Systems and methods for recommending medical tests

Abstract

The device 10 for recommending medical tests for a patient has at least one electronic controller 20 programmed to retrieve patient data stored in a database 32, compare the patient data with guidelines 34 for a plurality of medical tests, identify at least one recommended medical test from the plurality of medical tests based on the comparison, and output on at least one display device a recommendation 40 that the patient undergo the at least one recommended medical test.

Description

[Technical Field]

[0001] FIELD OF THE INVENTION The present invention relates generally to the health care arts, patient screening arts, patient medical testing arts, patient medical test recommendation arts, and related arts. [Background technology]

[0002] There are a variety of screening tests that can be performed on patients, which can improve health by detecting serious diseases early or, conversely, rule out serious diseases and provide peace of mind. Screening tests are typically performed based on certain guidelines (e.g., colonoscopy is recommended for patients over the age of 50) and require various inputs from blood tests, histopathology, specialized medical tests (e.g., colonoscopy), etc. Physicians typically prescribe screening tests for their patients based on the guidelines and their specific knowledge of the patient's existing medical condition and their own experience. Summary of the Invention [Problem to be solved by the invention]

[0003] This approach has certain drawbacks. Physicians typically have a large caseload and may not be able to recognize and prescribe useful screening tests for their patients. If a screening test is prescribed, the patient may undergo redundant lab work to collect patient data related to the test. For example, liver screening requires certain blood tests, and even if the patient has already undergone the blood test, they may still be rescheduled for liver screening. Another aspect of this problem is that, although a patient may undergo a specific lab test, which may provide the majority of the patient data needed to perform a useful screening test, physicians may not be aware of this situation and may not prescribe the screening test even when only a small amount of additional patient data is required (or perhaps no additional data collection occurs at all). [Means for solving the problem]

[0004] Certain improvements to overcome these problems and others are disclosed below.

[0005] In one aspect, an apparatus for recommending medical tests for a patient has at least one electronic processor that is programmed to retrieve patient data stored in a database, compare the patient data to guidelines for a plurality of medical tests, identify at least one recommended medical test from the plurality of medical tests based on the comparison, and output on at least one display device a recommendation that the patient undergo the at least one recommended medical test.

[0006] In another aspect, a non-transitory computer-readable medium stores instructions executable by at least one electronic processor for performing a method for recommending medical tests for a patient, the method comprising: obtaining patient data stored in a database, comparing the patient data to guidelines for a plurality of medical tests, identifying a plurality of recommended medical tests from the plurality of medical tests based on the comparison, ranking the plurality of recommended medical tests based on one or more ranking criteria, and outputting, on at least one display device, a list of a plurality of recommendations for the patient to undergo the plurality of recommended medical tests in accordance with the one or more ranking criteria.

[0007] In another aspect, a method for recommending medical tests for a patient includes retrieving patient data stored in a database; comparing the patient data to guidelines for a plurality of medical tests; identifying at least one recommended medical test from the plurality of medical tests based on the comparison; generating a user interface that displays at least one of the recommendations for the at least one recommended medical test, currently available patient data stored in an electronic medical record database, and additional data required to complete the at least one recommended medical test; and transmitting the recommendation to one or more of the patient, the patient's physician, and the patient's insurance company.

[0008] One advantage is that it provides an overview for the physician to look at existing tests and diagnoses and determine potential follow-up procedures.

[0009] Another advantage is that the results of a patient's existing tests can be used for future tests on the same patient.

[0010] Another advantage is in extracting medical claims from existing test results and recommending additional tests to be performed.

[0011] Another advantage resides in maximizing the use of blood or tissue samples taken from patients.

[0012] A given embodiment may provide none of the aforementioned advantages, or may provide one, two, more, or all of the aforementioned advantages, and / or may provide other advantages, as will be apparent to one of ordinary skill in the art upon reading and understanding this disclosure.

[0013] The disclosure may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating preferred embodiments and are not to be construed as limiting the disclosure. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a diagrammatical illustration of an exemplary screening test recommendation device for recommending screening tests for a patient according to the present disclosure; [Figure 2] 2 illustrates an exemplary flowchart process performed by the device of FIG. 1. [Figure 3] FIG. 2 illustrates an example of a recommendation engine for the device of FIG. 1. DETAILED DESCRIPTION OF THE INVENTION

[0015] In some embodiments disclosed herein, a recommendation engine is provided for recommending screening tests, identifying available patient data for screening tests, and identifying required additional lab tests. The recommendation engine mines a patient's electronic medical (or health) record and compares various screening test guidelines with the patient's existing medical condition and available patient data. Based on this comparison, various screening tests are identified as potentially useful for the patient. For each potential screening test, the recommendation engine determines what patient data is already available for use in the screening test and what patient data still needs to be collected, along with the lab tests required to collect it. As patient data becomes less reliable over time, the age of the patient data can be taken into account, discarding older patient data (e.g., if it is too old to be considered reliable), and retained patient data can be annotated with its acquisition date.

[0016] The recommendation engine scores each potential screening test based on criteria such as the clinical guidelines for the screening test itself (e.g., this test is highly recommended for all patients at this patient's age) and based on any underlying medical conditions, medical history, or other patient characteristics (e.g., a liver test may be ranked higher if the patient is noted to be a relatively heavy alcohol user; or a screening test may be ranked higher if the patient is of an ethnicity where the condition being tested for is more prevalent). Another possible ranking factor is how many additional lab tests are required to perform the test—if a screening test can be performed with only one lab visit, such a test may be ranked higher. Another possible ranking factor may be related to the patient's health insurance coverage; for example, if the patient's health insurance fully reimburses the cost of performing a screening test, such a test may be ranked higher. If the patient has previously undergone a screening test, the scoring can take into account the time interval at which it is recommended to take the screening test again. For example, if a screening test is recommended to be repeated every 5 years and the patient has had a screening test within the past 5 years, the screening test will not be recommended, or the previous test will be taken into account when scoring (e.g., if the guidelines call for testing every 5 years and the last previous test was 4 years ago, the screening test may still be recommended, but with a lower score).

[0017] In other embodiments, if sufficient patient data is available to immediately perform a screening test, the screening test is performed and test results are generated by processing the already available patient data.

[0018] In yet another aspect, the recommendation system can be invoked in connection with scheduling an ordered hematology lab test, histopathology lab test, or other physician-ordered patient test that involves obtaining a blood or tissue sample from the patient. The recommendation system can then evaluate whether the extracted blood or tissue sample can be used for an additional screening test (in addition to the scheduled patient test), and if so, can recommend that the additional screening test be performed. In a variant approach, the recommendation system can further evaluate (calculate) the amount of blood or tissue sample needed to perform both the scheduled patient test and the additional screening test; and if the amount of blood or tissue extracted in response to the physician-ordered patient test is insufficient to also perform the additional screening test, the recommendation can also specify the amount of blood or tissue that needs to be extracted to perform both tests.

[0019] Based on the aforementioned information, the user interface (UI) presents potential screening tests to the physician and, optionally, to the patient and, optionally, to the patient's insurance company. The range of recommended screening tests and the level of detail provided for each recommended screening test can be tailored to the recipient. Typically, the patient's physician is provided with the most information possible, including the recommended screening test, the last date the screening test was administered to the patient (if applicable), already available patient data regarding the test annotated by the age of the data, any additional patient data needed to perform the screening test, and any lab tests required to collect it. The information provided to the physician may also include guideline recommendations regarding testing triggers (e.g., age thresholds, pre-existing conditions for which a screening test is recommended, etc.).

[0020] To meet regulatory requirements, etc., the recommendation system typically cannot order the screening tests that are actually performed. Rather, a physician may override the recommendation engine by rejecting a recommended screening test or indicating that the test should be retested within one year (or when the patient reaches age 50, or some other selected time interval). Similarly, a physician can reject the use of old patient data that was not automatically removed by the recommendation system. In some embodiments, the physician can select and approve screening tests at the time the recommendation system accesses the hospital's scheduling system to automatically schedule the tests.

[0021] In other embodiments, the results of any screening tests performed automatically by the recommender system by processing already available patient data are also presented to the physician and entered into the patient's EMR upon the physician's review and approval. Additionally, the recommender system dynamically updates its recommendations, so that, for example, if the recommender system detects that a recommended screening test has just been performed and its results have been added to the patient's EMR, the screening test recommendation is removed.

[0022] In contrast, the patient UI is expected to provide less information and less control over the list of recommended screening tests. Typically, the patient UI may list only recommended screening tests with scores above a threshold, perhaps supplemented with information about already available patient data and lab tests required to collect missing patient data. Patients generally cannot refuse tests in the patient UI (rather, they should discuss the matter with their physician, who can consult with the patient to refuse the test). Automatically administered screening test results are not presented to patients via the patient UI. Alternatively, if applicable regulations require patients to have the ability to refuse tests and / or request to receive test results, the patient UI may provide these capabilities. To ensure regulatory compliance, it is contemplated that different patient UI features will be provided in different regulatory jurisdictions. The insurance company's UI is likely similar to the patient's, but more information may be excluded to comply with HIPAA or other applicable patient privacy regulations, and less control is provided.

[0023] Typically, one set of M patient data is required to perform a given screening test. However, in practice, this set M is often not fixed. There may be a minimum set of patient data necessary to perform a screening test, but additional patient data beyond that minimum may provide greater confidence in the test results. In such cases, the physician may be presented with the minimum patient data required to perform the screening test and additional patient data that would increase confidence. If the test results are generated automatically by processing already available patient data, a confidence metric may also be provided, and if the test result is positive, obtaining additional patient data may be recommended to increase confidence.

[0024] In any other aspect disclosed herein, a recommendation engine can be incorporated into a laboratory test ordering system to maximize utilization of collected blood or tissue samples or to maximize data acquisition during an invasive procedure, such as a colonoscopy. For example, if a physician orders a complete blood count (CBC), at the time of entering this order, the recommendation system can determine whether additional blood tests can be performed on the vial (or vials) of blood obtained for the CBC to provide sufficient patient data to perform the recommended screening test. If so, the recommendation system can recommend performing the additional blood test using the blood collected for the already ordered CBC. Optionally, it can take into account the amount of blood (or tissue) obtained for the ordered laboratory test, determine whether additional blood (or tissue) is needed to perform the recommended additional laboratory test, and, if so, recommend collecting the additional blood or tissue.

[0025] 1 , an exemplary screening test recommendation apparatus 10 is implemented on an electronic processor 20, such as a server computer or exemplary multiple server computers 20 (e.g., a server cluster or farm, cloud computing resources, etc.), that implements the presently disclosed medical (e.g., screening) test recommendation method 100. To execute the recommendation method 100, the electronic processor 20 accesses at least one non-transitory storage medium 26 that stores a screening (or medical) test database 36 and at least one database 32 that stores patient medical records. Exemplary database(s) 32 are electronic medical records (EMRs) 32. However, other designations (e.g., electronic health record (EHR)) can be used, and / or the one or more databases 32 can include domain-specific patient record databases, such as a Picture Archiving and Communication System (PACS) database and / or a Radiology Information System (RIS) that stores patient data specific to medical images, or a Cardiovascular Information System (CIS or CVIS) that stores patient data collected and maintained by the patient's cardiologist and / or cardiology department. As discussed, the medical test recommendation method 100 maintains a list 38 of medical tests recommended for the patient. The medical test recommendation method 100 pushes this list 38, or a portion thereof, to the physician's user interface (UI) device 12. For example, the UI device 12 can be the physician's desktop computer, a tablet or notebook computer owned or assigned by the physician, a mobile phone owned or assigned by the physician, various combinations thereof, and / or others.

[0026] The medical test recommendation method 100 pushes a recommendation 40 that a patient undergo at least one of the recommended medical tests to various user interface (UI) devices 12, 13, 14. For example, in the illustrative example, the screening test recommendation 40 can be pushed to a physician's UI device 12 (e.g., a physician's desktop computer, a tablet or exemplary notebook computer 12 owned or transferred by the physician, a cell phone owned or transferred by the physician, various combinations thereof, and / or a patient's UI device 13, 14 (e.g., a patient's desktop computer, a tablet or exemplary notebook computer 13 owned or transferred by the patient, an exemplary cell phone 14 owned or transferred by the patient, various combinations thereof, and / or others), and / or an insurance company's UI (not shown). Additionally, the physician's UI device 12 can reject or decline the recommended test. can be used to send feedback 41, such as an instruction to postpone the recommended test for a period of time, back to the recommendation method 100. Note that FIG. 1 shows UI devices 12, 13, 14 for the patient and the patient's physician. More generally, however, the medical test recommendation method 100 preferably serves a large patient population (e.g., all patients treated by a physician or group of physicians, all patients insured by a particular health insurance provider, etc.). Thus, generally, there may be a corresponding UI device 12, 13, 14 for each patient that pushes recommendations 40 specific to that patient. Similarly, a patient-specific test recommendation list 38 is suitably maintained by the recommendation method 100 for each patient.

[0027] In some embodiments, medical test recommendation method 100 also interacts with a hematology, histopathology, or other lab scheduler 18 that schedules patient tests that involve drawing a blood sample from a patient or extracting a tissue sample from a patient. For example, lab scheduler 18 can be implemented on a server or workstation 19 that is used to schedule patients for physician-ordered complete blood counts (CBCs) or other physician-ordered patient tests that require blood or tissue samples. Recommendation method 100 receives information about the scheduled physician-ordered patient tests, optionally in real time, and determines whether any of the medical tests on list 38 of medical tests (excluding physician-ordered patient tests, if any) can be performed using the blood or tissue sample collected for the physician-ordered patient test. If so, recommendation 40 can include a recommendation to perform additional tests from list 38 that can be completed using the blood or tissue sample already scheduled to be collected to perform the physician-ordered test. Optionally, this recommendation may additionally or alternatively be sent to the scheduling workstation 19 and / or to the patient.

[0028] The various UI devices 12, 13, 14, 19 may include typical components such as an electronic processor (e.g., a microprocessor), at least one user input device (e.g., a mouse, keyboard, trackball, and / or the like) 22, and a display device 24 (e.g., an LCD display, a plasma display, a cathode ray tube display, and / or the like). The electronic processor 20 is implemented as a recommendation engine that executes the recommendation method 100 and is operatively connected to memory 26 embodied as one or more non-transitory storage media 26. The non-transitory storage media 26 may include, by way of non-limiting example, one or more of a magnetic disk, RAID, or other magnetic storage medium, a solid-state drive, a flash drive, an electronically erasable read-only memory, an optical disk or other optical storage, various combinations thereof, and may also be, for example, a network storage device accessible by the server computer 20, an internal hard drive (e.g., if the electronic processor 20 is implemented on a desktop PC rather than on the illustrative server 20), various combinations thereof, and the like. It should be understood that any reference herein to one or more non-transitory storage media or media 26 is intended to broadly encompass a single medium or multiple media of the same or different types. Similarly, electronic processor 20 may be embodied as a single electronic processor or as two or more electronic processors. Non-transitory storage media 26 store instructions executable by at least one electronic processor 20 to perform recommended method 100.

[0029] At least one electronic processor 20 is configured, as described above, to execute recommendation method or process 100 for recommending one or more medical tests to a patient. Non-transitory storage medium 26 stores instructions readable and executable by at least one electronic processor 20 to perform the disclosed operations, including executing method or process 100. In some examples, method 100 may be performed at least in part by cloud processing.

[0030] With reference to FIG. 2 and continuing reference to FIG. 1, an exemplary embodiment of method 100 is diagrammatically illustrated as a flowchart. In operation 102, at least one electronic processor 20 is programmed to retrieve patient data stored in database 32. Database 32, as previously described, may be any suitable database, such as, for example, an electronic medical record database, an electronic health record database, an image archiving and communication system database, a radiology information system, etc. In some embodiments, a physician may determine the acquisition date of an item of patient data and remove any item of patient data with acquired data older than a predetermined time threshold. For example, if a patient is recommended to undergo a colonoscopy every three years, all patient items containing colonoscopies acquired more than three years ago may be discarded. In another approach, the predetermined time threshold may be a configuration parameter of recommendation method 100, and a physician may optionally override this (default) time threshold. Note that the predetermined time threshold may be different for different types of patient data, as different types of patient data may be considered “out of date” at different ages and some patient data (e.g., patient ethnicity) may never change.

[0031] In operation 104, the at least one electronic processor 20 is programmed to compare the patient data to a plurality of medical test guidelines 34 stored in a screening or medical test database 36 and identify at least one recommended medical test based on the comparison. The guidelines 34 and other data from the plurality of medical test database 36 may be stored on the non-transitory computer-readable medium 26. For example, the guidelines 34 for various screening tests are compared to the patient's existing medical condition and available patient data retrieved from the database 32. Based on this comparison, at least one recommended screening test is identified as potentially useful for the patient.

[0032] In operation 106, at least one electronic processor 20 is programmed to transmit the recommendations 40 to one or more UI devices 12, 13, 14 for display on the display device 24 an output of the recommendations 40 that the patient undergo at least one recommended medical test 38. The computational and control processes involved in displaying the recommendations 40 can be distributed in various ways between the server 20 and the UI devices 12, 13, 14. For example, the workstation UI devices 12, 13 may be running a web browser that connects to a web page hosted by the server 20, the browser running on the UI devices 12, 13 receiving the web page content including the recommendations 40 and displaying the web page content on the display device 24. In the case of a mobile device, such as the exemplary mobile phone 14, the mobile device can execute an application program (“app”) that receives the recommendations 40 and locally constructs a presentation of the recommendations to be shown on the display device 24. These are merely illustrative examples.

[0033] In some embodiments, the comparison process 104 includes determining missing information that is missing in the patient data and that, if available, would enable performance of at least one recommended medical test. In this embodiment, the output process 106 includes outputting an identification 42 of the missing information that, if available, would enable performance of at least one recommended medical test.

[0034] In another embodiment, the comparison process 104 includes identifying a viable medical test from among multiple medical tests in the screening test database 36 based on the comparison, where the available information for the comparison is sufficient to perform the medical test using already available patient data. For example, if the patient recently had blood test results collected for one medical test on the list 38 that can be used for the screening test, the screening test on the list is performed using the available blood test results, generating a screening test result 43. The output process 106 to UI devices 12, 13, 14 then includes displaying the test result 43 on the display device 24. This approach is suitable when the screening test can be performed algorithmically based on the patient data, without relying on subjective analysis or interpretation of the patient data. For example, a standard body mass index (BMI) screening test for obesity can be performed algorithmically using only the patient's height and weight as input. Thus, BMI screening can be performed automatically upon receiving the patient's recent (i.e., not outdated) height and weight measurements in the patient's records. In contrast, screening tests such as mammography screening, which require expert interpretation of medical images by a clinician, typically cannot be performed algorithmically (although in some cases, such screening tests can be performed algorithmically using automated image analysis by artificial intelligence algorithms). Algorithms for performing screening tests are suitably stored in screening test database 36 for tests that can be performed algorithmically.

[0035] Optionally, the at least one electronic processor 20 is programmed to generate a confidence value for the test result 43 based at least on information available in the patient data. From this, the at least one electronic processor 20 is programmed to determine missing supporting information from the patient data that, if available, would increase the confidence value. The output process 106 then includes recommendations 46 for obtaining supporting information to increase the confidence value. This aspect can be implemented in other ways, some of which do not require calculating a numerical confidence value. For example, the screening test database 36 can store, for a given screening test, a list of patient data for performing the test and a list of supporting patient data that, if available, would increase confidence in the result. In this case, the output process 106 includes recommendations 46 for obtaining the listed supporting information to increase confidence in the test result.

[0036] In optional operation 108, if multiple tests are recommended based on comparison operation 104, at least one electronic processor 20 is programmed to rank the multiple recommended medical tests 38 based on one or more ranking criteria. The one or more ranking criteria may include, for example, (i) one or more clinical guidelines 34 for each of the recommended medical tests 38 (e.g., whether the patient's age or gender dictates that a particular test should be prioritized over others), (ii) patient history or underlying conditions or characteristics identified from the retrieved patient data (e.g., whether the patient or the patient's family has a history of, for example, diabetes or heart disease), (iii) additional lab tests (e.g., blood tests, tissue samples, procedures such as colonoscopy) required to perform each of the multiple recommended medical tests based on the retrieved patient data, (iv) insurance reimbursement levels for each test based on the retrieved patient data (e.g., how much an insurance company will cover the cost of the test), and (v) patients who have already undergone a test based on the retrieved patient data, or other ranking criteria. These are merely exemplary and are not intended to limit the types of potential ranking criteria. From the ranking criteria, the at least one electronic processor 20 is programmed to generate a corresponding score for each criterion for each recommended test 38. The ranked list of medical tests 38 based on the scores can be displayed on the display device 24, for example, in descending order of the score of each test. Alternatively, only the top N ranked list can be displayed (i.e., only the top N ranked tests are displayed).

[0037] Optionally, the at least one electronic processor 20 is programmed to generate recommendations 40 with information about the recommended medical tests 38, including, for example, the recommended medical tests, the scope of each recommended medical test, the level of detail for the at least one recommended medical test, currently available patient data stored in the database 32, and additional data 42 required to complete the at least one recommended medical test. These are merely illustrative examples.

[0038] The recommendations 40 can be transmitted to a physician-operable user device 12 (e.g., a workstation, laptop, mobile device such as a smartphone or tablet, etc.). In some embodiments, the physician can provide user input 41 (e.g., via a mouse click, keystroke, finger swipe, etc.) indicating a rejection of one or more of the recommended medical tests 38. For example, the physician can reject a recommended medical test 38 (e.g., a diabetes screening) if the physician knows the patient has recently undergone such screening. In another example, the physician can reject a recommended medical test 38 if the patient knows the test is not suitable for the patient (e.g., a test for a medical condition that will manifest over a long-term period recommended for a patient with terminal cancer). The at least one electronic processor 20 is programmed to receive the input 41 from the physician indicating a rejection of at least one of the recommended tests 38 and to update the list of medical recommended screening tests 38 to remove the rejected test. The tests in the updated ranked list 38 can be automatically updated on the physician's user device, or an updated UI can be sent to the user device.

[0039] The recommendations 40 can also be pushed to another party's UI device, such as the patient's UI device 13, 14 or the patient's insurance company's UI device (not shown). The information sent to the patient and / or insurance company includes less information than the UI sent to the physician. The information sent to the patient can list the recommended medical tests that score above a threshold and, in some cases, can list information about already available patient data or lab tests needed to collect missing patient data. The patient generally cannot refuse the recommended tests. Rather, the presented UI preferably suggests that the patient discuss the recommended tests with the patient's physician. The insurance company's UI is likely to be similar to the patient's UI, except that more information may be excluded to comply with HIPAA (in the United States) or other applicable patient privacy regulations.

[0040] In another optional aspect, the device 10 can include or interact with a test procedure scheduling system 18, including (for example) a workstation 19, as described above. In this example, the at least one electronic processor 20 is programmed to schedule a physician-ordered test procedure that requires the collection of a blood or tissue sample from the patient. The at least one electronic processor 20 is then programmed to determine missing information in the patient data (retrieved from the database 32) that, if available, would enable the performance of at least one recommended medical test (e.g., generated by the comparison process 104). From there, the at least one electronic processor 20 is programmed to determine whether an unscheduled hematology or histopathology test can be performed on the blood or tissue sample already required to perform the physician-ordered test to obtain at least some of the missing information. Finally, the output process 106 of a recommendation 40 that the patient undergo at least one recommended medical test 38 includes outputting the recommendation to perform the unscheduled hematology or histopathology test on the display device 24.

[0041] FIG. 3 shows a visualization of the processing of the at least one electronic processor 20. As shown in FIG. 3 and described above, the at least one electronic processor 20 is programmed as a recommendation engine 52 to output a notification 40 for one or more recommended medical tests (or diagnoses) 38. The recommendation engine 52 receives as input information regarding age-dependent medical tests 54 (e.g., information from the medical test guidelines 34) and information 56 derived from patient data retrieved from the database 32. From the inputs 54 and 56, the recommendation engine 52 is programmed to execute a matching algorithm to determine any missing information needed to complete the recommended tests and to generate an output 58 including a recommendation 40 regarding the medical test or diagnoses 38. The recommendation engine 52 is also programmed to generate a medical claim output 60 from existing medical tests without requiring additional information.

[0042] The matching algorithm involves using a set of recommended age-related medical tests or diagnoses M (including a subset Mi of medical tests and / or diagnoses required to generate a medical claim) together with a set R of existing medical test / diagnoses for each patient. For each recommended age-related medical test / diagnosis Mi, completing Mi with respect to R according to Ci = Mi\R outputs the set of additional medical tests / diagnoses required to complete the information required for Mi. In this way, only missing medical tests / diagnoses need to be performed, thereby reducing effort and cost. Furthermore, by computing Ci for all Mi, a set Mi with an empty Ci provides medical claims that can already be generated from those Mi, i.e., existing medical tests / diagnoses, without additional cost and effort.

[0043] Another example of method 100 is described below. For example, a patient has reached an age where a specific cancer screening is recommended by medical testing 36 guidelines 34. Cancer screening requires that background information about previous family cases of the disease be correlated with some information obtained from a blood sample. The patient had a previous blood sample taken for another medical test two months prior. Data from that previous test is available to medical test recommendation engine 52. Engine 52, connected to a software application ("app") on the patient's user device (e.g., a smartphone), can now suggest that additional information needed to complete the screening test be obtained either directly through the app or via the patient's physician at their next scheduled appointment.

[0044] The present disclosure has been described with reference to preferred embodiments. Modifications and alterations may occur to those skilled in the art upon reading and understanding the foregoing detailed description. It is intended that the exemplary embodiments be construed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims

1. 1. A device for recommending medical tests to a patient, comprising: retrieving patient data stored in a database; comparing the patient data to guidelines for a plurality of medical tests and identifying at least one recommended medical test from the plurality of medical tests based on the comparison; outputting, on at least one display device, a recommendation that the patient undergo the at least one recommended medical test; at least one electronic processor programmed to execute The at least one electronic processor further comprises: determining missing information that is missing in the patient data, the missing information enabling performance of the at least one recommended medical test if available; and wherein the step of outputting a recommendation that the patient undergo the at least one recommended medical test includes outputting, on the display device, an identification of the missing information, if available, that enables performance of the at least one recommended medical test. Device.

2. The at least one electronic processor: comparing the patient data to the guidelines for the plurality of medical tests; identifying the viable medical test from the plurality of medical tests based on a comparison where available information is sufficient to perform the viable medical test; performing the available medical test using the available information to generate test results; outputting the test results on the at least one display device; 10. The apparatus of claim 1, programmed to:

3. 3. The device of claim 2, wherein the at least one electronic processor is programmed to perform a step of determining supporting information that is missing in the patient data and that, if available, would increase confidence in the test result, and the output of the test result includes a recommendation to obtain the supporting information.

4. 4. The apparatus of claim 1, wherein the at least one recommended medical test comprises a plurality of recommended medical tests, and wherein the at least one electronic processor is programmed to perform the step of ranking the plurality of recommended medical tests based on one or more ranking criteria.

5. The at least one electronic processor: identifying a clinical guideline ranking criterion for each of the recommended medical tests; generating a score for each recommended medical test based on clinical guidelines for each of the medical tests; 5. The apparatus of claim 4, wherein the apparatus is programmed to perform the ranking of the plurality of recommended medical tests by a process comprising:

6. The at least one electronic processor: identifying ranking criteria for patient conditions or characteristics identified from the retrieved patient data; generating a score for each of a plurality of recommended medical tests based on the patient's medical history or characteristics; 5. The apparatus of claim 4, wherein the apparatus is programmed to perform the ranking of the plurality of recommended medical tests by a process comprising:

7. The at least one electronic processor: identifying, based on the retrieved patient data, ranking criteria for additional laboratory tests required to perform each of the plurality of recommended medical tests; generating a score for each of the recommended plurality of medical tests based on additional laboratory tests required for each medical test; 5. The apparatus of claim 4, wherein the apparatus is programmed to perform the ranking of the plurality of recommended medical tests by a process comprising:

8. The at least one electronic processor: identifying ranking criteria for reimbursement levels for each medical test based on the retrieved patient data; generating a score for the recommended plurality of medical tests based on a reimbursement level for each medical test; 5. The apparatus of claim 4, wherein the apparatus is programmed to perform the ranking of the recommended medical tests by a process comprising:

9. The at least one electronic processor: identifying ranking criteria for patients who have undergone one of the medical tests based on the retrieved patient data; generating a score for the plurality of recommended medical tests based on the patient having undergone any of the medical tests; 5. The apparatus of claim 4, wherein the apparatus is programmed to perform the ranking of the plurality of recommended medical tests by a process comprising:

10. The outputting step includes: generating a user interface showing at least one of the recommendation for the at least one recommended medical test, the coverage of the at least one recommended medical test, the level of detail of the at least one recommended medical test, currently available patient data stored in the database, and additional data required to complete the at least one recommended medical test; sending at least said recommendation to a physician; 10. The device according to claim 1, further comprising:

11. the at least one electronic processor: receiving an input from a physician indicating a rejection of at least one of the recommended medical tests presented on the user interface; updating the user interface to remove the rejected medical test; 11. The apparatus of claim 10, programmed to:

12. The outputting step includes: generating a user interface that displays at least one of the recommendations for the at least one recommended medical test, currently available patient data stored in an electronic medical record database, and additional data required to complete the at least one recommended medical test; transmitting at least said recommendation to the patient or the patient's insurance company; 10. The device according to claim 1, further comprising:

Images