PRIORITY CLAIM

[0001] The present application claims priority to and the benefit of Indian provisional application number 202141041217, filed September 14, 2021, the entire contents of which are hereby incorporated by reference and relied upon.

BACKGROUND

[0002] Only ten years ago, there were virtually no virtual assistants or chatbots. Since that time, tens to hundreds of artificial intelligence-based virtual assistances and chatbots have been integrated into devices such as smartphones, smart-speakers, televisions, automobiles, and virtually any other network-enabled device. Some of the more popular voice-enabled virtual assistants include Siri by Apple®, Google Now by Google®, Cortana by Microsoft®, Alexa by Amazon®, and Teneo by Artificial Solutions®. There are also many popular chatbots used by various companies for customer service and sales.

[0003] Currently, known virtual assistants and chatbots only have a single point of entry of information, namely speech or text provided by a user. Oftentimes without additional context, a user’s question or request may be interpreted by the virtual assistant/chatbot logic as being vague or undefined. In other instances, the virtual assistant/chatbot logic is programmed to ask a multitude of follow up questions to arrive at a perceived correct answer. An issue with this approach is that users can quickly become frustrated with the interaction if they believe the virtual assistant/chatbot does not understand the request. Further, users can easily become annoyed if they have to provide answers to numerous questions.

[0004] The above-known issues can especially be problematic in medical environments. For instance, a patient experiencing a medical issue or an issue with a medical device may not have the patience to progress through a long hierarchy of questions to arrive at a solution or recommended action. This is especially true for more imminent medical emergencies where time is of the essence. There is accordingly a need for a virtual assistant/chatbot that is more responsive and adaptive to patient issues.

SUMMARY

[0005] The example system, apparatus, and methods disclosed herein are configured to provide a virtual assistant/chatbot that is integrated into a medical device, such as a dialysis machine or an infusion pump. The virtual assistant/chatbot is configured to combine speech or textual questions/answers from patients with medical device data from the medical advice to more quickly converge upon an answer or a recommended action for a patient. The virtual assistant/chatbot may be used to predict patient complications and/or assist a patient regarding operation of the medical device.

[0006] The example virtual assistant/chatbot described herein is configured with a data structure defining a logical sequence of questions and answers for resolving issues and/or predicting patient complications. The sequence of questions and answers may be configured in a node arrangement such that certain answers lead to additional questions for addition information from a patient. However, instead of relying on answers from a patient for all of the questions, at least some of the questions may be answered automatically (or at least partially) using medical device data from the integrated medical device.

[0007] In an example, a patient may have damaged or improperly functioning kidneys. As a result, the patient may use a home or clinic-based dialysis machine to undergo dialysis treatments to remove waste products from blood. One common type of dialysis is peritoneal dialysis (“PD”), in which a cleansing fluid, referred to as peritoneal dialysis fluid, is moved into a patient’s peritoneal cavity of their abdomen via a catheter. The cleansing fluid absorbs waste products during a dwell period. After the dwell period has ended, the cleansing fluid is removed from the patient’s peritoneal cavity with the absorbed waste products, thereby compensating for the patient’s damaged kidneys.

[0008] Oftentimes, a PD machine is used to pump a prescribed volume of the cleansing fluid into the patient’s peritoneal cavity. The PD machine permits the cleansing fluid to remain in the patient during the dwell period. After the dwell period, the PD machine drains the cleansing fluid, with the waste products (including ultrafiltration), from the patient’s peritoneal cavity.

[0009] During a dialysis treatment, a patient may feel discomfort in their peritoneal cavity due to accumulating fluids. However, the patient does not realize the accumulating fluid and waste product is the source of their discomfort. The patient activates the virtual assistant/chatbot on the PD machine and conveys their discomfort. Instead of having to ask a multitude of follow up questions, the virtual assistant/chatbot accesses a memory device on the PD machine that stores medical device data, such as past and current treatment data (including fill volumes, drain volumes, and estimated ultrafiltration removed). Logic associated with the virtual assistant/chatbot analyzes the treatment data to determine that drain volumes are less than expected given the known dialysis fluid fill volumes, which indicates an accumulation of dialysis fluid in the peritoneal cavity. As a result, the logic causes the virtual assistant/chatbot to ask one or more follow up questions specifically directed to low drain volumes to confirm the determination. After a verbal or textual confirmation from the patient, the virtual assistant/chatbot may resolve the issue by increasing a drain duration, reducing a next fill volume, or transmitting a message to a clinician indicative of the excess residual volume remaining in the patient.

[0010] In another example, the virtual assistant/chatbot is configured to ask a patient a series of health-related questions during or after a dialysis treatment session. Programmed logic is configured to use the answers in conjunction with medical device data from a medical device to determine probabilities that a patient is experiencing or may soon experience a medical complication, such as peritonitis, sepsis, a catheter exit site dislodgement/infection, a cardiac issue, or other medical issue. In some instances, the programmed logic is configured to use previous patient answers (e.g., patient data) such as their diet, activity level, or demographics are part of the medical complication determination. Moreover in some instances, the programmed logic is configured to compare probabilities of different medical complications to previously calculated probabilities to determine a trend. The programmed logic may generate an alert or other message for the patient or a clinician (for display on the medical device or smartphone) that is indicative of an increase in the medical complication probability or a predicted onset of the medical complication. [0011] In light of the disclosure herein and without limiting the disclosure in any way, in a first aspect of the present disclosure, which may be combined with any other aspect listed herein unless specified otherwise, a dialysis apparatus includes a pump configured to pump dialysis fluid, at least one valve configured to control a flow of dialysis fluid pumped by the pump, a user interface configured to enable data to be entered and information to be provided, and a control unit operably connected to the pump, the at least one valve, and the user interface. The dialysis apparatus also includes a first memory device storing (i) a treatment parameters for a dialysis prescription, (ii) results from performing one or more dialysis treatments (e.g., treatment data), (iii) diagnostic information related to the dialysis apparatus, and (iv) a current status of the dialysis apparatus. The dialysis apparatus further includes a second memory device storing an issue resolution data structure for an adaptive virtual assistant or chatbot, the data structure including a plurality of potential issues related to dialysis or operation of the dialysis apparatus, each issue including a resolution action or a hierarchy of questions and possible answers that are associated with a resolution action at a lowest level. The dialysis apparatus additionally includes a third memory device storing machine-readable instructions, which when executed by the control unit, cause the control unit to receive an indication via the user interface to initiate the adaptive virtual assistant or chatbot, provide a prompt for a patient of the dialysis apparatus to enter information, receive information from the patient related to an issue they are experiencing, and select an issue of the plurality of potential issues having a greatest match probability by comparing the received information from the patient and at least one of (i) to (iv) to the plurality of potential issues. If the issue includes a resolution action, the control unit causes the resolution action to be provided to the patient via the user interface. However, if the issue includes a hierarchy of questions, the control unit progresses through the hierarchy of questions with one or more prompts to receive further information and use at least one of (i) to (iv) in conjunction with patient answers until a resolution action is identified, and causes the identified resolution action to be provided to the patient via the user interface.

[0012] In accordance with a second aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the comparison includes key word matching between the received information and keywords associated with each of the plurality of issues.

[0013] In accordance with a third aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the resolution action includes a recommendation for at least one of setting up the dialysis apparatus, setting up a treatment to be performed by the dialysis apparatus, resolving a priming issue, addressing a warning, handling an alarm, or responding to an error.

[0014] In accordance with a fourth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the resolution action includes at least one of ordering a dialysis solution bag, ordering a dialysis cassette, calling or transmitting a message indicative of a need for medical assistance, making a reservation for a clinical appointment, modifying at least one of the dialysis treatment parameters, transmitting an alert to a clinician indicative of the corresponding issue, or causing the dialysis apparatus to shut down.

[0015] In accordance with a fifth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the treatment parameter includes at least one of a total amount of ultrafiltration to remove, a total fill volume, a treatment time, a fill amount per treatment cycle, a dwell duration, or a dialysis solution dextrose concentration.

[0016] In accordance with a sixth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, at least one of the second memory device or the third memory device is remote from the dialysis apparatus in a cloud computing environment, such that the control unit accesses the at least one of the second memory device or the third memory device via a network.

[0017] In accordance with a seventh aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the information is received as speech and the third memory device stores additional machine-readable instructions, which when executed by the control unit, cause the control unit to convert the speech to text. [0018] In accordance with an eighth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, a medical fluid system includes a server configured to host an adaptive virtual assistant or chatbot that addresses a plurality of potential issues related to dialysis or operation of medical devices, each issue including a resolution action or a hierarchy of questions and possible answers that are associated with a resolution action at a lowest level. The medical fluid system also includes a medical device communicatively coupled to the server via a network. The medical device comprises a pump configured to pump medical fluid, at least one valve configured to control a flow of medical fluid pumped by the pump, a user interface configured to enable data to be entered and information to be provided, a control unit operably connected to the pump, the at least one valve, and the user interface, and a first memory device storing (i) a treatment parameters for a medical fluid treatment prescription, (ii) results from performing one or more medical fluid treatments, (iii) diagnostic information related to the medical device, and (iv) a current status of the medical device. The medical device also includes a second memory device storing machine-readable instructions, which when executed by the control unit, cause the control unit to receive an indication via the user interface to initiate the adaptive virtual assistant or chatbot, provide a prompt for a patient of the medical device to enter information, receive information from the patient related to an issue they are experiencing, and transmit at least one of (i) to (iv) and the information from the patient to the server. The example server is configured to select an issue of the plurality of potential issues having a greatest match probability by comparing the received information from the patient and at least one of (i) to (iv) to the plurality of potential issues. If the issue includes a resolution action, the server causes the resolution action to be provided to the patient via the user interface. However, if the issue includes a hierarchy of questions, the server progresses through the hierarchy of questions with one or more prompts to receive further information and use at least one of (i) to (iv) in conjunction with patient answers until a resolution action is identified, and causes the identified resolution action to be provided to the patient via the user interface.

[0019] In accordance with a ninth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the comparison includes key word matching between the received information and keywords associated with each of the plurality of issues.

[0020] In accordance with a tenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the resolution action includes a recommendation for at least one of setting up the medical device, setting up a treatment to be performed by the medical device, resolving a priming issue, addressing a warning, handling an alarm, or responding to an error.

[0021] In accordance with an eleventh aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the resolution action includes at least one of ordering a medical fluid solution bag, ordering a medical fluid cassette, calling or transmitting a message indicative of a need for medical assistance, making a reservation for a clinical appointment, modifying at least one of the medical fluid treatment parameters, transmitting an alert to a clinician indicative of the corresponding issue, or causing the medical device to shut down.

[0022] In accordance with a twelfth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the information is received as speech and the second memory device stores additional machine-readable instructions, which when executed by the control unit, cause the control unit to convert the speech to text.

[0023] In accordance with a thirteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the medical device includes at least one of a peritoneal dialysis machine, a hemodialysis machine, a continuous renal replacement therapy (“CRRT”) machine, an infusion pump, or a patient-controlled analgesia (“PCA”) machine.

[0024] In accordance with a fourteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the server is cloudbased and communicatively coupled to a plurality of medical devices for providing issue resolution using the adaptive virtual assistant or chatbot.

[0025] In accordance with a fifteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, a dialysis apparatus comprises a pump configured to pump dialysis fluid, at least one valve configured to control a flow of dialysis fluid pumped by the pump, a user interface configured to enable data to be entered and information to be provided, a control unit operably connected to the pump, the at least one valve, and the user interface, and a first memory device storing (i) a treatment parameters for a dialysis prescription, (ii) results from performing one or more dialysis treatments, (iii) diagnostic information related to the dialysis apparatus, and (iv) a current status of the dialysis apparatus. The dialysis apparatus also includes a second memory device storing an medical complication data structure for an adaptive virtual assistant or chatbot, the data structure including a plurality of questions related to medical complications or symptoms, each question having at a plurality of possible answers, each answer having a probability component for one or more medical complications, at least some of the answers including a hierarchy of questions and possible answers having probability components for medical complications. The dialysis apparatus further includes a third memory device storing machine-readable instructions, which when executed by the control unit, cause the control unit to receive an indication via the user interface to initiate the adaptive virtual assistant or chatbot, provide a prompt for a patient of the dialysis apparatus to enter information, receive information from the patient related to an issue they are experiencing, determine probability component values for the medical complications using the received information from the patient and at least one of (i) to (iv), combine the probability component values associated with a same medical complication, and compare the combined probability component values for each medical complication to a threshold and generate a message for display on the user interface for each medical complication that is above its respective threshold. If a medical complication cannot be identified, the control unit progresses through the hierarchy of questions with one or more prompts to receive further information and uses at least one of (i) to (iv) in conjunction with patient answers until a medical complication is identified, and causes the identified medical complication to be provided to the patient via the user interface.

[0026] In accordance with a sixteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the third memory device stores additional machine-readable instructions, which when executed by the control unit, cause the control unit to transmit the message indicative of the medical complication to a server or a clinician device.

[0027] In accordance with a seventeenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the medical complications include peritonitis, catheter exit site infection, diabetes, sepsis, a catheter placement issue, a cardiac issue, and mental health degradation.

[0028] In accordance with an eighteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, at least one of the second memory device or the third memory device is remote from the dialysis apparatus in a cloud computing environment, such that the control unit accesses the at least one of the second memory device or the third memory device via a network.

[0029] In accordance with a nineteenth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the third memory device stores additional machine-readable instructions, which when executed by the control unit, cause the control unit to cause the user interface to display one or more graphics illustrating possible answers.

[0030] In accordance with a twentieth aspect of the present disclosure, which may be used in combination with any other aspect listed herein unless stated otherwise, the third memory device stores additional machine-readable instructions, which when executed by the control unit, cause the control unit to determine trends of the combined probability component values for each medical complication over at least one of three days, a week, two weeks, a month, or six months, and generate the message for display on the user interface for each medical complication that has a trend that is above a respective second threshold.

[0031] In a twenty -first aspect of the present disclosure, any of the structure, functionality, and alternatives disclosed in connection with any one or more of Figs. 1 to 13 may be combined with any other structure, functionality, and alternatives disclosed in connection with any other one or more of Figs. 1 to 13. [0032] In light of the present disclosure and the above aspects, it is therefore an advantage of the present disclosure to provide an adaptive virtual assistant/chatbot that uses data from a medical device to more quickly determine a need of a patient.

[0033] It is another advantage of the present disclosure to use data from a medical device, such as a dialysis machine, in conjunction with textual or spoken inputs from a patient to resolve an issue with the medical device or a treatment being provided by the medical device.

[0034] It is yet another advantage of the present disclosure to provide an adaptive virtual assistant/chatbot that uses data from a medical device in conjunction with textual or spoken inputs from a patient to predict patient medical complications.

[0035] Additional features and advantages are described in, and will be apparent from, the following Detailed Description and the Figures. The features and advantages described herein are not all-inclusive and, in particular, many additional features and advantages will be apparent to one of ordinary skill in the art in view of the figures and description. Also, any particular embodiment does not have to have all of the advantages listed herein and it is expressly contemplated to claim individual advantageous embodiments separately. Moreover, it should be noted that the language used in the specification has been selected principally for readability and instructional purposes, and not to limit the scope of the inventive subject matter.

BRIEF DESCRIPTION OF THE FIGURES

[0036] Fig. 1 illustrates an example adaptive medical system, according to an example embodiment of the present disclosure.

[0037] Figs. 2 and 3 illustrate alternative embodiments of the adaptive medical system of Fig. 2, according to example embodiments of the present disclosure.

[0038] Fig. 4 is a diagram illustrative of the machine-readable instructions for the adaptive medical system of Figs. 1 to 3, according to an example embodiment of the present disclosure.

[0039] Fig. 5 is a diagram of different possible components of medical device data that is related to the adaptive medical system of Figs. 1 to 3, according to an example embodiment of the present disclosure. [0040] Fig. 6 is a diagram of different possible components of patient data that is relayed to the adaptive medical system of Figs. 1 to 3, according to an example embodiment of the present disclosure.

[0041] Fig. 7 illustrates a diagram of an example procedure for resolving an issue related to a dialysis machine of the adaptive medical system of Figs. 1 to 3, according to an example embodiment of the present disclosure.

[0042] Fig. 8 is a diagram illustrative of the virtual assistant/chatbot logic for issue resolution, according to an example embodiment of the present disclosure.

[0043] Fig. 9 illustrates a diagram of an example procedure for predicting and/or diagnosing one or more medical complications in a patient that is using a dialysis machine, according to an example embodiment of the present disclosure.

[0044] Figs. 10 and 11 are diagrams illustrative of a virtual assistant/chatbot logic for determining a medical complication of a patient, according to an example embodiment of the present disclosure.

[0045] Fig. 12 shows an example graphic that may be displayed by a user interface of a dialysis machine in conjunction with a question regarding possible medical complications, according to an example embodiment of the present disclosure.

[0046] Fig. 13 is a diagram of a user interface of a dialysis machine displaying content related to the detection of a peritonitis medical complication, according to an example embodiment of the present disclosure.

DETAILED DESCRIPTION

[0047] An adaptive virtual assistant/chatbot for a medical device is disclosed. The virtual assistant/chatbot is configured to be integrated within the medical device. The integration may include storing virtually all of the virtual assistant/chatbot logic within the medical device. Alternatively, this may include placing at least some of the virtual assistant/chatbot within a networked cloud-based server network that interfaces with a processor at the medical device. The integration of the virtual assistant/chatbot logic with a medical device enables medical device data and/or patient data to be used in the analysis of an interaction with a patient to more quickly and accurately converge upon an appropriate action, recommendation, or medical complication prediction.

[0048] As described herein, the medical device data and/or patient data is acquired by the medical device during one or more treatment sessions with a patient. The medical device data may include programmed treatment parameters specified in a treatment prescription that detail how a medical device is to operate. The medical device data may also include medical treatment parameters that quantify results from a treatment that has been performed, such as a total volume of fluid drained from a patient’s peritoneal cavity during a treatment. In some instances, the medical device data may include device diagnostic information, such as indications of occlusions or pump operational issues. The medical device data may further include a current status of a medical device, such as whether the device is performing a priming operation or a patient fill cycle of a dialysis treatment.

[0049] Reference is also made to patient data, which may also be used by the virtual assistant/chatbot logic. Patient data refers to information that is received from a patient during one or more sessions. For instance, during or after a treatment session, the virtual assistant/chatbot logic is configured to ask a patient questions about demographics and/or medical history. The virtual assistant/chatbot logic is configured to store patient responses to the questions. Additionally or alternatively, the virtual assistant/chatbot logic may access a hospital server to access an electronic medical record (“EMR”) of the patient to acquire at least some of the patient data.

[0050] As disclosed herein, the virtual assistant/chatbot logic is configured to use one or more computational structures specified in a data structure that define an interaction with a patient based on one or more patient answers. It should be appreciated that virtually any data structure may be used. For example, questions, answers, and results may be stored as nodes and links in a graph database structure. In another example, questions, answers, and results may be stored in a relational, sequential, or hierarchical data structure. The linkages between answers and questions is configured to guide the virtual assistant/chatbot logic to select which subsequent action is to be performed. [0051] The use of readily available medical device data and/or patient data enables the virtual assistant/chatbot logic disclosed herein to more accurately and quickly identify a recommended action or predicted medical complication while reducing answers needed from a patient. In some instances, patient requests or questions that may initially have low context or seem unclear are more readily apparent by the virtual assistant/chatbot logic analyzing the medical device data and/or patient data in conjunction with the request. In an example, a patient may start an interaction with the virtual assistant/chatbot logic disclosed herein by stating that they “Fill Full”. By itself, this statement is unclear and could refer to food consumption, constipation, or their peritoneal cavity. However, when the virtual assistant/chatbot logic receives such a statement during a dwell phase of a peritoneal dialysis treatment, there is a high probability that the patient is referring to a feeling of their peritoneal cavity being overfilled. The uniqueness of the virtual assistant/chatbot logic disclosed herein is that the current status of the dialysis device is known through the medical device data. Accordingly, the integrated virtual assistant/chatbot logic is able to analyze the medical device data in conjunction with the patient request to more quickly determine a patient needs help with overfilling of dialysis fluid. In contrast, known virtual assistant/chatbot systems would need to go through a round of 20-questions before even remotely being able to identify the patient’s medical condition.

Adaptive Medical Environment

[0052] Referring now to the drawings, Fig. 1 illustrates an example adaptive medical system 100, according to an example embodiment of the present disclosure. The adaptive medical system 100 in the illustrated embodiment includes a dialysis machine 102 (e.g., a medical device) configured to provide renal failure therapy to one or more patients. Renal failure therapy helps a patient balance water and minerals. Renal failure therapy also helps excrete daily metabolic load by removing a patient’s toxic end products of nitrogen metabolism (urea, creatinine, uric acid, and others), which accumulate in blood and tissue. Renal failure therapy for the replacement of kidney function is critical to many people because the treatment is lifesaving.

[0053] In some examples, the dialysis machine 102 is a peritoneal dialysis (“PD”) machine. Here, the dialysis machine 102 is configured to infuse a dialysis solution, also called dialysis fluid or renal failure therapy fluid into a patient’s peritoneal cavity via a catheter. The dialysis fluid contacts the peritoneal membrane of the peritoneal cavity for a period of time, referred to as a dwell period. Waste, toxins and excess water pass from the patient’s bloodstream, through the peritoneal membrane and into the dialysis fluid due to diffusion and osmosis, i.e., an osmotic gradient occurs across the membrane. An osmotic agent in dialysis provides the osmotic gradient. The used or spent dialysis fluid is drained from the patient, removing waste, toxins and excess water from the patient. This cycle is repeated, e.g., multiple times.

[0054] There are various types of peritoneal dialysis therapies, including continuous ambulatory peritoneal dialysis (“CAPD”), automated peritoneal dialysis (“APD”), and tidal flow dialysis and continuous flow peritoneal dialysis (“CFPD”). CAPD is a manual dialysis treatment. Here, the patient manually connects an implanted catheter to a drain to allow used or spent dialysis fluid to drain from the peritoneal cavity. The patient then connects the catheter to a bag of fresh dialysis fluid to infuse fresh dialysis fluid through the catheter and into the patient. The patient disconnects the catheter from the fresh dialysis fluid bag and allows the dialysis fluid to dwell within the peritoneal cavity, wherein the transfer of waste, toxins and excess water takes place. After a dwell period, the patient repeats the manual dialysis procedure, for example, four times per day, each treatment lasting about an hour. Manual peritoneal dialysis requires a significant amount of time and effort from the patient, leaving ample room for improvement.

[0055] Automated peritoneal dialysis (“APD”) is similar to CAPD in that the dialysis treatment includes drain, fill and dwell cycles. APD machines, such as the dialysis machine 102, however, perform the cycles automatically, typically while the patient sleeps. APD machines free patients from having to perform the treatment cycles manually and from having to transport supplies during the day. APD machines connect fluidly to an implanted catheter, to a source or bag of fresh dialysis fluid and to a fluid drain. APD machines pump fresh dialysis fluid from a dialysis fluid source, through the catheter and into the patient’s peritoneal cavity. APD machines also allow for the dialysis fluid to dwell within the cavity and for the transfer of waste, toxins and excess water to take place. The source may include multiple sterile dialysis fluid bags.

[0056] APD machines pump used or spent dialysis fluid from the peritoneal cavity, though the catheter, and to the drain. As with the manual process, several drain, fill and dwell cycles occur during dialysis. A “last fill” occurs at the end of APD and remains in the peritoneal cavity of the patient until the next treatment.

[0057] In some embodiments, the dialysis machine 102 may be configured to perform hemodialysis (“HD”). During HD, the dialysis machine 102 is configured to use diffusion to remove waste products from a patient’s blood. A diffusive gradient occurs across the semi- permeable dialyzer between a patient’ s blood and an electrolyte solution called dialysate or dialysis fluid to cause diffusion. Hemofiltration (“HF”) is an alternative renal replacement therapy that relies on a convective transport of toxins from the patient’s blood. HF is accomplished by adding substitution or replacement fluid to the extracorporeal circuit during treatment (typically ten to ninety liters of such fluid). The substitution fluid and the fluid accumulated by the patient in between treatments is ultrafiltered over the course of the HF treatment, providing a convective transport mechanism that is particularly beneficial in removing middle and large molecules (in hemodialysis there is a small amount of waste removed along with the fluid gained between dialysis sessions, however, the solute drag from the removal of that ultrafiltrate is not enough to provide convective clearance).

[0058] Hemodiafiltration (“HDF”) is a treatment modality that combines convective and diffusive clearances. HDF uses dialysis fluid flowing through a dialyzer, similar to standard hemodialysis, to provide diffusive clearance. In addition, substitution solution is provided directly to the extracorporeal circuit, providing convective clearance.

[0059] The example dialysis machine 102 may be located in a center, a hospital, or a patient’s home. A trend towards home dialysis exists today in part because home dialysis can be performed daily, offering therapeutic benefits over in-center dialysis treatments, which occur typically bi- or tri-weekly. Studies have shown that frequent treatments remove more toxins and waste products than a patient receiving less frequent but perhaps longer treatments. A patient receiving treatments more frequently does not experience as much of a down cycle as does an incenter patient, who has built-up two or three days’ worth of toxins prior to treatment. In certain areas, the closest dialysis center can be many miles from the patient’s home causing door-to-door treatment time to consume a large portion of the day. Home dialysis may take place overnight or during the day while the patient relaxes, works or is otherwise productive. Much of the appeal of a home treatment for the patient revolves around the lifestyle flexibility provided by allowing the patient to perform treatment in his or her home largely according to his or her own schedule.

[0060] Any of the above dialysis modalities performed by the dialysis machine 102 may be run on a scheduled basis and may require a start-up procedure. For example, dialysis patients typically perform treatment on a scheduled basis, such as every other day, daily, etc. Dialysis treatment machines typically require a certain amount of time before treatment for setup, for example, to run a priming and/or disinfection procedure. During a priming procedure, a fluid is pumped through one or more dialysis tubes/lines and/or cassettes to remove air and/or in-line particulates. Priming dialysis tubes/lines and/or cassettes prevents air and/or the particulates from coming into contact with the patient.

[0061] The dialysis machine 102 in the illustrated embodiment includes at least one pump 104 configured to move fluid from a fluid source to a patient tube. The pump 104 may include any type of pump, including a peristaltic pump, a rotary pump, a gear pump, a linear actuator pump, a diaphragm pump, etc. The pump 104 may be operated to prime a patient tube with dialysis fluid. The pump 104 may also be operated to provide dialysis fluid from the fluid source to a patient when the patient tube is connected to a catheter that is inserted into a patient’s peritoneal cavity. Priming may alternatively or additionally be performed using gravity where, for example, a source of fluid 108 is provided at head height and permitted to flow through one or more tubes.

[0062] In some embodiments, the dialysis machine 102 includes a disposable cassette, which is connected fluidly to the tubes. The cassette may include one or more flexible membranes or chambers that operate with valves 106 and/or pumps 104 in the dialysis machine 102. Priming may include moving fluid through the disposable cassette in addition to the one or more tubes.

[0063] The fluid source 108 may include one or more containers of pre-mixed dialysis fluid. In some embodiments, the fluid source 108 may include containers or reservoirs of concentrate that have been mixed with pure water to form dialysis fluid. Additionally or alternatively, the fluid source 108 may include an on-line source, such as a source of purified water that is mixed with one or more concentrates to form dialysis fluid. Moreover, in some examples, the fluid source 108 may include a fluid preparation device that provides prepared dialysis fluid to the dialysis machine 102 via one or more fluid connections. [0064] The example dialysis machine 102 of Fig. 1 also includes a processor 110 and a memory 112. The processor 110 may include any type of device capable of processing inputs and performing one or more calculations to determine one or more outputs. The processor 110 may include a microcontroller, a controller, an application specific integrated circuit (“ASIC”), a central processing unit included on one or more integrated circuits, etc. The memory 112 may include any volatile or non-volatile data/instruction storage device. The memory 112 may include, for example, flash memory, random-access memory (“RAM”), read-only memory (“ROM”), Electrically Erasable Programmable Read-Only Memory (“EEPROM”), etc.

[0065] In the illustrated example, the dialysis machine 102 includes a first memory 112a stores medical device data and patient data (collectively referred to a medical information 114). A second memory 112b stores virtual assistant/chatbot logic 116, which may include one or more data structures and/or instructions that provide for an interaction with a patient to access whether a patient is experiencing an issue or may experience a medical complication. In some instances, the second memory 112b stores first virtual assistant/chatbot logic for accessing patient medical complications and second virtual assistant/chatbot logic for addressing a patient issue.

[0066] A third memory 112c is configured to store one or more instructions 118 that are executable by the processor 110 to cause the processor 110 to perform operations disclosed herein. The instructions 118 may be part of one or more software programs or applications. References herein to the processor 110 or the virtual assistant/chatbot logic 116 being configured to perform an operation may include embodiments where the memory 112c stores instructions 118 that are configured to cause the processor 110 and/or the virtual assistant/chatbot logic 116 to perform the described operation.

[0067] The example memory 112c is configured to store instructions 118 that cause the processor 110 to operate the dialysis machine 102. The operations performed by the processor 110 include providing control signals or instructions to the pump 102, which cause the pump 102 to move dialysis fluid from the fluid source 108 to a patient tube during a priming sequence or during a dialysis treatment. The operations performed by the processor 110 also include actuating the one or more valves 106 and receiving feedback signals. [0068] The example processor 110 is also configured to transmit one or more messages to a user interface 120 of the dialysis machine 102 for displaying or otherwise conveying information on a display screen, such as a touchscreen. The processor 110 may cause the user interface 120 to display instructions to a patient for preparing the dialysis machine 102 for a treatment, including actions to prepare for a priming sequence. The user interface 120 may also display or otherwise convey indications indicative of alert conditions. The user interface 120 may include a touchscreen overlay and/or electromechanical actuators, buttons, and/or switches to enable an operator to input information. The input may include a prompt from an operator to begin a priming sequence or a dialysis treatment.

[0069] As shown in Fig. 1, the user interface 120 includes a microphone 122 and a speaker 124. The microphone 122 is configured to receive and process sound waves into audio signals or digital data. The speaker 124 is configured to convert audio signals or digital data into audible voice commands or sounds. In some embodiments, the microphone 122 and/or the speaker 124 may be located outside of the user interface 120, such as within a housing of the dialysis machine 102. Further, while only one microphone 122 and speaker 124 are shown, the dialysis machine 102 may include two or more microphones 122 and/or speakers 124.

[0070] It should be appreciated that the dialysis machine 102 may include additional components for therapy preparation and/or performing dialysis therapies. The additional components may include pump actuators, compressors pneumatic equipment, valve actuators, heaters, online fluid generation equipment, fluid pressure sensors, fluid temperature sensors, conductivity sensors, air detection sensors, blood leak detection sensors, filters, dialyzers, balance chambers, sorbent cartridges, etc. In addition, the dialysis machine 102 may include one or more network connections (e.g., an Ethernet connection) to enable the processor 110 to receive data/prescriptions and transmit dialysis therapy status information to a remote or centralized server 128 via a network (e.g., the Internet) 130.

[0071] As shown in Fig. 1, the dialysis machine is communicatively coupled to the server 128 via the network 130. As disclosed herein, the server 128 includes any distributive computing environment (e.g., a cloud computing system), a workstation, a computer, etc. configured to remotely provide control and/or assistance to the dialysis machine 102. For example, the server 128 may transmit dialysis prescriptions and/or updates to the dialysis prescriptions to the dialysis machine 102 via the network 130. As disclosed herein, in some embodiments, the server 128 may host at least some of the virtual assistant/chatbot logic 116 or provide remote updates for the virtual assistant/chatbot logic 116.

[0072] The server 128 is communicatively coupled to one or more clinician device 132. While only one clinician device 132 is shown, in other embodiments the server 128 may be connected to tens or hundreds of clinician devices with a wired and/or wireless connection. The clinician device 132 is configured to provide an interface for viewing information hosted by the server 128. The interface may also enable a user to enter information into the clinician device 132 and/or interact with the dialysis machine 102 via the server 128. For example, a user interface clinician device 132 may be configured to display alerts and/or or other messages generated by the processor 110 of the dialysis machine 102 and/or the server 128. The user interface may enable a user to respond to the alert or message, thereby addressing the alarm or message.

[0073] The example network 130 may include any wide area network (e.g., the Internet), any local area network, any cellular network, or combinations thereof. The network 130 may include one or more routes, switches, and/or gateways across public and/or private network partitions. In some embodiments, the network 130 may be a local area network where the dialysis machine 102 and the server 128 are included within a same enterprise system.

[0074] Fig. 2 is another diagram of the adaptive medical system 100, according to an example embodiment of the present disclosure. In this example, the second memory 112b storing the virtual assistant/chatbot logic 116 is located at the server 128 instead of at the dialysis machine 102. Additionally, a first portion of the third memory 112c storing first instructions 118a are located at the dialysis machine 102, while a second portion of the third memory 112d storing second instructions 112b are located at the server 128.

[0075] In this illustrated configuration, the server 128 is configured to remotely provide the virtual assistant/chatbot logic 116 for the dialysis machine 102. To enable this configuration, the processor 110 at the dialysis machine 102 is programmed via the first instructions 118a to operate as an interface for the virtual assistant/chatbot. In other words, the dialysis machine 102, via the user interface 120, provides textual or audible questions to a patient and receives textual or spoken responses. The received responses are transmitted via the processor 110 to the server 128, which applies the virtual assistant/chatbot logic 116 to determine a follow-up question, recommended action, or other response. The server 128 transmits the response to the dialysis machine 102 via the network 130, which then either provides the response via the user interface 120 and/or updates operating parameters for a current and/or subsequent dialysis treatments.

[0076] In this embodiment, the medical device data and patient data, referred to as the medical information 114, is still stored locally at the first memory 112a. Thus, the dialysis machine 102 may transmit the medical information 114 to the server 128 to enable the virtual assistant/chatbot logic 116 to determine a next response or recommended action. In other instances, the server 128 is configured to access the first memory 112a for reading the medical information 114 to make a determination of a subsequent action to perform.

[0077] In some embodiments, the server 128 may include one or more APIs for receiving the medical information 114 and/or patient provided speech and/or text from the dialysis machine 102. In these examples, the processor 110 of the dialysis machine 102 is configured via the instructions 118 with addresses and/or port numbers of the APIs. Certain APIs may be defined for issue resolution, medical compilation prediction, and/or patient queries. The server 128 may also include one or more APIs through which response messages, treatment parameter adjustments, and/or recommended actions may be transmitted to the dialysis machine 102.

[0078] Fig. 3 is another diagram of the adaptive medical system 100, according to an example embodiment of the present disclosure. In this example, the first memory 112a is located at the server 128. As such, the virtual assistant/chatbot logic 116 has local access to the medical information 114 for determining a next step of the virtual interaction with a patient. The dialysis machine 102 may periodically transmit the medical information 114 to the server 102 during a dialysis treatment session, after a session has been completed, or at periodic time intervals (e.g., every hour, every ten minutes, etc.).

[0079] In some embodiments, the server 128 and the dialysis machine 102 may each include a first memory 112a. The dialysis machine 102 may include a first memory 112a for storing medical device data that is related to programming and/or operating one or more dialysis treatments. Additionally, the server 128 includes a first memory 112a for storing patient data, which may be determined through virtual interactions with the patient over multiple treatment sessions. In some instances, the server 128 may be communicatively coupled to a hospital information system. The server 128 uses this connection to obtain the medical information 114 from one more EMRs associated with the patient, one or more pharmacies, one or more medical laboratories, etc. There is virtually no limit to the number of information sources that may be connected to the server 128 for obtaining, aggregating, or otherwise having access to the medical information 114.

[0080] While Fig. 1 shows one dialysis machine 102, it should be appreciated that adaptive medical system 100 may include a plurality of dialysis machines 102 and/or other medical device types. For instance, the system 100 may include home-based dialysis machines 102 and/or clinicbased dialysis machines 102. In such a configuration, the server 128 may include a cloud-based computing system that provides a virtual assistant/chatbot service for the plurality of dialysis machines 102.

Virtual Assistant/Chatbot Embodiment

[0081] Fig. 4 is a diagram illustrative of the instructions 118 of Figs. 1 to 3, according to an example embodiment of the present disclosure. As discussed above, the instructions 118 are stored in the third memory 112c and specify how the virtual assistant/chatbot logic 116 are to operate. The instructions 118 are used by the processor 110 of the dialysis machine 102 and/or the server 128 to perform operations of the virtual assistant/chatbot logic 116. The blocks shown in Fig. 4 represent certain operations defined by the instructions 118. In other embodiments, some of the blocks may be combined, further partitioned, or include additional blocks.

[0082] The example instructions 118 include a speech-text conversion module 402, which is operationally connected to the microphone 122 of the user interface 120 of the dialysis machine 102. The speech-text conversion module 402 receives digital data and/or analog signals that include recorded human speech. The speech-text conversion module 402 is configured to convert the digital data and/or analog signals to text using one or more speech-to-text algorithms.

[0083] The speech-text conversion module 402 transmits the converted text to a language processing module 404, which is configured to use one or more algorithms to amend the received text based on known or learned accents, speech, or slang of a user. The language processing module 404 may include a library of known accents, speech, and/or slang for different types of users. The language processing module 404 selects appropriate textual modifiers based on how well a patient matches certain accents, speech, and/or slang. In some instances, the language processing module 404 may use one or more machine learning algorithms for identifying and/or modifying text based on the identified accent, speech, or slang of a user. The language processing module 404 outputs modified text that takes into account a user’s accent, speech, or slang. For instance, the language processing module 404 may receive a textual input from the speech-text conversion module 402 that includes combinations of vowels, constants, and breaks. After filtering through the language processing module 404, the string of vowels, constants, and breaks is modified into textual words and/or phrases, which are input into a speech recognition module 406.

[0084] The speech recognition module 406 is configured to operate one or more natural language processing algorithms to determine a meaning of received words or phrases. The speech recognition module 406 in some embodiments identifies a meaning of a string of words or phrases, the with identified meaning being stored as metadata, in separate data fields, or otherwise appended to the words and/or phrases. The speech recognition module 406 may analyze words or phrases to identify that a question is being asked, and a subject of the question. In this example, the speech recognition module 406 appends that the words or phrases correspond to a question and keywords that are associated with the question. The processing performed by the speech recognition module 402 adds formatted information that enables subsequent analysis based on more defined linguistic parameters.

[0085] In some embodiments, the speech recognition module 406 is configured to search for certain keywords for starting a virtual session with a patient. If a virtual session is not already in progress, the speech recognition module 406 listens or otherwise processes text and phrases for certain keywords or phrases that are indicative to start a session. For example, the virtual assistant/chatbot may be called “Claria”. Accordingly, the speech recognition module 406 searches for the term “Claria” or similar spellings. If a match is made, the speech recognition module 406 begins a virtual session with a patient and processes the phrase that includes the “Claria” term. At this point, the speech recognition module 406 processes subsequent words and/or phrases as part of the virtual conversation. However, if a match is not made, the speech recognition module 406 discards the text from further processing so as to refrain from recording other patient conversations or ambient room sounds. It should be appreciated that the virtual assistant/chatbot may also be activated by a patient by selecting a corresponding icon that is displayed by the user interface 120 of the dialysis machine 102.

[0086] The example instructions 118 of Fig. 4 also define an adaptive engine 408 that is configured to apply text from the speech recognition module 406 and/or the medical information 118 to the virtual assistant/chatbot logic 116. The potential inputs into the adaptive engine 408 accordingly include text from the speech recognition module 406 and/or text from a chatbot program provided by the user interface 120 of the dialysis machine 102. The inputs also include the medical information 118 from the first memory 112a. The example adaptive interface 408 includes an input interface 410 that is configured to receive text and/or other inputs entered into the user interface 120 of the dialysis machine 102. The inputs may include selection of an icon that causes the adaptive interface 408 to begin a virtual interactive session with a patient. The inputs may also include text that is entered into a chat session via the user interface 120. In some instances, the adaptive engine 408 launches a virtual chat session by opening a virtual chat or text messaging session on the user interface 120. Text entered by a user into a field or text box is transmitted by the user interface 120 to the adaptive engine 408 via the input interface 410. In some instances, the input interface 410 may include one or more application programming interfaces (“APIs”) that connect to the text messaging application, which enable routing of entered text to the adaptive engine 408. In addition to text, the input interface 408 may accept images, video, emojis, or selection indications of displayed options.

[0087] To receive the medical information 114 from the first memory 112a, the instructions 118 define a memory interface 412. The example memory interface 412 is configured to request or otherwise receive the medical information 114 from the first memory 112a. In some embodiments, the adaptive interface 408 is configured to use the memory interface 412 to request the medical information 114 after detecting that a patient has begun a virtual interactive session. In another embodiments, the adaptive interface 408 may search the first memory 112a for relevant medical information 114 (or request relevant medical information from the first memory 112a) using the memory interface 412 when the virtual assistant/chatbot logic 116 is configured to use certain medical information for answering a question, determining a recommended action, or otherwise predicting a medical complication.

[0088] As discussed above, the medical information 114 includes medical device data and patient data. The medical device data may be received from control logic 414 that operates one or more pumps 104, valves 106, and/or other components of the dialysis machine 102. Fig. 5 is a diagram of different possible components of medical device data 502, according to an example embodiment of the present disclosure. As shown in Fig. 5, the medical device data 502 includes treatment parameters 504 that be defined in one or more dialysis prescriptions or programs. The treatment parameters 504 define how the control logic 414 is to perform a treatment. The control logic 414 transforms the treatment parameters 504 into one or more signals or control instructions for the pumps 104, values 106, and other components.

[0089] For PD, the treatment parameters 504 may include one or more of a total fill volume, a number of cycles, a fill rate, a fill volume per cycle, a dwell time, a drain rate, an expected ultrafiltration (“UF”) removed per cycle or treatment, a dialysis fluid concentration (e.g., dextrose concentration), or a treatment schedule. For HD, the treatment parameters 504 may specify a treatment time, a blood circulation rate, a dialysis fluid circulation rate, a dialysis fluid volume, a treatment schedule, or a dialysis fluid concentration. For an infusion therapy, the treatment parameters 504 may include an infusion rate, a volume of fluid to be infused, a fluid type/volume, a drug or component concentration of the fluid, and a total infusion time. The treatment parameters 504 may be received from the server 128 and/or entered via the user interface 120. At least some of the treatment parameters 504 may be modified by a patient, a clinician, and/or the adaptive engine 408.

[0090] The medical device data 502 may also comprise medical result data 506 (e.g., treatment data), which describes or documents how a dialysis treatment was performed. As shown in Fig. 5, the medical result data 506 may include for a PD treatment at least one of dates/times that treatments were performed, a number of cycles per treatment, a fill volume, a drain volume, an estimated or measured amount of UF removed, and any events that occurred during a treatment. The events may include an alarm, an alert, a patient entry conflicting with a limit or threshold, a line occlusion, a line leak/disconnection, pausing of a treatment, etc. The medical result data 506 is determined by the control logic 414 based on feedback from the pumps 104, the valves, 106, and other dialysis components of the dialysis machine 102. The control logic 414 may also determine some of the medical result data 506 based on specified conditions for detecting a line occlusion by measuring line pressure using a pressure sensor or estimating UF removal based on differences between fill and drain volumes/weights. In some instances, the medical result data 506 may also include physiological data when the dialysis machine 102 is connected to or includes one or more sensors, such as a blood pressure sensor or cuff, a weight scale, a heart rate sensor, an ECG sensor, etc. Altogether, the medical result data 506 provides a summary of how a dialysis treatment was performed.

[0091] As shown in Fig. 5, the medical device data 502 may also include device information 508. The control logic 414 is configured to determine the device information 508 according to one or more specified conditions. For instance, the device information 508 may include a correct status of the dialysis machine 102, such as whether the machine is in a priming sequence, a cleaning/disinfection sequence, about to start a cycle of a PD treatment, progressing through a fill phase, progressing through a dwell phase, progressing through a drain phase, or finished with a cycle or treatment. The device information 508 may also include diagnostic information, such as faults detected in one or more pumps 104, valves 106, or other dialysis components.

[0092] Turning to Fig. 6, the medical information 114 stored in the first memory 112a may also include patient data 602. As mentioned above, the patient data 602 relates to information specific for patient that cannot be readily determined through monitoring of the dialysis machine 102. The patient data 602 may include patient activity information 604, patient demographic information 606, and patient medical information 608. The patient activity information 604 is determined through one or more question and answer sessions with a patient. Additionally or alternatively, the patient activity information 604 may be determined from a patient’s EMR, which is accessed by the adaptive engine 408 and/or the server 128 via a server interface 416.

[0093] In an example, a PD treatment may take three four hours to complete. During this time, a patient is fluidly connected to the dialysis machine 402 during at least dialysis fluid fill and drain phases of a cycle. During this time, it can get boring for a patient. To help will the time, the adaptive engine 408 may be configured to determine from the device information 408 that a fill phase will occur for the next 30 minutes. Also, the adaptive engine 408 is configured to determine from the speech recognition module 406 that the ambient environment is quiet. Based on these conditions, the adaptive engine 408 may be configured to begin a virtual session with a patient to fill the time to acquire some patient activity information 604, which may include diet information, fluid intake, medications, activity level, mental state, and sleep pattern. The adaptive engine 408 may prompt the patient with simple questions, such as “What have you eaten today?”, “Which medications have you taken?”, “What have you done today?”, and “How are you feeling?” and “How did you sleep?”. Patient responses are recorded by the processor 110 via the adaptive engine 408 as the patient activity information 604. The adaptive engine 408 may timestamp the patient activity information 604 to enable trends to be determined. In some instances, the adaptive engine 408 may use the virtual assistant/chatbot logic 116 for asking questions and determining if follow up questions are needed. In addition to asking questions during a treatment, the adaptive engine 408 may ask questions after a treatment or before a treatment at times that are not likely to interrupt a patient.

[0094] In addition to prompting the patient for patient activity information 604, the adaptive engine 408 may prompt the patient for the patient demographic information 606. Alternatively, the adaptive engine 408 may use the server interface 416 to obtain the patient demographic information 606 from one or more EMRs. As shown in Fig. 6, the patient demographic information 606 may include gender, age, weight, race, ethnicity, and geographic location. The adaptive engine 408 may use the virtual assistant/chatbot logic 116 to obtain the patient demographic information 606 from a patient before, during, or after treatments.

[0095] The adaptive engine 408 may also prompt a patient for the patient medical information 608 using the same virtual session. Alternatively, the adaptive engine 408 may use the dialysis machine 102 to obtain at least some of the patient medical information 608 using a connected physiological sensor. In yet other instances, the adaptive engine 408 uses the server interface 416 to obtain the patient medical information 608 from one or more EMRs. Altogether, the patient data 602 provides a summary of a health of a patient that may include at least some subject information that is obtained through a virtual session.

[0096] Returning to Fig. 4, the instructions 118 specify the server interface 416 for connecting to one or more APIs of the server 428. The server interface 416 includes a network address and/or ports of the server 428 to which certain medical information 114 is to be transmitted. In instances where at least some of the virtual assistant/chatbot logic 116 is located at the server 128, the server interface 416 provides for the transmission of virtual session initiation indications and patient responses to the server 128. Further, the server interface 416 is configured to receive questions, recommended actions, and other virtual session responses from the server 128, which are routed to the adaptive engine 408 for processing. As discussed above, the server interface 416 is also configured to enable one or more patient EMRs to be accessed by the adaptive engine 408 for obtaining medical information 114 of a patient. The instructions 118 and/or the virtual assistant/chatbot logic 116 may also be periodically updated by the server 128 via the server interface 416.

[0097] In some embodiments, the server interface 416 is configured to provide for the transmission of alerts, clinician messages, and requests for assistance. The server interface 416 may transmit alerts or clinician messages to the server 128, which are then routed to an appropriate clinician device 132 (e.g., a clinician device that is registered or subscribed to a patient of the dialysis machine 102). The alerts or messages may be indicative of an issue or medical complication that a patient is experiencing. The server interface 416 may also provide connectivity via the network 130 to an e911 service. Some alerts may be more medically severe, such as heart issues, loss of consciousness, or severe peritonitis, which may require immediate medical attention. When these types alerts are generated, the adaptive engine 408 is configured to transmit an alert or message to an e911 service. The alert or message may include an address of the patient, a name of the patient, and information related to the alert.

[0098] The instructions 118 of Fig. 4 also include an output interface 418 that is communicatively coupled to the adaptive engine 408. The output interface 418 is configured to convert data into text for display on the user interface 120 of the dialysis machine 102. The output interface 418 is also configured to convert data into speech, which is transmitted to the speaker 124 using an audio generator 420, which is configured to convert digital or analog signals into sound waves. The audio generator 420 may include one or more audio filters and/or amplifiers for producing audible speech.

[0099] The output interface 418 is also configured to open a chat program after receiving a message from the adaptive engine 408 indicative that a patient has selected to initiate a chat virtual session using the user interface 120. For the virtual chat, the output interface 418 causes a chat program to be displayed on the user interface, where questions and other text from the adaptive engine 408 are displayed. Patient responses are received via the input interface 410, as discussed above.

[00100] In some instances, the output interface 418 receives an indication from the adaptive engine 408 as to whether text is to be converted to audio or displayed as text. In some configurations, the virtual assistant/chatbot logic 116 provides a virtual session in which some information is conveyed audibly while other information is conveyed via the user interface 120. For example, the output interface 418 may cause a question to be transmitted as audio signals through the audio generator 420 while at the same time displaying at least some of the question via the user interface 120. In other instances, questions may be transmitted as audio while a virtual avatar or other graphic is displayed on the user interface 120 to simulate a more realistic interaction. In some instances, a question provided by the virtual assistant/chatbot logic 116 is specified to be asked through audio while possible answers are provided for selection via the user interface 120. As discussed below, many questions have defined possible answers in the virtual assistant/chatbot logic 116 such that the possible answers to a question can be identified and displayed by the output interface 418 when a question is being transmitted. The output interface 418 may also display determined recommended actions, clinician response messages, alarms/alerts, or indications of detected medical complications.

[00101] As discussed in more detail below, the virtual assistant/chatbot logic 116 determines the sequence of questions and graphics that are transmitted through the output interface 418. The adaptive engine 408 and the virtual assistant/chatbot logic 116 can be used to assist with issues of the dialysis machine 102 such as errors, warnings, setting up a treatment, or recurring alarms. The adaptive engine 408 and the virtual assistant/chatbot logic 116 can also be configured to predict patient complications such as peritonitis or catheter issues. Further, the adaptive engine 408 and the virtual assistant/chatbot logic 116 can be used for general assistance for ordering dialysis fluid bags, calling a clinic, booking a medical appointment, or calling an ambulance service. Further, the adaptive engine 408 and the virtual assistant/chatbot logic 116 may use a schedule in the medical information 114 to remind a patient about an upcoming or missed treatment or provide advice on their diet/lifestyle.

Issue Resolution Embodiment

[00102] As discussed above, the processor 110 executing the instructions 118 and the virtual assistant/chatbot logic 116 is configured to provide assistance for resolving issues related to the dialysis machine 102 or treatments performed by the dialysis machine 102. The virtual assistant/chatbot logic 116 may be configured to assist with issues related to at least one of setting up the dialysis machine 102 (including installing a cassette or tubing), performing software upgrades, setting up a treatment, performing priming of the cassette or tubing, handling warning issues, handling alarms, handling errors, ordering dialysis fluid containers or other disposables, sending a message to a clinician, ordering medical assistance, or scheduling a medical appointment. It should be appreciated that the types of issues that may be addressed by the virtual assistant/chatbot logic 116 is virtually endless and may be updated periodically based on population modeling of issues experienced by dialysis patients.

[00103] Fig. 7 illustrates a diagram of an example procedure 700 for resolving an issue related to the dialysis machine 102 or a treatment using the virtual assistant/chatbot logic 116, according to an example embodiment of the present disclosure. The example processor 110 is configured to execute or operate the machine-readable instructions 118, which are described by the procedure 700. Although the procedure 700 is described with reference to the flow diagram illustrated in Fig. 7, it should be appreciated that many other methods of performing the acts associated with the procedure 700 may be used. For example, the order of many of the blocks may be changed, certain blocks may be combined with other blocks, and many of the blocks described are optional. For example, block 706 may be omitted if only a text-based chatbot is used. [00104] To begin, the example processor 110 receives an indication 701 or determines that a patient wants to start a virtual session or chatbot (block 702). As discussed above, the indication 701 may include a speech indication that includes a keyword such as “Claria” or may be an indication selected on the user interface 120 of the dialysis machine 102. After receiving the indication 701, the processor 110 receives a patient response or query 703 (block 704). The response or query 703 may be a verbal question such as “How do I turn off this alarm!”. In some instances, the patient response or query 703 may be included with the indication 701. In other instances, the response or query 703 is provided via the user interface 120, which may provide a text entry box for a patient to type a question or provide a list of issue options for a patient to select.

[00105] If the response or query from the patient is provided via speech, the example processor 110 uses the instructions 118 to convert the speech to text (block 706). The processor 110 also uses the instructions 118 to access the medical information 114 at the first memory 112a and/or at the server 128. The processor 110 then uses the instructions 118 to compare the text provided by the patient and/or the medical information 114 to a data structure that contains the virtual assistant/chatbot logic 116 (block 710).

[00106] As described above, in this embodiment, the virtual assistant/chatbot logic 116 includes listing of possible issues and keywords associated with the issues. The processor 110 uses the instructions 118 to perform keyword matching to determine which issue a patient is likely referring. In the example, above, a patient indicates that help is needed with an alarm, but fails to identify the type of alarm (e.g., an occlusion alarm, a dialysis fluid leakage alarm, a low container alarm, a pumping alarm, a overfill alarm, etc.). In this instance, the virtual assistant/chatbot logic 116 specifies keywords associated with each alarm type. Instead of asking the patient more questions, the processor 110 uses the medical result data 506 of the medical device data 502 (e.g., medical information 114) to determine that an occlusion alarm is active. This additional piece of information enables the processor 110 to determine that the issue relates to an occlusion alarm.

[00107] Fig. 8 is a diagram illustrative of the virtual assistant/chatbot logic 116 for issue resolution, according to an example embodiment of the present disclosure. As shown, the virtual assistant/chatbot logic 116 includes a data structure of possible issues. Each issues has one or more keywords that are related to the issue. The keywords are preselected based on an analysis of medical information and patient responses from a populations of patients related to the issue. For example, an issue for an occlusion alarm includes keywords such as “alarm, alert, noise, flashing, beeping, and warning”. The keywords may also include diagnostic identifiers generated by the dialysis machine 102 associated with an occlusion detection. These keywords may include diagnostic trouble codes, field identifiers for occlusion detection, or event identifiers.

[00108] The processor 110 is configured to compare the received text and medical information 114 to each of the higher-order issues. The processor 110 selects an issue with a greatest matching score or probability based on the comparison with the keywords. The virtual assistant/chatbot logic 116 may prevent a selection if a match does not exceed a match threshold, such as 60% or 75%. In this instance, the virtual assistant/chatbot logic 116 includes follow up questions to ask the patient based on which issues have the greatest matching scores.

[00109] Fig. 8 also shows that there is a hierarchy of questions and answers for at least some of the possible issues. These additional questions and answers may further refine an issue to a more precise issue to provide a better resolution. For example, a higher-order issue may refer just to alarms. The lower level questions and answers provide keywords and criteria of different types of alarms. In some instances, the answers may be determined directly from the medical information 114 without further input from the patient, as in the occlusion alarm example discussed above. However, if any of the lower level answers cannot be determined, the processor 110 uses the listed questions associated with each possible answer of a higher-level issue for selecting which questions are provided to the patient for follow up. The hierarchy of the virtual assistant/chatbot logic 116 provides an issue/sub-issue transversal that causes the processor 110 to quickly converge upon a likely issues experienced by the patient.

[00110] At some point in the hierarchy, answers for a sub-issue are associated with a resolution or recommended action. The processor 110 determines when there is a match or likely match above a threshold with a resolution or recommended action. Based on this match, the processor 110 transmits the recommendation or resolution to the patient. In some instances, a resolution or recommendation may be provided for higher level issues without needing to progress through a hierarchy of sub-issues. The virtual assistant/chatbot logic 116 of Fig. 8 accordingly defines how the processor 110 is to interact with a patient to identify which issue is being experienced by a patient.

[00111] Returning to Fig. 7, the processor 110 determines uses the comparison of the keywords from the patient and the medical information 114 to determine if an issue can be identified (block 712). If an issue cannot be identified, the processor 110 uses the virtual assistant/chatbot logic 116 to determine which issues most closely match the keywords from the patient and the medical information 114. The processor 110 then determines additional questions (or prompts for additional information) specified by the virtual assistant/chatbot logic 116 for those semi-matching issues (block 714). The processor 110 then transmits one or more messages 715 with the additional questions to the patient (block 716). As discussed above, the processor 110 transmits the messages 715 as audio and/or as text displayed on the user interface 120 of the dialysis machine 102. In some instances, the messages 715 may include options that are selectable by a patient for converging upon an issue. The processor 110 then returns to block 704 for the next patient response.

[00112] Returning to block 712, if the issue is identified, the processor 110 determines from the virtual assistant/chatbot logic 116 whether a resolution can be identified (block 718). If the issue (or sub-issue) is associated with a resolution within the virtual assistant/chatbot logic 116, the processor 110 transmits a message 719 that includes information for resolving the issue (block 720). In some embodiments, the processor 110 is configured to resolve the issue by changing one or more treatment parameters 504 of the dialysis machine according to defined conditions. The message 719 may include a recommended action to perform for resolving the issues. For an occlusion alarm, this may include a recommendation to check tubing for crimps or blockages. For more serious issues, the processor 110 may also transmit the message 719 to the server 128 (for routing to the clinician device 132) or an emergency service.

[00113] The processor 110 may be configured to record feedback from a patient if the issue was resolved. If the issues was resolved, the processor 110 notes the success for later auditing and ends the example procedure 700. If the issue was not resolved, the processor 110 may use the virtual assistant/chatbot logic 116 to determine any follow up questions to identify another resolution. Alternatively, the processor 110 may request manual assistance from a help desk, clinician, or other personnel.

[00114] Returning to block 718, if a resolution is not identified, the processor 110 determines from the virtual assistant/chatbot logic 116 whether there are sub-issues or questions available (block 722). In other words, the processor 110 progresses down the hierarchy of the virtual assistant/chatbot logic 116 that is illustrated in Fig. 8. If there are further questions or prompts, the processor 110 determines additional questions (or prompts for additional information) specified by the virtual assistant/chatbot logic 116 for the next-lower questions/issues (block 714). The processor 110 then transmits one or more messages 715 with the additional questions to the patient (block 716). As discussed above, the processor 110 transmits the messages 715 as audio and/or as text displayed on the user interface 120 of the dialysis machine 102. The processor 110 then returns to block 704 for the next patient response.

[00115] However, if no further prompts or questions are available, the processor 110 is unable to determine a recommended for the patient. The processor 110 may transmit a message 723 for display on the user interface 120 indicative that a resolution cannot be determined (block 724). The processor 110 may also transmit the message 723 to a clinician or help desk to request manual assistance. The processor 110 may record feedback indicative that the issue could not be resolved to enable later auditing to determine updates for the virtual assistant/chatbot logic 116 for resolving the issue in the future.

Patient Medical Complication Embodiment

[00116] As discussed above, the processor 110 executing the instructions 118 and the virtual assistant/chatbot logic 116 is configured to predict whether a patient may experience a medical complication in the future or determine that a patient is currently experiencing a medical complication. In this embodiment, the virtual assistant/chatbot logic 116 may be configured to determine whether a patient has or is likely to get peritonitis, a catheter exit site infection, diabetes, or sepsis. The virtual assistant/chatbot logic 116 may also determine if a patient is having an issue with a catheter, having a cardiac issue, is experiencing a mental health issue, or is at risk for not adhering to their prescribed treatments. It should be appreciated that the types of medical complications that may be addressed by the virtual assistant/chatbot logic 116 is virtually endless and may be updated periodically based on population modeling of medical complications experienced by dialysis patients.

[00117] Fig. 9 illustrates a diagram of an example procedure 900 for predicting and/or diagnosing one or more medical complications in a patient that is using the dialysis machine 102, according to an example embodiment of the present disclosure. The example processor 110 is configured to execute or operate the machine-readable instructions 118, which are described by the procedure 900. Although the procedure 900 is described with reference to the flow diagram illustrated in Fig. 9, it should be appreciated that many other methods of performing the acts associated with the procedure 900 may be used. For example, the order of many of the blocks may be changed, certain blocks may be combined with other blocks, and many of the blocks described are optional. For example, block 906 may be omitted if only a text-based chatbot is used.

[00118] To begin, the example processor 110 receives an indication 901 or determines that a patient wants to start a virtual session or chatbot (block 902). As discussed above, the indication 901 may include a speech indication that includes a keyword such as “Claria” or may be an indication selected on the user interface 120 of the dialysis machine 102. After receiving the indication 901, the processor 110 receives a patient response or query 903 (block 904). The response or query 903 may be a verbal question such as “my stomach feels warm”. In some instances, the patient response or query 903 may be included with the indication 901. In other instances, the response or query 903 is provided via the user interface 120, which may provide a text entry box for a patient to type a question or provide a list of medical options for a patient to select.

[00119] If the response or query from the patient is provided via speech, the example processor 110 uses the instructions 118 to convert the speech to text (block 906). The processor 110 also uses the instructions 118 to access the medical information 114 at the first memory 112a and/or at the server 128. The processor 110 then uses the instructions 118 to determine a probability component for each possible medical complication (block 910).

[00120] In some embodiments, the virtual session illustrated in Fig. 9 may be initiated by the processor 110 rather than the patient. For example, the processor 110 may use the medical device data 502 to determine the dialysis machine 102 is in the middle of a fill cycle where there is some patient inactivity. The processor 110 may also initiate a virtual session before or after a dialysis treatment. To pass the time, the processor 110 initiates a virtual conversation by asking, for example, “how are you feeling?” The processor 110 uses the initial answers from the patient to progress through the virtual assistant/chatbot logic 116 discussed in connection with Figs. 10 and 11. In some instances, the processor 110 may only ask questions to determine the patient data 602 if such information is not known or stale before or while asking other questions defined within the virtual assistant/chatbot logic 116. For example, the processor 110 may ask a patient about their activities, which may be used to determine complications related to mental health, calorie bum, or general conditions of a patient’s body/mind.

[00121] Fig. 10 is a diagram of the virtual assistant/chatbot logic 116 for determining a medical complication of a patient, according to an example embodiment of the present disclosure. As shown in Fig. 10, the processor 110 compares the medical information 114 of the patient and the response or query 903 to calculate probability components 1002. Each probability component corresponds to a different medical complication for a different piece of information. The processor 110 determines a value for each probability component 1002 by comparing keywords assigned to that probability component to the medical information 114 of the patient and the response or query 903. If there is a match or near match, the probability is set to 100% for that component. If there is not a match, the probability is set to 0% for that component. If there is a partial match, the probability may be set to a value between 0% and 100% based on a degree of match.

[00122] The processor 110 uses the virtual assistant/chatbot logic 116 to calculate a probability component for each significant term or phrase in the patient’s response 903, as described in Fig. 11. In the above-example, the terms “stomach” and “warm” are compared individually to the probability components 1002 for matches. Further, the medical information 114 is compared separately, where most of the information may not contribute to a probability component. However, discrepancies in expected UF removed and drain volumes, alarms, physiological data, or indications of premature treatment ending via the patient may contribute to one or more probability components 1002. [00123] As shown in Fig. 10, the processor 110 combines probability components 1002 for the same medical condition 1004 to arrive at a probability of the patient having or predicted to have that medical complication. The virtual assistant/chatbot logic 116 may define weights or averages for combining the different probability components 1002. If a probability is not above a threshold, such as 40%, 50%, 60%, 75%, etc., the processor 110 determines that the patient is not experiencing a complication or will experience a complication. However, for each medical complication with a probability above the threshold, the virtual assistant/chatbot logic 116 causes the processor 110 to generate a medical complication notification 1006. It should be appreciated that more than one medical complication may be identified.

[00124] Fig. 10 also shows that the virtual assistant/chatbot logic 116 may provide for the evaluation of trends 1008. The probability components may be stored and compared overtime to see if a one or more probability components or probabilities of a medical complication have been increasing. A steady or sharp increase may cause the processor 110 to generate a medical complication notification 1006. The trends 1008 may be evaluated over distinct time periods, such as a three-day trend, a weekly-trend, a bi-weekly trend, a monthly trend, and a six- month trend.

[00125] Fig. 11 illustrates a hierarchy of the virtual assistant/chatbot logic 116, according to an example embodiment of the present disclosure. As shown in Fig. 11, the virtual assistant/chatbot logic 116 includes a hierarchy of questions, possible answers, and follow-up questions. Many of the questions have one or more possible answers. The virtual assistant/chatbot logic 116 causes the processor 110 to audibly or textually ask a question, which may be selected based on the initial patient query 903 or selected when a virtual session is automatically launched to elicit information from a patient during treatment lulls.

[00126] A patient response to a question is matched to the answers using the keyword matching described above in connection with Figs. 7 and 8. Specifically, a set of keywords may be assigned to each answer. Significant words and phrases from a patient and/or the medical information 114 is compared to the words or phrases of each possible answer for a given question. The processor 110 determines which answer mostly closely matches the patient response or medical information 114. Example possible answers to the question “how are you feeling” may include at least one of “I’m good”, “I’m not well”, “I’m feeling tired”, “I have pain”, “I feel cold”, “I have a fever, etc.

[00127] The selected answer may correspond to a sub-question having its own possible answers. The hierarchy continues until a patient answer corresponds to a probability component 1002 for a medical complication. The defined probability component may have a value of 0%, 100%, or a value therebetween for each medical complication based on a known relationship between the probability component and the different medical complications. For example, a fever may correspond highly to peritonitis and sepsis and less highly to catheter placement issues or a patient’s mental health. The processor 110 cycles through the virtual assistant/chatbot logic 116 for different sets of questions and corresponding hierarchies until one or more medical complications are identified or it is determined the patient has no risk of developing a medical complication.

[00128] Below are some example questions that are part of the virtual assistant/chatbot logic 116 that are used for predicting different medical complications.

• Example list of questions to predict peritonitis: o Leakage near catheter o Constipation o Catheter block o Catheter Kink o Diarrhea o Low urine output o Thirst o Inability to pass stool or gas o Fatigue o Fluid removal o Cloudy dialysis fluid o White flecks, strands or clumps (fibrin) in the dialysis fluid o History of peritonitis or infection or other diseases o Details of medication

• Example list of questions to predict sepsis: o Patches of discolored skin o Decreased urination o Changes in mental ability o Problems breathing o Abnormal heart functions o Chills due to fall in body temperature

• Example list of questions to predict mental illness: o Lack of interest o Language and voice pattern o Lack of activity o Missed therapy session

• Example list of questions to predict catheter exit site infection: o Swelling near exit site o Redness around exit site o Soreness when you touch the area o Drainage around exit site

• Example list of questions to predict a catheter problem: o Drain or fill duration o Fluid line pressure o Residual volume in the peritoneal cavity

[00129] The questions and answers of the virtual assistant/chatbot logic 116 shown in Fig. 11 may be determined from statistical data of a start and progression of the different medical complications from a population of patient data. The questions and answers may be periodically updated by the server 128 based on additional data from patient populations. In some embodiments, the server 128 is configured to use artificial intelligence and/or machine learning to determine the questions, answers, and/or the probability components for each of the complications. Further, in some embodiments, the virtual assistant/chatbot logic 116 may determine probabilities related to a chance of a patient experiencing the medical complication and separate probabilities that a patient already has developed the medical complication. The related complication information may convey to a patient or a clinician whether the patient is expected to have a medical complication in the future or likely currently has the medical complication.

[00130] Returning to Fig. 9, the processor 110 determines if the calculated medical complication probabilities exceed a threshold for identification of a medical complication (block 912). If at least one medical complication is above the threshold, the processor 110 is configured to generate and transmit a message 913 for display on the user interface 120 of the dialysis machine 102 that is indicative of the medical complication or a likelihood of developing a medical complication (block 914). The processor 110 may also transmit the message 913 to the server 128 for relaying to the clinician device 132. In some embodiments, if the probability of the medical complication is relatively high (e.g., over 90% for example), the processor 110 may generate an alert or provide a more serious indication of the complication for the patient and/or the clinician. In serious instances related to severe infection, heart trouble, etc., the processor 110 may generate message for a medical service provider. The processor 110 may store the message 913 indicative of the medical complication with the medical information 114 of the first memory 112a, which may be used for identifying issues identified by the patient in subsequent virtual sessions.

[00131] Returning to block 912, if the medical complication 912 cannot be readily determined from an initial patient response, the processor 110 determines if further prompts or questions are available (block 916). If not further prompts or questions are available in the hierarchy of the virtual assistant/chatbot logic 116, the processor 110 logs the patient condition as an exception. The processor 110 may associate the patient answers and the medical information 114 with the exception, which is transmitted to the server 128 for further analysis. The processor 110 also generates a message 917 indicated that the medical complication cannot be identified (block 918). The processor 110 may send the message 917 to the server 128 to trigger manual intervention from a clinician or other healthcare provider.

[00132] Returning to block 916, if there are further prompts or questions, the example processor 110 is configured to progress to the next question/prompt within the virtual assistant/chatbot logic 116 (block 920). The processor 110 then transmits a message 921 for display on the user interface 120 and/or for audible transmission via the speaker 124 that includes the next selected question (block 922). The processor 110 then returns to block 904 for the patient response. The processor 110 calculates new probability components for the medical complications based on the new patient response. The processor 110 may then combine or add the new probability components with the probability components from the previous patient response to calculate new medical complication probabilities, which are compared to the thresholds for a positive identification. The processor 110 continues to update the probability components through a progression of the hierarchy of questions shown in Fig. 11 for the virtual assistant/chatbot logic 116 until a lowest-level answer is reached. During this progression, the processor 110 may also analyze probability trends to one or more delta thresholds to determine if a positive medical complication identification can be made.

[00133] In some embodiments, the processor 110 prompts a patient to provide feedback regarding a medical question for which that may not know how to answer, or answer properly. In these embodiments, the virtual assistant/chatbot logic 116 is configured to provide a visual cue or graphic for display on the user interface 120 of the dialysis machine to help a patient answer a question. Fig. 12 shows an example graphic 1200 that may be displayed by the user interface 120 in conjunction with a question regarding possible medical complications, according to an example embodiment of the present disclosure.

[00134] In the illustrated example, the virtual assistant/chatbot logic 116 is progressing down a hierarchical tree regarding a peritonitis medical complication. One of the questions may ask a patient about a color of their effluent fluid. The degree of infection is associated with how yellow the effluent appears, with darker yellow fluid being indicative of a more severe infection (due to an accumulation of more white blood cells, which makes the effluent appear yellow). When asked, most patients simply respond that the effluent appears yellow or light yellow, which may be too vague for accurately determining a level of infection. Instead, the processor 110 displays the graphic 1200 with different color labels in conjunction with causing the question to be transmitted via audio. The patient is able to view the possible selections and convey the labeled term that best matches their effluent. In some embodiments, the user interface 120 is configured to enable a patient to touch which effluent graphic most closely matches their effluent to respond. The processor 110 accordingly provides sufficient examples or prompts to elicit more detailed information from a patient for making a medical complication diagnosis.

[00135] In other embodiments, the processor 110 may prompt a patient to use a camera on the dialysis machine 102 or a camera on their smartphone (which may be connected via Bluetooth® or a Wi-Fi connection to the dialysis machine 102) to record an image of their effluent. The recorded image is used by the processor 110 as part of the patient response. For instance, the processor 110 may perform image analysis on the recorded image to determine a color of the effluent. The determined color is provided as an answer to the question for subsequent medical complication diagnosis.

[00136] Fig. 13 is a diagram of the user interface 120 of the dialysis machine 102 displaying content 1300 related to the detection of a peritonitis medical complication, according to an example embodiment of the present disclosure. In the illustrated example, the content 1300 is transmitted via one or more messages 913 from the processor 110 after determining that the patient has a relatively high probability of developing peritonitis.

[00137] The content 1300 includes a warning message 1302 indicative of the medical complication. The content 1300 also includes an information section 1304 regarding the dialysis treatment and/or relevant medical information 114 associated with the patient. The content 1300 also includes an icon 1306 to begin another virtual session, perhaps to ask how peritonitis can be avoided. The content 1300 also includes icons 1308 that enable a patient to schedule an appointment with a clinician to treat their peritonitis or request immediate medical assistance.

Conclusion

[00138] It should be understood that various changes and modifications to the presently preferred embodiments described herein will be apparent to those skilled in the art. Such changes and modifications can be made without departing from the spirit and scope of the present subject matter and without diminishing its intended advantages. It is therefore intended that such changes and modifications be covered by the appended claims.