Display control device and computer program
The display control device and computer program facilitate transparent evaluation and comparison of multiple inference models, addressing the lack of transparency and consistency in machine learning-based clinical judgments by integrating model evaluation and historical data display.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- NIHON KOHDEN CORP
- Filing Date
- 2022-11-18
- Publication Date
- 2026-06-30
AI Technical Summary
Existing inference models generated through machine learning often lack transparency and consistency, making it difficult for users to validate their results and leading to unsatisfactory usability when multiple models are used for clinical judgments.
A display control device and computer program that integrates multiple inference models, displaying their results with identifiers and allowing users to evaluate and compare them, along with the ability to input evaluation information and display historical data for enhanced decision-making.
Enhances usability by enabling users to validate and compare inference results across models, improving confidence in clinical judgments through transparent evaluation and historical analysis.
Smart Images
Figure 0007882761000001 
Figure 0007882761000002 
Figure 0007882761000003
Abstract
Description
Technical Field
[0001] The present disclosure relates to a device for controlling information displayed on a display device. The present disclosure also relates to a computer program executable by a processor mounted on the device.
Background Art
[0002] In the medical clinical field, an inference model generated using machine learning technology to substitute or assist the clinical judgment of medical staff is known. For example, Patent Document 1 discloses a model that infers whether a specific type of arrhythmia has occurred in a subject based on whether a predetermined type of peak exists in the electrocardiogram waveform data acquired from the subject.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] Improve the usability when there are multiple inference models for providing a specific clinical judgment.
Means for Solving the Problems
[0005] One aspect example provided by the present disclosure is a display control device, an interface that receives subject information regarding a subject, a processor that controls information displayed on a display device, and is provided with, wherein the processor inputs the subject information into a first inference model to obtain a first inference result corresponding to a clinical judgment for the subject obtained for a predetermined item, The subject information is input into the second inference model to obtain a second inference result corresponding to the clinical judgment obtained for the subject for the predetermined items. The first inference result is displayed on the display device along with a first identifier that identifies the first inference model. The second inference result is displayed on the display device along with a second identifier that identifies the second inference model.
[0006] One example of an embodiment provided herein is a computer program executable by a processor mounted on a display control device, By being executed, the display control device will We accept information about the target individuals. The subject information is input into the first inference model to obtain a first inference result corresponding to the clinical judgment obtained for the subject for a predetermined item. The aforementioned biological information is input into a second inference model to obtain a second inference result corresponding to the clinical judgment obtained for the subject for the predetermined items. The first inference result is displayed on the display device along with a first identifier that identifies the first inference model. The second inference result is displayed on the display device along with a second identifier that identifies the second inference model.
[0007] In recent years, various inference models that output clinical judgments based on input biometric information have been put into practical use, but the inference results are not always satisfactory to the user. In particular, when such inference models are generated through machine learning, it is not always possible to explicitly describe the basis for the inference results, and it may be difficult to gain confidence in the validity of the inference results.
[0008] According to the configurations described in each of the above examples, it is possible to provide new methods for consideration and analysis for users to make clinical judgments themselves, such as evaluating the validity of the inference results from each inference model by comparing different clinical judgments for the same item output from multiple inference models, or evaluating in parallel the inference results of multiple inference models that provide clinical judgments for multiple items that can be referenced for the same purpose. Therefore, it is possible to improve usability when multiple inference models exist to provide a specific clinical judgment. [Brief explanation of the drawing]
[0009] [Figure 1] This illustrates the functional configuration of a display control device according to one embodiment. [Figure 2] Figure 1 shows an example of the screen displayed on the display device. [Figure 3] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 4] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 5] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 6] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 7] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 8] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 9] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 10] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 11] Another example of the screen displayed on the display device shown in Figure 1 is shown. [Figure 12] Another example of the detailed information displayed on the display device in Figure 1 is shown. [Figure 13] Another example of the detailed information displayed on the display device in Figure 1 is shown. [Figure 14]Another example of the detailed information displayed on the display device of FIG. 1 is shown. [Figure 15] The configuration of a model evaluation system according to an embodiment is illustrated.
Mode for Carrying Out the Invention
[0010] Examples of embodiments will be described in detail below while referring to the accompanying drawings.
[0011] As used herein, the expression "at least one of A and B" for two entities A and B means including the case where only A is specified, the case where only B is specified, and the case where both A and B are specified. Each entity of A and B may be singular or plural unless otherwise specified.
[0012] As used herein, the expression "at least one of A, B, and C" for three entities A, B, and C means including the case where only A is specified, the case where only B is specified, the case where only C is specified, the case where A and B are specified, the case where B and C are specified, the case where A and C are specified, and the case where all of A, B, and C are specified. Each entity of A, B, and C may be singular or plural unless otherwise specified. The same applies when there are four or more entities to be described.
[0013] FIG. 1 illustrates the functional configuration of a display control device 10 according to an embodiment. The display control device 10 is a device for controlling information displayed on a display device 20.
[0014] The display control device 10 includes an input interface 11. The input interface 11 is configured as a hardware interface that receives subject information SB of a subject 30. The subject information SB may be in the form of digital data or analog data according to the specifications of sensors and measuring devices.
[0015] If the subject information SB is in the form of analog data, the input interface 11 includes an appropriate conversion circuit, including an A / D converter. This description also applies to other signals and data that the input interface 11 can accept, as described later.
[0016] As used herein, the term "subject information" includes not only information relating to the subject's biological parameters, but also information relating to the subject's attributes, health status, and settings of devices connected to the subject. Examples of information relating to biological parameters include general vital signs (respiratory rate, blood pressure, heart rate, pulse rate, etc.) and test results (white blood cell count, serum creatinine level, urine specific gravity, etc.). Examples of information relating to attributes include height, weight, age, and sex. Examples of information relating to health status include medical history, underlying diseases, complications, surgical history, medication history, and length of hospitalization. Examples of information relating to settings include ventilator flow rate and whether or not a pacemaker is implanted. "Subject information" may be obtained from the subject through various sensors and measuring devices, or it may be entered by healthcare professionals or the subject themselves through various input devices.
[0017] The display control device 10 includes a processor 12. The processor 12 is configured to input subject information SB to the first inference model 41 and the second inference model 42.
[0018] The first inference model 41 is an inference algorithm configured to output a first inference result corresponding to a clinical judgment obtained for a predetermined item when subject information SB obtained from subject 30 is input. The first inference result may represent the clinical judgment deterministically or probabilistically. In particular, in the latter case, the inference algorithm may be generated through various machine learning methods.
[0019] The second inference model 42 is an inference algorithm configured to output a second inference result corresponding to a clinical judgment obtained based on a different method than the first inference model 41 for the predetermined items, when subject information SB obtained from the subject 30 is input. The second inference result may represent the clinical judgment deterministically or probabilistically. In particular, in the latter case, the inference algorithm may be generated through various machine learning processes.
[0020] As used herein, the phrase "clinical judgments obtained based on different methods" includes the following: • Using algorithms provided by different vendors resulted in differing clinical judgments for the same item. • Using different versions of algorithms provided by the same vendor may result in different clinical judgments for the same item. • By using algorithms generated through different machine learning methods, different clinical judgments obtained for the same item can be obtained. • By using algorithms generated through machine learning with differing training data (including cases where the assigners of training labels differ), different clinical judgments obtained for the same item can be analyzed. • By using an algorithm that requires different inputs, different clinical judgments are obtained for the same item. • Different clinical judgments obtained regarding differing items
[0021] The display control device 10 is equipped with an output interface 13. The processor 12 is configured as a hardware interface that outputs a display control signal DC to the display device 20. The display control signal DC is configured to display the first inference result on the display device 20 along with a first identifier that identifies the first inference model 41, and to display the second inference result on the display device 20 along with a second identifier that identifies the second inference model 42.
[0022] The output interface 13 allows the display control signal DC to be either a digital or analog signal, depending on the specifications of the display device 20. If the display control signal DC is an analog signal, the output interface 13 includes an appropriate conversion circuit, including a D / A converter. This description also applies to other signals and data that the output interface 13 can output, as described later.
[0023] Figure 2 shows an example of a model selection screen 21 displayed on the display device 20 based on the display control signal DC.
[0024] The model selection screen 21 includes the first identifier 211. The first identifier 211 identifies "Infusion Model 1". "Infusion Model 1" is an example of the first inference model 41. "Infusion Model 1" is an inference algorithm that outputs the probability that subject 30 is dehydrated when heart rate and blood pressure are input as subject information SB for subject 30. "Dehydrated state" is an example of a predetermined item, and "probability of being dehydrated" is an example of a clinical judgment. On the model selection screen 21, this probability is displayed together with the first identifier 211.
[0025] The model selection screen 21 includes a second identifier 212. The second identifier 212 identifies "Infusion Model 2". "Infusion Model 2" is an example of the second inference model 42. "Infusion Model 2" is an inference algorithm that outputs the probability that subject 30 is dehydrated when subject information SB of subject 30, such as heart rate, blood pressure, and infusion volume, is input. On the model selection screen 21, this probability is displayed along with the second identifier 212.
[0026] In other words, the parameters input as subject information SB may differ between the first inference model 41 and the second inference model 42.
[0027] As an alternative example, "Lactate Value Model 1" can be adopted as the first inference model 41 and "Lactate Value Model 2" as the second inference model 42. Both "Lactate Value Model 1" and "Lactate Value Model 2" are input as subject information SB for the subject 30, including height, weight, transcutaneous arterial oxygen saturation (SpO2), end-tidal carbon dioxide concentration (EtCO2), hemoglobin level (Hb), heart rate, cardiac output (CO), respiratory rate, and peripheral body temperature.
[0028] "Lactate Level Model 1" is an inference algorithm that outputs the blood lactate level [mmol / L] of subject 30 for the given input. "Blood lactate level" is an example of a predetermined item, as well as an example of a clinical judgment. "Lactate Level Model 2" is an inference algorithm that infers whether the blood lactate level of subject 30 exceeds 2 [mmol / L] for the given input, and outputs either "high" or "low". "Whether the blood lactate level is high or low" is an example of a clinical judgment. In other words, the manner in which the clinical judgment for a predetermined item is shown may differ between the first inference model 41 and the second inference model 42.
[0029] The number of inference models into which the processor 12 inputs subject information SB may be three or more. In this case, three or more identifiers, each identifying one of the three or more inference models, are displayed on the model selection screen 21 of the display device 20.
[0030] Figure 3 shows another example of the model selection screen 21 displayed on the display device 20 based on the display control signal DC. In this example, multiple inference models that provide different clinical judgments are displayed in order to comprehensively assess the circulatory dynamics of the subject 30. That is, multiple inference models that provide clinical judgments on multiple items that can be referenced for a common purpose can be displayed on the display device 20.
[0031] As illustrated in the figure, the way in which clinical judgments are indicated may be not only by numbers and letters, but also by shapes, symbols, colors, etc.
[0032] In recent years, various inference models that output clinical judgments based on input biometric information have been put into practical use, but the inference results are not always satisfactory to the user. In particular, when such inference models are generated through machine learning, it is not always possible to explicitly describe the basis for the inference results, and it may be difficult to gain confidence in the validity of the inference results.
[0033] With the configuration described above, it is possible to evaluate the validity of the inference results from each inference model by comparing differing clinical judgments on the same item output from multiple inference models, or to evaluate in parallel the inference results of multiple inference models that provide clinical judgments on multiple items that can be referenced for the same purpose. This provides new methods for consideration and analysis that enable users to make their own clinical judgments. Therefore, it is possible to improve usability when multiple inference models exist to provide a specific clinical judgment.
[0034] As illustrated in Figure 2, the model selection screen 21 includes a first evaluation GUI 213 and a second evaluation GUI 214. The first evaluation GUI 213 is a GUI for inputting an evaluation of the inference results for "Infusion Model 1". The second evaluation GUI 214 is a GUI for inputting an evaluation of the inference results for "Infusion Model 2". The smiling image is configured to trigger a click or tap operation when the inference result is given a high rating. The worried face image is configured to trigger a click or tap operation when the inference result is given a low rating.
[0035] The first evaluation GUI 213 and the second evaluation GUI 214 can be examples of the user interface 50 illustrated in Figure 1. That is, the input interface 11 of the display control device 10 may be configured to receive evaluation information EV corresponding to the evaluation of at least one of the inference results from the first inference model 41 and the inference results from the second inference model 42.
[0036] The user interface 50 for inputting evaluation information EV may take any appropriate form, such as voice input, eye-tracking input, or gesture input, in addition to or instead of the cursor operation or touch operation described above.
[0037] The processor 12 is configured to store evaluation information EV in the storage 60. The storage 60 is a storage device that can be implemented using semiconductor memory, a hard disk drive, a magnetic tape drive, etc. The evaluation information EV may be sent directly from the processor 12 to the storage 60, or it may be sent via the output interface 13.
[0038] With this configuration, it is possible to store information evaluating the validity of the inference results output by each inference model. Furthermore, since evaluation can be performed by referencing multiple inference models displayed on the display device 20, it is possible to add a new metric: not only an absolute evaluation of each inference model, but also a relative evaluation among multiple inference models.
[0039] Furthermore, the evaluation information EV entered through the user interface 50 may include attribute information of the user who performed the evaluation. Attribute information may include at least one of the following: name, gender, age, medical department, affiliated organization, job title, position, assigned ward, assigned patients, length of employment, length of hospital stay, professional skills, and acquired qualifications.
[0040] The evaluation information EV can be output from the output interface 13 in the form of a report from an output device. Examples of output devices include printers, speakers, displays, and data transmission devices.
[0041] Attribute information may be entered via text input, biometric authentication, or by optically or magnetically reading attribute information stored on a user-portable card or tag, or by reading it via proximity wireless communication.
[0042] For example, even among individuals with the same job responsibilities, evaluations of the same reasoning results may differ depending on their level of experience. By adding evaluator attribute information to evaluation data, metrics such as reliability and objectivity can be introduced to the evaluation. In other words, it becomes possible to evaluate the evaluation itself.
[0043] The processor 12 may read evaluation information EV stored in the storage 60 as needed and display it on the display device 20. The evaluation information EV may be read directly from the storage 60 or received through the input interface 11. In other words, the processor 12 may output a display control signal DC from the output interface 13 to display the evaluation information EV on the display device 20.
[0044] Figure 4 shows another example of the model selection screen 21 displayed on the display device 20. In this example, "Urine Volume Model 1" is identified by the first identifier 211, and "Urine Volume Model 2" is identified by the second identifier 212.
[0045] Figure 5 illustrates the display of pre-entered evaluation information EV for "Urine Volume Model 1". In this example, the evaluation information EV is displayed when the first identifier 211 is tapped by the user. Note that the evaluation information EV may be displayed on the model selection screen 21 at all times, depending on the layout specifications of the model selection screen 21.
[0046] Specifically, as illustrated in Figure 1, an instruction signal IS is generated in response to a display instruction for evaluation information EV input to the user interface 50. When the instruction signal IS is received by the input interface 11, the processor 12 reads the evaluation information EV corresponding to the instruction signal IS from the storage 60 and outputs a display control signal DC from the output interface 13 to display the evaluation information EV on the display device 20.
[0047] As illustrated in Figure 5, the evaluation information EV may include more detailed comments on the inference model. These comments can be entered into the user interface 50 via text or voice input by the user.
[0048] With this configuration, users can make clinical decisions while referring not only to the inference results of each inference model, but also to the accumulated evaluations of that inference model.
[0049] Figure 6 shows another example of the model selection screen 21 displayed on the display device 20. In this example, the colors of the first identifier 211 and the second identifier 212 are different. The color of the second identifier 212 indicates that the evaluation of "Urine Volume Model 2" is low. In other words, the processor 12 may be configured to change the display mode of at least one of the first identifier 211 and the second identifier 212 depending on the level of evaluation of the inference model. In this example, a display control signal DC that changes the color of the second identifier 212 to indicate that the evaluation of "Urine Volume Model 2" is low is output from the output interface 13.
[0050] Changes in the display manner can also be made by changing, in addition to or replacing, the color of the identifier, at least one of the shape of the identifier, the position of the identifier, and the text displayed with the identifier.
[0051] This configuration improves the accessibility of the accumulated evaluations for each inference system. It also facilitates guidance to more specific evaluation information for each inference system. Figure 7 illustrates how a user who wants to check the reason for the low evaluation of "Urine Volume Model 2" displays the evaluation information EV in the same way as the example shown in Figure 5.
[0052] In addition to or instead of the examples shown so far, evaluation information EV may include the number of times a particular evaluation was made, the number of times the inference model was used, and the number of evaluations relative to the number of uses.
[0053] In addition to or instead of the above-mentioned evaluation information EV, more detailed information relating to the inference results of each inference model may be displayed on the display device 20. An example of such detailed information is the change in the inference results over time.
[0054] Figure 8 illustrates the state in which the detailed information DT for "Infusion Model 1" is displayed when the first identifier 211 is tapped on the model selection screen 21 illustrated in Figure 2. As mentioned above, in the state illustrated in Figure 2, the probability that subject 30 is currently dehydrated, as inferred by "Infusion Model 1," is displayed along with the first identifier 211. The detailed information DT illustrated in Figure 8 shows the change in this probability, inferred by "Infusion Model 1," over time from the past to the present.
[0055] In this example, when the first identifier 211 is tapped by the user, the detailed information DT is displayed. The detailed information DT may also be displayed permanently on the model selection screen 21, depending on the layout specifications of the model selection screen 21.
[0056] Specifically, the processor 12 may be configured to save data corresponding to the inference result for "infusion model 1" to a storage area such as storage 60 each time an inference result is obtained. As illustrated in Figure 1, an instruction signal IS is generated in response to a display instruction for detailed information DT input to the user interface 50. When the instruction signal IS is received by the input interface 11, the processor 12 reads the detailed information DT corresponding to the instruction signal IS from storage 60 or the like, and outputs a display control signal DC from the output interface 13 to display the detailed information DT on the display device 20.
[0057] Figure 9 illustrates the state in which detailed information DT for "Infusion Model 2" is displayed in the same way as when the second identifier 212 is tapped on the model selection screen 21 as exemplified in Figure 2.
[0058] For example, even if two inference models produce the same inference result corresponding to the same clinical judgment at a given point in time, their inference algorithms may differ, resulting in different histories of inference results. The configuration described above allows for the examination of the validity of the inference results, including their history.
[0059] Furthermore, if the inference algorithm is designed to output future predictions for a specific clinical judgment, the time axis used to show the changes in the inference results over time in the detailed information DT may include future points in time.
[0060] As illustrated in Figure 10, the displayed detailed information DT may include an event identifier 215 indicating the occurrence of an event related to the subject information SB that occurred at a specific point in time. In this example, the event identifier 215 indicates that "increased fluid volume" was performed.
[0061] For example, a user can specify the type of event and the time when the event occurred through the user interface 50. The user interface 50 generates an instruction signal IS to add the information related to the specified event to the detailed information DT. When the instruction signal IS is received by the input interface 11, the processor 12 outputs a display control signal DC from the output interface 13 to add the event identifier 215 corresponding to the instruction signal IS to the display device 20.
[0062] Furthermore, a configuration may also be adopted in which the occurrence of a predetermined event related to a specific clinical judgment output by the inference model is automatically detected and reflected in the detailed information DT.
[0063] When multiple inference models exist that perform clinical judgment inferences on the same item, their behavior in response to specific events may differ due to differences in inference algorithms. With the above configuration, it becomes easier to evaluate the validity of the inference results from the perspective of their behavior in response to specific events. In addition, it becomes easier to understand what kinds of events influence the behavior of the inference models.
[0064] As illustrated in Figure 11, the displayed detailed information DT may include evaluation identifiers that indicate the evaluation given to the inference result at a specific point in time. In this example, evaluation identifiers 216a and 216b indicate that the inference result received a high rating. Evaluation identifier 216c indicates that the inference result received a low rating.
[0065] For example, a user can specify a specific point in time in the temporal changes of the inference result and an evaluation of the inference result at that point in time through the user interface 50. The user interface 50 generates an instruction signal IS to add the information related to the specified evaluation to the detailed information DT. When the instruction signal IS is received by the input interface 11, the processor 12 outputs a display control signal DC from the output interface 13 to add and display the evaluation identifier corresponding to the instruction signal IS on the display device 20.
[0066] With the configuration described above, it becomes possible to visualize which behaviors of the inference model are being evaluated and how, making it easier to understand the basis for the evaluation of that inference model.
[0067] The processor 12 can change the display manner of evaluation identifiers according to the elapsed time since the evaluation information EV was received. In the example shown in Figure 11, evaluation identifier 216b is displayed lighter than other evaluation identifiers. The processor 12 is configured to display evaluation identifiers that have been evaluated for a longer period of time in a lighter manner.
[0068] With this configuration, it is possible to distinguish between relatively new and relatively old evaluations on the same screen, allowing for an evaluation of the evaluation itself from the perspective of the time elapsed since the evaluation was made.
[0069] Figure 12 shows another example of the detailed information DT displayed on the display device 20. In this example, when the subject information SB input to the inference model includes multiple biometric parameters, the contribution of each biometric parameter to the inference result output from the inference model is displayed in the form of a bar graph. In addition to or instead of the contribution, the measured values of each biometric parameter may be displayed.
[0070] Figure 13 shows another example of the contribution display included in the detailed information DT. In this example, each of the multiple bioparameters included in the subject information SB is displayed in the form of a bar graph, showing whether it is acting to increase or decrease the value output from the inference model, along with the degree of that effect. A bar graph extending to the right indicates that the corresponding bioparameter is acting to increase the value output from the inference model. A bar graph extending to the left indicates that the corresponding bioparameter is acting to decrease the value output from the inference model. In addition to or instead of the contribution, the measured values of each bioparameter may also be displayed.
[0071] Figure 14 shows another example of the contribution display included in the detailed information DT. In this example, the extent to which each part of the measurement waveform of the bioparameters included in the subject information SB contributes to the inference results output from the inference model is displayed in the form of a heatmap.
[0072] With reference to Figures 12 to 14, the configurations for each example described make it possible to visualize how the subject information SB influences the inference results, thus making it easier to evaluate the validity of the inference results from the perspective of the behavior in response to specific biological parameter inputs. In addition, it becomes easier to understand which biological parameters influence the behavior of the inference model.
[0073] The display control device 10, the first inference model 41, the second inference model 42, the display device 20, the user interface 50, and the storage 60 described above can constitute a model evaluation system 70 for evaluating multiple inference models. The model evaluation system 70 can take various forms.
[0074] Figure 15 shows an example of the configuration of the model evaluation system 70. The display device 20, the user interface 50, and the storage 60 are each connected to the display control device 10 via a communication network N.
[0075] The model evaluation system 70 may include a server device 71. The server device 71 is communicated with the display control device 10 via a communication network N. At least one of the first inference model 41 and the second inference model 42 may be mounted on the display control device 10 or on the server device 71. In the example shown in Figure 15, the first inference model 41 is mounted on the display control device 10 and the second inference model 42 is mounted on the server device 71. When multiple server devices are connected to the communication network N to constitute the model evaluation system 70, the first inference model 41 and the second inference model 42 may be mounted on different server devices.
[0076] The display device 20 may be mounted on the display control device 10. The user interface 50 may be mounted on the display device 20 or on the display control device 10. The storage 60 may be mounted on the display control device 10 or on the server device 71.
[0077] As illustrated in Figure 1, N display control devices 10 (101 to 10N; N is an integer of 2 or more) can be communicated to the storage 60. That is, evaluation information EV output from each display control device 10 can be centrally stored in the storage 60.
[0078] This configuration not only facilitates the accumulation of more evaluations of the inference model, but also facilitates the sharing of evaluations by multiple healthcare professionals. Furthermore, it enables ensemble learning of new inference models using the inference results and evaluation information collected from each display control device.
[0079] The processor 12 of the display control device 10, which has the various functions described above, can be implemented by a general-purpose microprocessor that works in cooperation with general-purpose memory. Examples of general-purpose microprocessors include CPUs, MPUs, and GPUs. Examples of general-purpose memory include ROMs and RAMs. In this case, the ROM may store a computer program that implements the function in question. ROM is an example of a non-temporary computer-readable medium that stores computer programs. The general-purpose microprocessor selects at least a portion of the program stored in the ROM and loads it onto the RAM, and then works in cooperation with the RAM to execute the above-described process. The computer program may be pre-installed in the general-purpose memory, or it may be downloaded from an external server device via a communication network and then installed in the general-purpose memory. In this case, the external server device is an example of a non-temporary computer-readable medium that stores computer programs.
[0080] The processor 12 may be implemented by a dedicated integrated circuit such as a microcontroller, ASIC, or FPGA capable of executing the above-mentioned computer program. In this case, the above-mentioned computer program is pre-installed in a memory element included in the dedicated integrated circuit. This memory element is an example of a computer-readable medium that stores the computer program. The processor 12 can also be implemented by a combination of a general-purpose microprocessor and a dedicated integrated circuit.
[0081] The various configurations described herein are merely examples to facilitate understanding of this disclosure. Each configuration example can be modified or combined with other configuration examples as appropriate.
[0082] The configurations listed below also constitute part of this disclosure. Item 1: An interface for receiving information about the target person, A processor that controls the information displayed on the display device, It is equipped with, The aforementioned processor, The subject information is input into the first inference model to obtain a first inference result corresponding to the clinical judgment obtained for the subject for a predetermined item. The subject information is input into the second inference model to obtain a second inference result corresponding to the clinical judgment obtained for the subject for the predetermined items. The first inference result is displayed on the display device along with a first identifier that identifies the first inference model. The second inference result is displayed on the display device along with a second identifier that identifies the second inference model. Display control device. Item 2: The interface receives evaluation information corresponding to an evaluation of at least one of the first inference result and the second inference result, The processor stores the evaluation information in storage. The display control device described in item 1. Item 3: The aforementioned evaluation information includes attribute information of the person who performed the evaluation. The display control device described in item 2. Item 4: The processor causes the evaluation information to be displayed on the display device. A display control device as described in item 2 or 3. Item 5: The processor changes the display mode of at least one of the first identifier and the second identifier according to the level of the evaluation. A display control device as described in any of items 2 to 4. Item 6: The processor causes the display device to show at least one of the changes in the first inference result over time and the changes in the second inference result over time. Display control devices as described in items 1 to 5. Item 7: The interface receives event information indicating the occurrence of an event related to the subject information at a specific point in time. The processor displays an identifier corresponding to the event at a position corresponding to the time in the time progression. The display control device described in item 6. Item 8: The interface receives evaluation information corresponding to an evaluation of at least one of the first inference result and the second inference result at a specific point in time. The processor displays an identifier corresponding to the evaluation at a position corresponding to the time in the time course. A display control device as described in item 6 or 7. Item 9: The processor changes the display mode of the identifier according to the elapsed time since the evaluation information was received. The display control device described in item 8. Item 10: The processor causes the display device to display, for at least one of the first inference result and the second inference result, the measured values of the multiple biological parameters included in the subject information and at least one of their contributions to the inference result. A display control device as described in any of items 1 through 9. [Explanation of symbols]
[0083] 10: Display control device, 11: Input interface, 12: Processor, 20: Display device, 211: First identifier, 212: Second identifier, 215: Event identifier, 216a~216c: Evaluation identifier, 30: Subject, 41: First inference model, 42: Second inference model, 60: Storage, DT: Detailed information, EV: Evaluation information, SB: Subject information
Claims
1. An interface for receiving information about the target person, A processor that controls the information displayed on the display device, It is equipped with, The aforementioned processor, The subject information is input into the first inference model to obtain a first inference result corresponding to the clinical judgment obtained for the subject for a predetermined item. The subject information is input into the second inference model to obtain a second inference result corresponding to the clinical judgment obtained for the subject for the predetermined items. The first inference result is displayed on the display device along with a first identifier that identifies the first inference model. The second inference result is displayed on the display device along with a second identifier that identifies the second inference model. Display control device.
2. The interface receives evaluation information corresponding to an evaluation of at least one of the first inference result and the second inference result, The processor stores the evaluation information in storage. The display control device according to claim 1.
3. The aforementioned evaluation information includes attribute information of the person who performed the evaluation. The display control device according to claim 2.
4. The processor causes the evaluation information to be displayed on the display device. The display control device according to claim 2.
5. The processor changes the display mode of at least one of the first identifier and the second identifier according to the level of the evaluation. The display control device according to claim 2.
6. The processor causes the display device to show at least one of the changes in the first inference result over time and the changes in the second inference result over time. The display control device according to claim 1.
7. The interface receives event information indicating the occurrence of an event related to the subject information at a specific point in time. The processor displays an identifier corresponding to the event at a position corresponding to the time in the time progression. The display control device according to claim 6.
8. The interface receives evaluation information corresponding to an evaluation of at least one of the first inference result and the second inference result at a specific point in time. The processor displays an identifier corresponding to the evaluation at a position corresponding to the time in the time course. The display control device according to claim 6.
9. The processor changes the display mode of the identifier according to the elapsed time since the evaluation information was received. The display control device according to claim 8.
10. The processor causes the display device to display, for at least one of the first inference result and the second inference result, the measured values of the multiple biological parameters included in the subject information and at least one of their contributions to the inference result. The display control device according to claim 1.
11. A computer program that can be executed by a processor mounted on a display control device, By being executed, the display control device will We accept information about the target individuals. The subject information is input into the first inference model to obtain a first inference result corresponding to the clinical judgment obtained for the subject for a predetermined item. The subject information is input into the second inference model to obtain a second inference result corresponding to the clinical judgment obtained for the subject for the predetermined items. The first inference result is displayed on the display device along with a first identifier that identifies the first inference model. The second inference result is displayed on the display device along with a second identifier that identifies the second inference model. Computer program.