resources.

[0072] In some embodiments, as discussed above, the system 100 may further include an imaging device 120, such as a computed tomography (CT) imaging device, connected to the processing device 110. In such an embodiment, the processor circuitry 111 retrieves the plurality of images at the complexity analysis module 200 from the imaging device 120.

[0073] Figure 3 illustrates a method for generating a medical procedure schedule in accordance with this disclosure.

[0074] As shown, a method is provided for generating a medical procedure schedule. The method is often tailored for specific types of medical procedures, and may be used to allocate resources specific to those types of medical procedures efficiently. For example, the method as described is provided for scheduling procedures in cardiac catheterization laboratories (cathlabs).

[0075] Such a cathlab would typically be used for cardiac procedures for treating coronary artery disease. However, procedures may vary in terms of complexity and resource requirements, as well as time required. Because cathlabs are a limited resource, the ability to accurately predict a time requirement for a procedure allows a facility to increase their utilization of available cathlabs. Accordingly, the method provided herein facilitates improved utilization of resources.

[0076] The method includes first retrieving a plurality of images (300) at a complexity analysis module 200. Each image retrieved is associated with a corresponding medical procedure 110a of a plurality of medical procedures 110a, 110b, 110c.

[0077] The method then proceeds with processing (310) each image at the complexity analysis module 200 in order to extract (320) at least one value for at least one corresponding complexity metric. The method then generates (330) a medical procedure characterization for each medical procedure 110a, 110b, 110c based at least partially on the corresponding complexity metric (extracted at 320).

[0078] The processing of the images (310) may include identifying plaque (313) in the corresponding image and determining (316) a composition of the plaque. The composition of the plaque may define a value for the corresponding complexity metric. For example, the plaque may be classified as calcified, soft, or mixed.

[0079] In some embodiments, in addition to the complexity metric (extracted at 320), the method may further generate (340) an urgency metric. The urgency metric may be defined by, or at least partially based on, the composition of the plaque identified in the corresponding image.

[0080] The processing of the images (at 310) may further include identifying (350) at least one lesion to be treated. Each lesion identified may be assigned a classification value (360) based at least partially on a centerline, lumen, and attenuation of the corresponding lesion. In such an embodiment, the complexify metric (extracted at 320) may be based at least partially on the number of lesions identified and the corresponding classification value for each lesion.

[0081] It is noted that the term lesion may refer to any number of types of damage to be addressed by a medical procedure. For example, the lesions discussed herein may each be a chronic total occlusion, stenosis, thrombus, or plaque buildup. It is, therefore, understood that the lesions identified and classified in the context of the method may correspond to various types of medical issues that require intervention. Similarly, the term medical procedure, as used herein, may refer to a number of different types of surgeries or other interventions that may be taken in response to the identification of a lesion.

[0082] While the complexity metric (extracted at 320) is based at least partially on the corresponding images (retrieved at 300), in some embodiment, patient information beyond imaging data may be considered in defining complexity. As such, patient age or prior treatment history may be used to refine the complexify metric for a particular medical procedure 110a.

[0083] Figure 4 illustrates a scoresheet for use in embodiments of the method of FIG. 3. Classification of each lesion (at 360) may be based on a number of factors. In some embodiments, the classification value for any lesion determined to be a chronic total occlusion (CTO) is further based on at least one of a determination that an entry for the CTO is blunt, as opposed to tapered, a determination that plaque comprising the occlusion is calcified, an evaluation of bending in the CTO segment, and a length of the occlusion. Such an approach to classification is shown in the J-CTO score sheet provided as FIG. 4.

[0084] Figure 5 illustrates an example of scoring that could be used in the context of the method of FIG. 3. In some embodiments, classification of each lesion (at 360) may be based on a determination of a location of the lesion, a percentage of stenosis, and a number of vessels impacted. Such an approach to classification is discussed in more detail below, with respect to the CAD-RADS score sheet provided as FIG. 5.

[0085] In some embodiments, the classifications of individual lesions (at 360) may further inform the urgency metric generated (at 340).

[0086] In some embodiments, manual adjustments to complexity (at 320) of individual cases or urgency (at 340) may be applied prior to generation of a schedule. Such manual adjustments may be based on patient characteristics, such as age or treatment history, as noted above, or they may be based on clinical evaluations not otherwise accounted for. Accordingly, as discussed below, the schedule may continue to be optimized even if an evaluation of a particular medical procedure 110a requires manual adjustment.

[0087] The medical procedure characterizations (generated at 330) are then retained (370) in the case database 220 which then contains a queue of cases waiting to be assigned. The method further generates (380) a time estimation for each medical procedure 110a, 110b, 110c, based on the corresponding medical procedure characterization. The time estimation (generated at 380) is then typically incorporated into the medical procedure characterization or otherwise stored therewith (370) in the case database 220.

[0088] It is noted that once the medical procedure characterization is generated (at 330) and retained (at 370) in the case database 220, the characterization may continue to be augmented or modified. As such, while the discussion herein may discuss the storage of additional or newly generated data or metrics alongside the medical procedure characterization in the case database 220, it will be understood that all such data or metrics could similarly be considered to be integrated into the corresponding medical procedure characterization in the case database. Accordingly, the time estimation (generated at 380) may be generated at least partially based on the characterization. However, after generating the time estimation, the method incorporates such a time estimation into the characterization of the corresponding medical procedure. [0089] In some embodiments, the method includes further defining (390) a confidence metric for the time estimation (generated at 380). The confidence metric is based at least partially on the classification values for lesions associated with the medical procedure (assigned at 360). The confidence metric may then be incorporated into or stored with the medical procedure characterizations (370). As such, the medical procedure characterizations (generated at 330) may incorporate, be partially based on, or be stored with various time metrics providing both a time estimation (generated at 380) and, in some cases, a confidence metric (defined at 390). As such, the case database 220 may include, for each medical procedure 110a, 110b, 110c a projected time for the procedure and a level of confidence in that projected time.

[0090] Further, as noted above, the medical procedure characterizations (generated at 330) may further incorporate, be partially based on, or be stored with, the urgency metric (generated at 340), which may at least partially be based on a classification of a lesion (at 360) as a short-term risk, an evaluation of plaque stability, or how focused a location of the lesion is.

[0091] The medical procedure characterizations (generated at 330) may then comprise an estimate of the number and type of lesions to be treated which may in turn be used to generate a time estimation (at 380). The time estimate may be a sophisticated prediction based on the other aspects of the characterization, as noted above, and it may be based on an Al model, such as a convolutional neural network (CNN) as discussed below. Similarly, the time estimate may be based on historic treatment data stored in a database and processed using such a CNN or other predictive model, or it may be based on a decision tree. Alternatively, it may be based on simple predefined rules, such as, e.g., 45 minutes per simple lesion.

[0092] The method then proceeds to determine (400), at the available resource module 230, a set of resources available for performing the plurality of medical procedures 110a, 110b, 110c. The resources available include at least one operating room 240, such as a cathlab, and at least one medical team 250. The resources available may be limited to those available from a particular organizational unit, such as a particular cardiology department, hospital, or hospital network.

[0093] The available resource module 230 typically determines and/or stores, along with a listing of available resources, characteristics of each available resource. Accordingly, each medical team of the at least one medical team 250 available for allocation to a medical procedure may be associated with a set of skills. Such a set of skills may include, for example, a measure of experience for each medical team. Further, medical teams may have different compositions. As such, only certain procedures may require, for example, an anesthesiologist. Similarly, certain procedures may require the presence of specific individuals, such as a head surgeon, and as such, those individuals may be added to an existing medical team, or may be allocated as a resource separate from the medical team.

[0094] Similarly, the available resource module 230 may define multiple operating rooms 240, such as cathlabs, available, and each operating room may be associated with defined capabilities. For example, some procedures may require specialized equipment, such as an intravascular laser module, and such equipment may be available only in specific rooms.

[0095] Further, the available resource module 230 may define a time of availability for each available resource. For example, rooms 240 may be available at specific times or for specific amounts of time. Similarly, medical teams 250 may be available at specific times or for specific amounts of time. Rules may be associated with the availability of such medical teams 250 or rooms 240, such as the amount of time that each resource may be used for on a particular day or consecutively. Similarly, medical teams 250 may define preferences for specific types of procedures or for specific times of day.

[0096] The available resource module may further include in the set of resources a catalog of available consumables, such as guidewires, balloons, stents, and contrast agent. Such consumables may be associated with specific rooms 240, or they may instead be available for allocation.

[0097] The method then proceeds with allocating (410) at least one operating room 240 and at least one medical team 250 to each medical procedure 110a, 110b, 110c and generating a schedule (420) based on the time estimation associated with each medical procedure and the allocated operating room and medical team. Where each medical team 250 is associated with a set of skills and/or a measure of experience, the allocation of the medical team (at 410) to a specific medical procedure 110a may be based at least partially on the measure of experience.

[0098] For example, in some embodiments, the complexity metric (extracted at 320) associated with a medical procedure 110a may indicate a level of experience required for a medical team 250 to perform the corresponding procedure. As such, where a medical procedure 110a has a value for the complexity metric above a defined threshold, the scheduling module 260 may allow the allocation of a particular medical team 250 only if the medical team has a measure of experience above a defined threshold. Further, the level of experience may be associated with a specific type of medical procedure, thereby defining a particular medical team 250 as specialists. In some embodiments, multiple measures of experience may be associated with a particular medical team, with each measure of experience defining experience associated with a particular type of procedure.

[0099] Similarly, where each of several available operating rooms 240 are associated with defined capabilities, the allocation of an operating room to a particular medical procedure 110a by the scheduling module 260 may be based on a comparison between the corresponding medical procedure characterization and the capabilities associated with the operating room.

[00100] In embodiments in which an urgency metric (generated at 340) corresponds to a particular medical procedure 110a, the schedule may be generated (at 420) at least partially based on the urgency metric. Accordingly, a short-term risk associated with a particular lesion may result in the scheduling of the corresponding medical procedure 110a at an earlier time than such procedure would otherwise be scheduled.

[00101] The schedule generated (at 420) may further consider other characteristics associated with resources at the available resource module 230. For example, the schedule may balance urgency associated with a particular medical procedure 110a against the availability of a medical team 250 with the skills or measure of experience required to perform the procedure based on the associated complexity metric.

[00102] Further, where a confidence metric is defined (at 390), such a confidence metric may be used to determine scheduling (at 420) in order to increase predictability. For example, a medical procedure 110a associated with a low value for the confidence metric, and therefore associated with a large potential error around the time estimation (generated at 380) may be scheduled later in the day, so as to avoid a risk of having to reschedule later procedures.

[00103] Further, in some embodiments, the time estimation (generated at 380) may ultimately depend on which team of several available medical teams 250 are assigned to the particular medical procedure. For example, a team with a lot of experience with a particular type of medical procedure may take less time to complete the corresponding procedure than a team with less experience. Similarly, some surgeons or medical teams may simply work faster than others.

[00104] As such, the time estimation (generated at 380) may be a dimensionless work unit. Each medical team 250 may then be associated with at least one time modifier, such that the time estimation may be converted to units of time by modifying the work unit for a particular medical procedure based on the time modifier associated with a particular medical team.

[00105] For example, the time modifier may be a simple variable that can be multiplied with the dimensionless work unit in order to provide a value in units of time. As such, a team with an experienced surgeon who typically works quickly may have a lower value for such a time modifier than a team with a less experienced or slower surgeon.

[00106] In some embodiments, a medical team 250 may be associated with a plurality of distinct time modifiers. Each time modifier may be associated with a specific type or group of medical procedures. The scheduling module may then determine which of the plurality of distinct time modifiers is to be applied for a specific medical procedure based on the corresponding medical procedure characterizations (generated at 330).

[00107] In some embodiments, the allocation by the scheduling module 260 of the at least one operating room 240 and the at least one medical team 250 is at the same time as the generation of the schedule. Such allocation may then be performed based on an overall optimization metric. The scheduling module 260 may then attempt to minimize or maximize the overall optimization metric.

[00108] In some embodiments, such an overall optimization metric is based on weighted measures of skill level and preferences associated with individual medical teams 240 of the at least one medical team. The overall optimization metric may be minimized or maximized based on, for example, an implementation of a knapsack problem.

[00109] In some embodiments, the complexity metric extracted (at 320) from each image and associated with the corresponding medical procedure 110a, 110b, 110c is based on a predictive model. Similarly, the urgency metric (generated at 340), the classification value for a lesion (assigned at 360), the time estimation (generated at 380) and/or the confidence metric (defined at 390) may be based on a predictive model. Such a predictive model may be an Al model and may be based on a convolutional neural network (CNN). In some embodiments, the predictive model may be trained based on previous iterations of the method of generating a medical procedure schedule.

[00110] As such, over time, the complexity metric may generate more accurate time estimations, and/or the time estimations and the corresponding confidence metric may better predict which characteristics of a medical procedure are likely to reduce predictability. Accordingly, feedback loops may be implemented to determine whether estimated efforts and times were realistic and to adjust future estimates or to suggest optimizations in terms of team composition, prior room assignment, etc.

[00111] Similarly, coupled with long-term outcome data, success rates may be monitored and correlated with image based features generated and evaluated during the extraction of the complexity metric (at 320). Data generated during implementation of the scheduling module may be used in other ways as well. The expected real procedure costs can be estimated and compared to estimated income, such as by predicting Diagnosis Related Group (DRG) numbers.

[00112] The scheduling module 260 may monitor actual progress in real time. Accordingly, if a particular medical procedure 110a exceeds the estimated time, and such additional time was not already accounted for by the confidence metric, the method may proceed to reschedule subsequent procedures or otherwise optimize scheduling for remaining procedures.

[00113] The methods according to the present disclosure may be implemented on a computer as a computer implemented method, or in dedicated hardware, or in a combination of both. Executable code for a method according to the present disclosure may be stored on a computer program product. Examples of computer program products include memory devices, optical storage devices, integrated circuits, servers, online software, etc. Preferably, the computer program product may include non-transitory program code stored on a computer readable medium for performing a method according to the present disclosure when said program product is executed on a computer. In an embodiment, the computer program may include computer program code adapted to perform all the steps of a method according to the present disclosure when the computer program is run on a computer. The computer program may be embodied on a computer readable medium.

[00114] While the present disclosure has been described at some length and with some particularity with respect to the several described embodiments, it is not intended that it should be limited to any such particulars or embodiments or any particular embodiment, but it is to be construed with references to the appended claims so as to provide the broadest possible interpretation of such claims in view of the prior art and, therefore, to effectively encompass the intended scope of the disclosure.

[00115] All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the principles of the disclosure and the concepts contributed by the inventor to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.