Viewing assistance device, viewing assistance method, and viewing assistance program

Abstract

This viewing assistance device comprises a processor. The processor acquires situation information regarding a situation when a user views medical care data, and outputs viewing recommendation levels for medical care data of a plurality of viewing candidates derived on the basis of the situation information.

Description

Browsing Support Device, Browsing Support Method, and Browsing Support Program

[0001] The present disclosure relates to a browsing support device, a browsing support method, and a browsing support program.

[0002] Regarding the medical treatment of patients, a technology for making recommendations regarding medical treatment to a user according to the situation is known. For example, in the technology described in Japanese Unexamined Patent Application Publication No. 2023-149917, a technology for determining the medical treatment stage of a patient and presenting a recommended treatment to a user who is a medical worker from among a plurality of treatments that can be performed on the patient is described.

[0003] By the way, in the medical field, depending on the situation when a user views medical treatment data, the medical treatment data desired to be viewed may be different, and it is desired that the desired medical treatment data can be easily viewed.

[0004] However, in the above conventional technology, since it presents a recommended treatment based on the current medical treatment stage of the patient, it does not recommend the medical treatment data that the user desires to view, and such a technology has been desired.

[0005] The present disclosure has been made in view of the above circumstances, and an object thereof is to provide a browsing support device, a browsing support method, and a browsing support program that can easily view the medical treatment data desired by the user.

[0006] In order to achieve the above object, the browsing support device of the present disclosure includes a processor, and the processor acquires situation information regarding the situation when a user views medical treatment data, and outputs a browsing recommendation degree for medical treatment data of a plurality of browsing candidates derived based on the situation information.

[0007] Also, in order to achieve the above object, the browsing support method of the present disclosure acquires situation information regarding the situation when a user views medical treatment data, and outputs a browsing recommendation degree for medical treatment data of a plurality of browsing candidates derived based on the situation information.

[0008] Furthermore, in order to achieve the above objectives, the browsing support program of this disclosure acquires situational information regarding the user's browsing of medical data, and causes a computer to execute a process that outputs a browsing recommendation score for multiple candidate medical data based on the situational information.

[0009] According to this disclosure, users can easily access the medical data they desire.

[0010] This is a diagram illustrating the overview of the browsing support device of the embodiment. This is a block diagram showing an example of the configuration of the browsing support device of the embodiment. This is a functional block diagram showing an example of the configuration of the browsing support device of the embodiment. This is a diagram showing an example of a settings screen. This is a diagram illustrating the outline of the processing of the derivation unit. This is an example of a display screen. This is a flowchart showing an example of the browsing support processing flow by the browsing support device of the embodiment.

[0011] Embodiments of this disclosure will be described in detail below with reference to the drawings. These embodiments are not intended to limit the technology of this disclosure.

[0012] As shown in Figure 1, the browsing support device 10 of this embodiment is connected to the database group 12 via the network N. The browsing support device 10 and the database group 12 are able to communicate with each other via the network N.

[0013] Database group 12 includes various databases that are referenced when obtaining status information, which will be described in detail later. As will be described in detail later, examples include information on attendance management in hospitals, scheduled information which is information on staff schedules, and scheduled information which is information on patients' schedules.

[0014] Furthermore, database group 12 includes multiple medical data sets that users may view. The medical data includes at least one of the following: medical image data such as MRI (Magnetic Resonance Imaging) images and CT (Computed Tomography) images; test result data such as pathology test results; and medical document data such as medical documents. Specific examples include data related to radiology, endoscopy, physiology, pathology, ultrasound, treatment plans, medical questionnaires, patient consent forms or appointment slips for examinations and procedures, explanations regarding examinations and procedures, and discharge summaries.

[0015] The specific storage location of the database group 12 is not limited. For example, the database group 12 may be stored on different server devices depending on the type of database. For example, medical image data may be stored in a PACS (Picture Archiving and Communication System).

[0016] The browsing support device 10 of this embodiment is a device that supports users, such as doctors and technicians, in browsing medical data by outputting a browsing recommendation level for multiple candidate medical data based on situation information regarding the user's browsing status at the time of browsing. The browsing support device 10 refers to the database group 12 via the network N and acquires situation information and candidate medical data. The "browsing recommendation level" is the degree to which the user is recommended to browse that medical data, and for example, the probability that the user will browse that medical data may be used.

[0017] Figure 2 shows a block diagram illustrating an example of the configuration of the browsing support device 10 of this embodiment. As shown in Figure 2, the browsing support device 10 of this embodiment comprises a control unit 20, a storage unit 22, a communication interface (I / F) unit 24, an operation unit 26, and a display unit 28. The control unit 20, storage unit 22, communication interface unit 24, operation unit 26, and display unit 28 are connected to each other via a bus 29, such as a system bus or a control bus, enabling the exchange of various types of information.

[0018] The control unit 20 in this embodiment controls the overall operation of the browsing support device 10. The control unit 20 is a processor and includes a CPU (Central Processing Unit) 20A. The control unit 20 is also connected to the storage unit 22, which will be described later. The control unit 20 may also include a GPU (Graphics Processing Unit).

[0019] The operation unit 26 is used for the user to input information related to viewing medical data. The operation unit 26 is not particularly limited and may include, for example, various switches, a touch panel, a stylus, a mouse, and a microphone for voice input. The display unit 28 displays the final output document and various information. The display unit 28 may include, for example, a liquid crystal display or an LED (Light Emitting Diode) display. The operation unit 26 and the display unit 28 may be integrated to form a touch panel display.

[0020] The communication interface unit 24 communicates various types of information with external devices of the browsing support device 10, such as the database group 12, via wireless or wired communication over the network N.

[0021] The memory unit 22 includes a ROM (Read Only Memory) 22A, a RAM (Random Access Memory) 22B, and a storage unit 22C. The ROM 22A has various programs and the like that are pre-stored in it, which are executed by the CPU 20A. The RAM 22B temporarily stores various data. The storage unit 22C stores the browsing support program 30 executed by the CPU 20A, and various other information. In this embodiment, the storage unit 22C also has a pre-stored learned derivation model 32, which will be described in detail later. The storage unit 22C is a non-volatile memory unit, and examples include an HDD (Hard Disk Drive) and an SSD (Solid State Drive).

[0022] Furthermore, Figure 3 shows a functional block diagram of an example configuration of the browsing support device 10 of this embodiment. As shown in Figure 3, the browsing support device 10 comprises a reception unit 40, a status information acquisition unit 42, a medical data acquisition unit 44, an output unit 46, and a display control unit 48. As an example, in the browsing support device 10 of this embodiment, the CPU 20A of the control unit 20 executes the browsing support program 30 stored in the storage 22C, so that the CPU 20A functions as the reception unit 40, the status information acquisition unit 42, the medical data acquisition unit 44, the output unit 46, and the display control unit 48.

[0023] The reception unit 40 receives instructions regarding browsing assistance made by the user via the operation unit 2. In the browsing assistance device 10 of this embodiment, the user can make instructions regarding browsing assistance. Figure 4 shows an example of a setting screen 80 displayed on the operation unit 26 when the user makes instructions regarding browsing assistance. The user makes instructions regarding browsing assistance by making settings on the setting screen 80. Instructions regarding browsing assistance include, for example, instructions indicating the range of medical data to be extracted as browsing candidates, or instructions indicating the range of medical data to which the derivation conditions used to derive the browsing recommendation level will be changed.

[0024] In the example shown in Figure 4, the user can give instructions by checking the respective checkboxes to determine whether to leave the recommendation of candidate medical data to the viewing support device 10, to always display only the latest data, to display reports if they are attached, and to always hide (not display) reports for various types of medical data such as "diagnostic documents," "MRI," "CT," "general radiology," "upper endoscopy," "lower endoscopy," "ultrasound," and "tissue biopsy." In the initial state, i.e., when no instructions have been received from the user, the checkbox corresponding to leaving the recommendation to the viewing support device 10 is checked. In this embodiment, instructions regarding viewing support are given based on information such as the type of medical data stored in association with the medical data, whether the medical data is the latest data or not, and whether a report is attached to the medical data, but this is not limited to any information stored in association with the medical data. For example, information stored in association with the medical data may include information on the medical department that requested the creation of the medical data, information on the target organ of the medical data, information on the freshness of the medical data, and viewing priority information such as triage. If the instructions regarding access support involve information on the freshness of medical data, for example, the user can give instructions indicating a period such as "within one month," which may change the conditions for deriving the recommendation level for viewing medical data within the period specified by the user.

[0025] By configuring settings on the settings screen 80, the user can specify the derivation conditions used to derive the viewing recommendation level. These derivation conditions include conditions for which medical data should have a high or low recommendation level. For example, the user can specify which types of medical data should have a high recommendation level, i.e., which types of medical data should be prioritized, among multiple types of medical data. For example, in the example shown in Figure 4, for MRI, where "Always show only the latest" is checked, the derivation condition is set to give the highest viewing recommendation level to the latest medical data related to MRI.

[0026] Furthermore, by configuring settings on the settings screen 80, the user can specify which medical data from which viewing recommendations should be derived from among multiple viewing candidate medical data. For example, the user can specify that only MRI-related medical data from among multiple types of medical data should have viewing recommendations derived. For example, in the example shown in Figure 4, medical data related to medical documents that have the "Always hide" checkbox checked should not have viewing recommendations derived. Therefore, among the multiple types of viewing candidate medical data 60, medical data other than medical documents will be specified as the target for which viewing recommendations should be derived.

[0027] The reception unit 40 receives the settings made by the user on the settings screen 80, acquires the received settings, and outputs them to the derivation unit 46.

[0028] The status information acquisition unit 42 acquires status information regarding the circumstances when a user views medical data. The circumstances at the time of viewing include at least one of the user's circumstances, the patient's circumstances, the circumstances regarding the medical data, and the circumstances of the device being used for viewing. The status information is information used to identify these circumstances.

[0029] User information includes at least one of the user's attendance status and work status. Depending on the attendance status, the medical data a user may want to view may differ. For example, the medical data a user may want to view during a conference may differ from the medical data a user may want to view during ward rounds. Therefore, status information regarding the user's attendance status may be acquired. Examples of such status information include the day of the week, time of day, before / during / after a conference, before / during / after outpatient consultations, before / during / after ward rounds, before / after a shift change or handover, etc. Also, depending on the user's work status, the medical data a user may want to view may differ. For example, the medical data a user may want to view when preparing for an examination may differ from the medical data a user may want to view while actually performing the examination. Therefore, status information regarding the user's work status may be acquired.

[0030] The method by which the status information acquisition unit 42 acquires status information representing at least one of the user's attendance status and work status is not limited. For example, the status information acquisition unit 42 may acquire status information corresponding to the current date and time information (including the day of the week) for which the user is being assisted by the browsing support device 10 from the attendance sheet (schedule sheet) managed by the hospital, which is included in the database group 12.

[0031] Furthermore, as a method for identifying attendance status or work status from situational information, for example, if a certain medical department has outpatient hours during a specific time slot on a certain day of the week, then the attendance status of a doctor in that department can be identified as being in outpatient practice during that time slot. Alternatively, even without explicitly identifying attendance status, attendance status can be identified by indirectly estimating the correspondence between attendance status and viewing trends from the current date and time information (including the day of the week) provided by the viewing support device 10 to assist the user. Specifically, attendance status can be indirectly identified from viewing trends, such as the fact that a doctor in a certain medical department often views CT images at a specific time on a particular day.

[0032] Furthermore, the patient's status refers to the patient's condition corresponding to the medical data viewed by the user. For example, this includes the time elapsed since hospitalization, discharge, and surgery, as well as the treatment stage and the patient's condition. Therefore, the status information in this case includes information representing at least one of the patient's diagnosis and treatment stage.

[0033] Furthermore, as a method for the status information acquisition unit 42 to identify the patient's status, for example, the status information acquisition unit 42 may identify the patient's status by referring to the patient's medical record information included in the database group 12. Alternatively, for example, the patient's status related to surgery may be identified from the hospital's surgical management system included in the database group 12.

[0034] Furthermore, regarding recommendations for medical data, if multiple modes such as conference mode, outpatient mode, and ward mode are available for user selection, the user's situation and the patient's situation may be identified based on the mode selected by the user.

[0035] Furthermore, status information regarding the status of medical data includes, for example, information on the freshness of the medical data. Here, "freshness of medical data" refers to the newness of the medical data based on the time information from which the status information acquisition unit acquires the status information. For example, this could include the time elapsed since the creation of the medical data, whether or not it is the latest information, whether or not it is the latest information for each type of medical data such as modality, whether or not it has been viewed by a user in the past, or whether or not it has been viewed by other users in the same profession or department as the user. For example, medical data may be associated with time information indicating the time of creation and its type, and status information may be acquired by comparing the freshness of the medical data by comparing the time information of the medical data for each type. Alternatively, viewing information such as the user who viewed the medical data may be associated and stored for each piece of medical data, and status information may be acquired by determining whether or not a user has viewed it in the past. Note that the time information includes information indicating at least one of the date and time, day, month, and year.

[0036] Furthermore, when the status information acquisition unit 42 acquires status information regarding the status of medical treatment data, the status information is stored in the database group 12. For example, when a user views certain medical treatment data, the user who viewed the data and viewing information such as the date and time of viewing are stored in association with the medical treatment data.

[0037] Furthermore, the status of the viewing device refers to the status of the device on which the user views the medical data, i.e., the device on which the medical data is displayed. For example, if the display area is small, the user may not view detailed data. Also, the type of device and the size of the display area may affect the time required to access the medical data. For this reason, it is also advisable to obtain the status of the viewing device. In this case, the status information may include, for example, the type of device, the size of the display area, the resolution, and location information.

[0038] The status information acquisition unit 42 outputs the acquired status information to the derivation unit 46.

[0039] The medical data acquisition unit 44 acquires candidate medical data 60 from the database group 12. The method by which the status information acquisition unit 42 acquires candidate medical data 60 is not limited. For example, if the user inputs the identification information of the patient to be viewed, the medical data associated with that patient may be acquired as candidate medical data 60 based on the identification information. Alternatively, the status information acquisition unit 42 may acquire candidate medical data 60 according to the status information acquired by the status information acquisition unit 42. The medical data acquisition unit 44 outputs the acquired candidate medical data 60 to the output unit 46.

[0040] The derivation unit 46 derives a viewing recommendation level for each of the multiple viewing candidate medical data 60 based on the situation information. In this embodiment, the derivation unit 46 identifies the situation at the time of viewing based on the situation information and derives the viewing recommendation level using the derivation model 32. As shown in Figure 5, the derivation unit 46 inputs the situation at the time of viewing 50 and the viewing candidate medical data 60 into the derivation model 32 and obtains the viewing candidate medical data 60 to which the viewing recommendation level has been assigned, which is output from the derivation model 32. In the example shown in Figure 5, the state in which a viewing recommendation level has been output for each of the viewing candidate medical data 60-1 to 60-4 included in the viewing candidate medical data 60 is shown.

[0041] The derivation model 32 used to derive the viewing recommendation score can be obtained, for example, by training a machine learning model as follows. As a machine learning model, for example, LightGBM (Light Gradient Boosting Machine), which is a type of gradient boosting decision tree model, can be used.

[0042] At a certain point, there are 10 pieces of medical data for a certain patient from A to J. If the medical data actually viewed at that time is C, the viewed medical data C is a positive example (Label = 1), and the other 9 pieces of medical data are negative examples (Label = 0) to construct the teacher data. The parameters of the machine learning model are learned so that the browsing recommendation degree output by the machine learning model for positive examples is high, and the browsing recommendation degree output for negative examples is low. In other words, the parameters of the model are updated (learned) so that the output value of the machine learning model is large (i.e., close to 1) for positive examples and small (i.e., close to 0) for negative examples. In addition, based on the confidence level of the situation to be described later, the learning data may be weighted.

[0043] Also, the hyperparameters of the machine learning model are adjusted using a part of the learning data. Specifically, using Optuna, which is one of the Bayesian optimization tools, the hyperparameters are adjusted so that the prediction error becomes small for 60 pieces of medical data of browsing candidates obtained after that period when learning with medical data for a certain period.

[0044] The learned machine learning model learned in this way is stored in advance in the storage 22C of the browsing support device 10 as described above in this embodiment.

[0045] When the reception unit 40 is receiving the above settings, the derivation unit 46 derives the recommendation degree of the medical data 60 of the browsing candidates based on the received settings.

[0046] For example, if the medical data 60 selected as viewing candidates is to exclude the medical data specified by the user, the derivation unit 46 may exclude the medical data specified by the user from the viewing candidate medical data 60 acquired by the medical data acquisition unit 44, and then input the remaining viewing candidate medical data 60 into the derivation model 32. Alternatively, for example, the derivation unit 46 may input all the viewing candidate medical data 60 acquired by the medical data acquisition unit 44 into the derivation model 32, and then exclude the medical data specified by the user from the viewing candidate medical data 60 to which a viewing recommendation score has been assigned, output from the derivation model 32. When using a derivation model 32 that derives a viewing recommendation score considering the correlation between medical data, it is preferable to apply the latter method.

[0047] On the other hand, for example, if it is to always recommend the medical data instructed by the user, the medical data acquisition unit 44 may exclude the medical data instructed by the user from the medical data candidates 60 acquired, and the remaining medical data candidates 60 may be input into the derivation model 32 to obtain a viewing recommendation score, and the viewing recommendation score for the medical data instructed by the user may be set higher than these. Alternatively, for example, the derivation unit 46 may input all the medical data candidates 60 acquired by the medical data acquisition unit 44 into the derivation model 32, and among the medical data candidates 60 to which a viewing recommendation score has been assigned output from the derivation model 32, the viewing recommendation score assigned to the medical data instructed by the user may be changed to the highest value.

[0048] Also, for example, when preferentially recommending the medical data instructed by the user, for example, the derivation unit 46 may perform weighting so as to increase or decrease the recommendation degree derived by the derivation model 32 according to the user's instruction. As a specific aspect, for example, when the reception unit 40 receives an instruction to preferentially view MRI, the derivation unit 46 makes the recommendation degree of the MRI data relatively larger than the viewing recommendation degree derived by the derivation model 32 among the medical data 60 of the viewing candidates. For example, the derivation unit 46 doubles the recommendation degree of the MRI data compared to the viewing recommendation degree derived by the derivation model 32. Also, for example, when the reception unit 40 receives an instruction that it does not want to view CT very much, the derivation unit 46 makes the recommendation degree of the CT data relatively smaller than the viewing recommendation degree derived by the derivation model 32 among the medical data 60 of the viewing candidates. The derivation unit 46, for example, sets the recommendation degree of the MRI data to 1 / 2 times the viewing recommendation degree derived by the derivation model 32.

[0049] Also, the derivation unit 46 may derive the viewing recommendation degree based on the confidence level for the situation. The "confidence level for the situation" refers to the degree of likelihood of that situation. For example, regarding the user's attendance status, if the probability of being in a conference is 70% and the probability of being in an outpatient medical treatment is 30%, the viewing recommendation degree assuming each situation at a ratio of 7:3 is derived. Specifically, the derivation unit 46 derives, as the viewing recommendation degree, a value obtained by adding a value obtained by multiplying the viewing recommendation degree in the case of being in a conference by 0.7 and a value obtained by multiplying the viewing recommendation degree in the case of being in an outpatient medical treatment by 0.3.

[0050] The confidence level may be derived, for example, based on the periodic pattern in which the situation occurs, and more specifically, based on the degree of deviation from the periodic pattern. The periodic pattern in which the situation occurs may be obtained from the results of analyzing the user's or patient's schedule, or the user's input during browsing. A specific example of how to derive confidence levels based on periodic patterns will be explained. First, date and time information is obtained. For example, date and time information for 13:00 on the 21st (Thursday) is obtained. Next, past situations are obtained at regular intervals based on the obtained date and time information. In the above example, if the regular interval is every week, the situation at 13:00 on the 7th (Thursday) and the situation at 13:00 on the 14th (Thursday) are obtained. Next, it is determined whether the situations are common between the past data. In the above example, if the situation at 13:00 on the 7th (Thursday) and 13:00 on the 14th (Thursday) is ward rounds, then the confidence level that the situation at 13:00 on the 21st (Thursday) is ward rounds will be high. On the other hand, if the situation at 13:00 on the 7th (Thursday) and 13:00 on the 14th (Thursday) was ward rounds, the degree of confidence that the situation at 13:00 on the 21st (Thursday) was during a conference would be lower.

[0051] In addition, as a concrete example of a more quantitative approach than the above, the degree of deviation from the cyclical pattern may be derived based on the number of times the situation occurred within a predetermined number of times during the same time period in the past, specifically, within the same time period on the same day of the week. For example, if 40 out of 50 times (80%) were ward rounds and 10 times were other situations, the deviation for ward rounds would be 20% (100 - 80 = 20). In this case, the "time period" can be in one-hour units, AM / PM units, or daily units, but it is preferable to broaden the time period as it can absorb schedule discrepancies.

[0052] Furthermore, the confidence level may be derived based on the completion rate or the number of characters in the input fields for the schedule information. For example, when a user enters schedule information such as work schedules, if they enter detailed information, the confidence level of that schedule (situation) can be considered high. Therefore, regarding the user's situation, if the completion rate for all input fields for the schedule information is high, or if the number of characters is large, the confidence level of that situation will be high. Also, for example, in the case of a patient with many blank or undecided fields in their hospitalization plan or treatment plan, the situation can be considered uncertain. Therefore, regarding the patient's situation, if the completion rate for all input fields for the schedule information is low, or if the number of characters is small, the confidence level of that situation will be low.

[0053] Furthermore, for example, the confidence level may be derived based on whether or not the user provided instructions regarding browsing assistance. In this embodiment, the confidence level may also be derived based on whether or not the user made a setting in the settings screen 80 (see Figure 4) described above, or whether or not they changed it from the default setting.

[0054] Furthermore, the derivation unit 46 may derive a viewing recommendation level according to the time required to access the candidate medical data 60. For example, the viewing recommendation level derived in the manner described above may be weighted according to the time required to access. Alternatively, the viewing recommendation level may be increased or decreased in stages according to the time required to access.

[0055] Furthermore, the longer the time required to access the candidate medical data 60, the higher the recommendation level for viewing will be.

[0056] When using access time in this way, it is possible to estimate that the access time will be longer if the medical data 60 of the viewing candidates are less fresh, as they may be more difficult to access or find. Also, the access time may differ depending on the device on which the medical data 60 of the viewing candidates are displayed. For example, if the display size is large, the number of medical data 60 of the viewing candidates that can be displayed simultaneously will increase, so the time the user spends scrolling the screen to view the medical data 60 of the viewing candidates will be shorter, and therefore the access time may be shorter. Also, for example, whether the display device is a touch panel display or whether a mouse is connected may change the time the user spends scrolling the screen to view the medical data 60 of the viewing candidates, and therefore the access time may change. For this reason, it is also possible to estimate the access time using information about the device on which the medical data 60 of the viewing candidates are displayed.

[0057] The derivation unit 46 outputs the viewing recommendation score derived as described above to the display control unit 48. The manner in which the derivation unit 46 outputs the viewing recommendation score is not limited. For example, it is also possible to output each of the candidate medical data 60 by assigning a numerical value or rank as an ordinal scale according to the derived viewing recommendation score. Alternatively, it is also possible to output a list (details described later) in which the candidate medical data 60 are arranged in order of viewing recommendation score.

[0058] Furthermore, the derivation unit 46 may output all of the medical data 60 of the viewing candidates for which the viewing recommendation level has been derived, or it may output only the medical data 60 of the selected viewing candidates. In this case, for example, the number of medical data 60 of the viewing candidates to be selected may be a number corresponding to the viewing recommendation level. An example of a mode in which a number corresponding to the viewing recommendation level is to select a predetermined number of medical data 60 of the viewing candidates, starting with those with the highest viewing recommendation levels. Another example is to select medical data 60 of the viewing candidates whose viewing recommendation level is equal to or greater than a predetermined value. Another example is to select the medical data 60 of the viewing candidates such that the sum of their viewing recommendation levels is equal to or greater than a threshold. As a specific example, if the viewing recommendation levels of the five medical data 60 of the viewing candidates are "0.5", "0.3", "0.25", "0.2", and "0.1", and the threshold is "1", then the mode in which the medical data 60 of the three viewing candidates with viewing recommendation levels of "0.5", "0.3", and "0.25" will be selected.

[0059] Furthermore, the number of selected items may be increased or decreased depending on the number of medical data 60 candidates whose viewing recommendation level exceeds a predetermined value. For example, if the viewing recommendation level of all medical data 60 candidates is below the predetermined value, the number of selected items may be increased relatively compared to other cases. For example, the number of selected items may be doubled. Also, the number of selected items may be increased depending on the number of medical data 60 candidates whose viewing recommendation level is below the predetermined value.

[0060] Furthermore, the number of candidate medical data 60 available for viewing varies depending on the medical department. For example, the more candidate medical data 60 available, the more difficult it becomes to meet the user's needs. Therefore, the number of selections may be increased or decreased depending on the medical department to which the user belongs or the medical department to which the patient receives treatment. Alternatively, the user's viewing habits may be accumulated, and the number of selections may be increased or decreased accordingly. A specific example is increasing the number of selections if the user tends to view a large number of items.

[0061] The display control unit 48 controls the display of information regarding medical data that the user is recommended to view, according to the recommendation level. The display device is the device on which the user views the medical data, and is not limited to the display unit 28 of the viewing support device 10. When the user views the medical data on a device other than the viewing support device 10, the display control unit 48 outputs the recommendation level and candidate medical data 60, or the above list (details described later), to the viewing device via the network N.

[0062] Figure 6 shows an example of how a list of medical data is displayed according to its recommended viewing level. Figure 6 is an example of how a list 64 is displayed based on a recommended viewing level derived from a situation where a user is viewing medical data 61, which is an MRI image, during a patient's medical examination. In the example shown in Figure 6, the medical data 60 with a higher recommended viewing level are displayed at the top of the list 64. By displaying the medical data 60 with a higher recommended viewing level at the top of the list 64 in this way, the medical data 60 with a higher recommended viewing level are more likely to catch the user's eye.

[0063] In other words, List 64 shows the recommended viewing level for the candidate medical data 60. As such, since List 64 in this embodiment shows the recommended viewing level, the user can decide which medical data to view by referring to the recommended viewing level. For this reason, the viewing support device 10 of this embodiment makes it easy to view the desired medical data.

[0064] Furthermore, if there is a candidate medical data 60 that the user has instructed to be prioritized for viewing, it will be displayed at the top of the list 64 in a way that makes it distinguishable from other candidate medical data 60. In the example shown in Figure 6, candidate medical data 60-1, which is an endoscopic image that the user has instructed to be prioritized for viewing, is highlighted at the top of the list 64 by being surrounded by a thick line.

[0065] Furthermore, as shown in Figure 6, information indicating whether the candidate medical data 60 is the latest medical data may also be displayed. In particular, if there are multiple medical data of the same type taken of the same body part, it is advisable to display information indicating whether it is the latest information. Also, as shown in Figure 6, information indicating whether the user has already read the data may also be displayed.

[0066] Next, the operation of the browsing support device 10 of this embodiment will be described with reference to the drawings. Figure 7 shows a flowchart illustrating an example of the flow of browsing support processing performed in the browsing support device 10 of this embodiment. In this embodiment, the browsing support device 10 executes the browsing support processing shown as an example in Figure 7 by having the CPU 20A of the control unit 20 execute the browsing support program 30 stored in the storage 22C based on a user's start instruction made by the operation unit 26.

[0067] In step S100 of Figure 7, the reception unit 40 determines whether it has received the instructions or settings instructed by the user on the settings screen 80, as described above. If it has received the instructions, the determination in step S100 becomes positive, and the process proceeds to step S102. In step S102, the reception unit 40 obtains the settings and then proceeds to step S104.

[0068] On the other hand, if the settings are not accepted, the determination in step S100 becomes a negative determination, and the process proceeds to step S104. In step S104, the status information acquisition unit 42 acquires status information regarding the user's status when viewing medical data, as described above.

[0069] In the next step S106, the medical data acquisition unit 44 acquires the medical data 60 of the viewing candidates as described above.

[0070] In the next step S108, the derivation unit 46 derives the viewing recommendation level for the candidate medical data 60. In this embodiment, as described above, the derivation unit 46 inputs the viewing situation 50 and the candidate medical data 60 into the derivation model 32 and obtains the candidate medical data 60 to which the viewing recommendation level has been assigned, output from the derivation model 32 (see Figure 5).

[0071] In the next step S110, the display control unit 48 displays candidate medical data 60 according to the recommended viewing level. In this embodiment, as described above, the derivation unit 46 displays a list 64 of candidate medical data 60 arranged according to the recommended viewing level (see Figure 6).

[0072] In the next step S112, the display control unit 48 determines whether or not to terminate this process. If it does not terminate, the determination in step S112 is negative, and the process returns to step S100, repeating steps S100 to S110. On the other hand, if it decides to terminate this process, the determination in step S112 is positive, and the browsing recommendation process shown in Figure 7 is terminated.

[0073] As described above, the browsing support device 10 of this embodiment includes a status information acquisition unit 42 and an output unit 46. The status information acquisition unit 42 acquires status information regarding the user's status when viewing medical data. The display control unit 48 outputs a viewing recommendation score for a plurality of viewing candidate medical data, which is derived based on the status information.

[0074] Thus, the browsing support device 10 of this embodiment outputs a browsing recommendation score derived based on situational information for multiple candidate medical data, and can present the browsing recommendation score to the user. This allows the user to decide which medical data to view based on the browsing recommendation score. Therefore, the browsing support device 10 of this embodiment makes it possible for the user to easily view the medical data they desire.

[0075] It goes without saying that the configuration and operation of the browsing support device 10, etc., described in each of the above embodiments are merely examples and can be modified as needed without departing from the spirit of the present invention. It also goes without saying that the above embodiments may be combined as appropriate.

[0076] For example, the system may accumulate and analyze the correspondence between the medical data currently being viewed by the user and the medical data to be viewed next, thereby deriving a recommendation level for viewing candidate medical data 60 based on the medical data currently being viewed.

[0077] Furthermore, although the above embodiment describes a configuration in which the derived model 32 is pre-stored in the storage 22C, the derived model 32 may also be stored in an external device. In this case, the derivation unit 46 of the browsing support device 10 utilizes the derived model 32 stored in the external device via the network N. Alternatively, the control unit 20 of the browsing support device 10 may generate the derived model 32. That is, the browsing support device 10 may function as a learning device that learns the derived model 32.

[0078] Furthermore, in this embodiment, each process is executed on any computer. Alternatively, any computer may execute these processes using a processor as hardware, a program as software, or a combination thereof. In that case, the processor is configured to work in cooperation with the program to execute the various processes in this embodiment, and can function as a unit or means in this embodiment. The execution order of the processes by the processor is not limited to the order described and may be changed as appropriate. Any computer may be a general-purpose computer, a computer designed for a specific purpose, a workstation, or any other system capable of executing each process.

[0079] A processor may consist of one or more hardware components, and the type of hardware is not limited. For example, a processor may consist of programmable logic devices such as a CPU (Central Processing Unit), MPU (Micro Processing Unit), FPGA (Field Programmable Gate Array), dedicated circuits for performing specific processing such as an ASIC (Application Specific Integrated Circuit), a GPU (Graphics Processing Unit), or an NPU (Neural Processing Unit). Furthermore, the type of hardware may be a combination of different types of hardware. When multiple hardware components are configured to execute one or more processes of a processor, these components may reside in physically separate devices or in the same device. Furthermore, in any embodiment, the order of the processes performed by the processor is not limited to the order described above and may be changed as appropriate. The hardware components are composed of electrical circuits (circuits) and the like, which are combinations of circuit elements such as semiconductor elements.

[0080] Furthermore, the program may be firmware or software such as microcode. Alternatively, the program may be, for example, a set of program modules, each function of which may be implemented by a processor configured to perform its respective function. The program may be program code or multiple code segments stored on one or more non-temporary computer-readable media (e.g., storage media or other storage). The program may be divided and stored on multiple non-temporary computer-readable media located on physically separate devices. Program code or code segments may represent any combination of procedures, functions, subprograms, routines, subroutines, modules, software packages, classes, or instructions, data structures, or program statements. Program code or code segments may be connected to other code segments or hardware circuits by sending and receiving information, data, arguments, parameters, or memory contents.

[0081] Furthermore, although the above embodiment describes a configuration in which the browsing support program 30 is pre-stored (installed) in the storage 22C of the memory unit 22, the invention is not limited to this configuration. The browsing support program 30 may be provided in the form of a recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the browsing support program 30 may be provided in the form of a download from an external device via a network.

[0082] Furthermore, the technology disclosed herein extends to all program products. Program products include all forms of products for providing programs. For example, program products include programs provided via networks such as the Internet, and non-temporary computer-readable recording media such as CD-ROMs, DVDs, and USB memory sticks on which programs are stored.

[0083] The following additional information is disclosed regarding the above-described embodiments.

[0084] (Note 1) A browsing support device comprising a processor, wherein the processor acquires status information regarding the circumstances when a user is viewing medical data, and outputs a viewing recommendation score for multiple candidate medical data based on the status information.

[0085] (Note 2) The browsing support device described in Note 1, wherein the aforementioned situation includes at least one of the attendance status and work status of the user viewing the medical data.

[0086] (Note 3) The browsing support device according to Note 1 or Note 2, wherein the status includes the status of the patient corresponding to the medical data, and the status information includes information representing at least one of the patient's diagnosis and treatment stage.

[0087] (Note 4) The browsing support device described in any one of Notes 1 to 3, wherein the situation includes the situation relating to the medical data, and the situation information includes freshness information of the medical data.

[0088] (Note 5) The above situation is a viewing support device as described in any one of Notes 1 to 4, including the status of the device used to view the medical data.

[0089] (Note 6) The viewing support device described in any one of Notes 1 to 5, which includes at least one of the following: medical image data, test result data, and medical document data.

[0090] (Note 7) The recommended viewing level is the viewing support device described in any one of Notes 1 to 7, which is an ordinal scale of the viewing candidates.

[0091] (Note 8) The browsing support device according to any one of Notes 1 to 7, wherein the processor outputs a list of medical data selected from the plurality of browsing candidate medical data according to the browsing recommendation level.

[0092] (Note 9) The browsing support device according to any one of Notes 1 to 8, wherein the processor derives the browsing recommendation level based on the derivation conditions in accordance with the user's instructions.

[0093] (Note 10) The browsing support device according to any one of Notes 1 to 8, wherein the processor outputs a browsing recommendation score for a plurality of medical data specified by the user from among the plurality of medical data of browsing candidates.

[0094] (Note 11) The browsing support device according to any one of Notes 1 to 10, wherein the status includes the user's attendance status or the status of a patient corresponding to the medical data, the status information includes the user's or patient's schedule information and the date and time information when the medical data is viewed, and the processor derives the browsing recommendation level based on the status identified by the status information.

[0095] (Note 12) The browsing support device according to any one of Notes 1 to 11, wherein the processor derives the browsing recommendation score based on the situation and the degree of confidence in the situation.

[0096] (Note 13) The occurrence of the above situation has a periodic pattern, and the browsing support device according to Note 12 derives the confidence level based on the degree of deviation from the periodic pattern.

[0097] (Note 14) The browsing support device according to Note 12, wherein the status information includes schedule information representing the schedule related to the status, and the processor derives the degree of confidence of the status based on the completion rate or number of characters for the entire set of input items of the schedule information.

[0098] (Note 15) The processor is a browsing support device according to any one of Notes 1 to 14, which derives the browsing recommendation score using a trained machine learning model.

[0099] (Note 16) The browsing support device according to any one of Notes 1 to 15, wherein the processor also outputs information indicating whether or not it is the latest medical data when at least a portion of the items included in the medical data of the plurality of browsing candidates are the same.

[0100] (Note 17) The browsing support device according to any one of Notes 1 to 16, wherein the processor also outputs information indicating whether or not each of the plurality of browsing candidates has been read by the user.

[0101] (Note 18) The browsing recommendation level increases as the time required to access the medical data of the browsing candidate increases. (Browse support device as described in any one of Notes 1 to 17)

[0102] (Note 19) The processor is a browsing support device as described in Note 18, which estimates that the access time will be longer the less fresh the medical data of the browsing candidate is.

[0103] (Note 20) The browsing support device according to Note 18, wherein the processor estimates the access time using information about the device that displays the candidate medical data for browsing.

[0104] (Note 21) A browsing support method that obtains status information regarding the user's viewing of medical data and outputs a viewing recommendation score for multiple candidate medical data based on the status information.

[0105] (Note 22) A browsing support program that causes a computer to perform a process to obtain status information regarding the user's viewing of medical data and to output a viewing recommendation score for multiple candidate medical data based on the status information.

[0106] (Note 23) A computer program product that includes a browsing support program that causes the processor to execute a process to obtain status information regarding the circumstances when a user is viewing medical data, and to output a viewing recommendation score for multiple candidate medical data that is derived based on the status information.

[0107] (Note 24) A computer-readable storage medium containing a computer-readable storage medium that stores a browsing support program for which the processor is instructed to obtain status information regarding the circumstances when a user is viewing medical data, and to execute a process that outputs a recommendation level for viewing multiple candidate medical data based on the status information.

[0108] Furthermore, the disclosure of Japanese Patent Application No. 2024-191802, filed on 31 October 2024, is incorporated herein by reference in its entirety. In addition, all documents, patent applications, and technical standards described herein are incorporated herein by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

Claims

1. A browsing support device comprising a processor, wherein the processor acquires status information regarding the user's browsing status of medical data, and outputs a browsing recommendation score for multiple candidate medical data based on the status information.

2. The browsing support device according to claim 1, wherein the situation includes at least one of the attendance status and work status of the user viewing the medical data.

3. The browsing support device according to claim 1, wherein the situation includes the situation of a patient corresponding to the medical data, and the situation information includes information representing at least one of the patient's diagnosis and treatment stage.

4. The browsing support device according to claim 1, wherein the situation includes the situation relating to the medical data, and the situation information includes freshness information of the medical data.

5. The browsing support device according to claim 1, wherein the situation includes the situation of a device for viewing the medical data.

6. The browsing support device according to claim 1, wherein the plurality of candidate medical data for viewing include at least one of medical image data, test result data, and medical document data.

7. The browsing support device according to claim 1, wherein the browsing recommendation level is an ordinal scale of the browsing candidates.

8. The browsing support device according to claim 1, wherein the processor outputs a list of medical data selected from the plurality of browsing candidate medical data according to the browsing recommendation level.

9. The browsing support device according to claim 1, wherein the processor derives the browsing recommendation level based on derivation conditions corresponding to the user's instructions.

10. The browsing support device according to claim 1, wherein the processor outputs a browsing recommendation score for a plurality of medical data specified by the user from among the plurality of candidate medical data for browsing.

11. The browsing support device according to claim 1, wherein the situation includes the user's attendance status or the patient's status corresponding to the medical data, the situation information includes the user's or patient's schedule information and the date and time information when the medical data is viewed, and the processor derives the browsing recommendation level based on the situation identified by the situation information.

12. The browsing support device according to claim 1, wherein the processor derives the browsing recommendation score based on the situation and the degree of confidence in the situation.

13. The browsing support device according to claim 12, wherein the occurrence of the aforementioned situation has a periodic pattern, and the processor derives the confidence level based on the degree of deviation from the periodic pattern.

14. The browsing support device according to claim 12, wherein the status information includes schedule information representing a schedule related to the status, and the processor derives a degree of confidence in the status based on the completion rate or number of characters for all input items of the schedule information.

15. The browsing support device according to claim 1, wherein the processor derives the browsing recommendation score using a trained machine learning model.

16. The browsing support device according to claim 1, wherein the processor also outputs information indicating whether or not the data is the latest medical data when at least a portion of the items included in the medical data of the plurality of browsing candidates are the same.

17. The browsing support device according to claim 1, wherein the processor also outputs information indicating whether or not each of the plurality of browsing candidates has been read by the user.

18. The browsing support device according to claim 1, wherein the browsing recommendation level increases as the time required to access the medical data of the browsing candidate increases.

19. The browsing support device according to claim 18, wherein the processor estimates that the access time is longer the less fresh the medical data of the browsing candidate is.

20. The browsing support device according to claim 18, wherein the processor estimates the access time using information about a device that displays the candidate medical data for browsing.

21. A browsing support method that obtains status information regarding the user's viewing of medical data and outputs a viewing recommendation score for multiple candidate medical data based on the status information.

22. A browsing support program that causes a computer to perform a process of obtaining status information regarding the user's viewing of medical data and outputting a viewing recommendation score for multiple candidate medical data based on the status information.

Images