Diagnosis assistance system, diagnosis assistance device and diagnosis assistance method

Abstract

The embodiment relates to a diagnosis support system, a diagnosis support device and a diagnosis support method. It makes it easy to grasp medical information about a patient. The diagnosis support system of the embodiment comprises a storage unit, an extraction unit, a calculation unit and a display control unit. The storage unit stores medical information, which includes patient information related to the patient's condition and intervention information related to the intervention on the patient. The extraction unit extracts the change point of the medical information. The calculation unit calculates a first change amount representing the change of the patient information before and after the change point and a second change amount representing the change of the intervention information before and after the change point. The display control unit performs a display based on the first change amount and the second change amount in a display mode set corresponding to at least one of the first change amount and the second change amount.

Description

[0001] References to related applications

[0002] This application enjoys the benefit of priority of Japanese patent application No. 2019-182726 filed on October 3, 2019, and the entire contents of this Japanese patent application are cited in this application. Technical Field

[0003] The embodiments relate to a diagnosis support system, a diagnosis support device, and a diagnosis support method. Background Art

[0004] Doctors make various decisions based on patients' past examination information, treatment information, and other various medical information related to patients. However, depending on the patient, a large amount of medical information may be recorded, and it is not easy to grasp all of it. Summary of the invention

[0005] An object of the present invention is to make it easy to grasp medical information about a patient.

[0006] The diagnostic support system of the embodiment comprises a storage unit, an extraction unit, a calculation unit and a display control unit. The storage unit stores medical information, which includes patient information related to the patient's condition and intervention information related to the intervention on the patient. The extraction unit extracts the change point of the medical information. The calculation unit calculates a first change amount indicating the change of the patient information before and after the change point and a second change amount indicating the change of the intervention information before and after the change point. The display control unit performs display based on the first change amount and the second change amount in a display mode set corresponding to at least one of the first change amount and the second change amount.

[0007] According to the diagnosis support system of the embodiment, medical information about a patient can be easily grasped. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 This is a block diagram showing an example of the configuration of the diagnosis support system according to the first embodiment.

[0009] Figure 2 It is a diagram showing an overview of the processing of the diagnosis support device according to the first embodiment.

[0010] Figure 3 This is a diagram showing an example of medical information according to the first embodiment.

[0011] Figure 4 It is a diagram showing an example of the extraction process of the change point according to the first embodiment.

[0012] Figure 5It is a diagram showing an example of the extraction process of the change point according to the first embodiment.

[0013] Figure 6 This is a diagram showing an example of the extraction process of the first period and the second period according to the first embodiment.

[0014] Figure 7 This is a diagram showing an example of the extraction process of the first period and the second period according to the first embodiment.

[0015] Figure 8 This is a diagram showing an example of the calculation process of the first change amount according to the first embodiment.

[0016] Fig. 9 This is a diagram showing an example of the calculation process of the second change amount according to the first embodiment.

[0017] Fig. 10A This is a diagram showing an example of the display mode setting process according to the first embodiment.

[0018] Fig. 10B This is a diagram showing an example of the display mode setting process according to the first embodiment.

[0019] Fig.11A It is a diagram showing a display example of the first embodiment.

[0020] Fig. 11B It is a diagram showing a display example of the first embodiment.

[0021] Fig. 11C It is a diagram showing a display example of the first embodiment.

[0022] Fig.11D It is a diagram showing a display example of the first embodiment.

[0023] Fig.12 This is a flowchart for explaining a series of processes of the diagnosis support system according to the first embodiment.

[0024] Fig.13 It is a diagram showing a display example of the second embodiment. DETAILED DESCRIPTION

[0025] Hereinafter, embodiments of the diagnosis support system, the diagnosis support device, and the diagnosis support method will be described in detail with reference to the accompanying drawings. In addition, the diagnosis support system, the diagnosis support device, and the diagnosis support method of the present application are not limited to the embodiments described below.

[0026] First, a first embodiment will be described. In the first embodiment, a diagnosis support system 1 including a diagnosis support device 10 and a database 20 will be described.

[0027] Figure 1 1 is a block diagram showing an example of the structure of the diagnosis support system 1 according to the first embodiment. Figure 1 As shown in FIG. 1 , the diagnosis support system 1 of the first embodiment includes a diagnosis support device 10 and a database 20. Figure 1 As shown, the diagnosis support device 10 and the database 20 are connected to each other via a network NW.

[0028] The database 20 is a storage device that stores medical information about patients. The database 20 is an example of a storage unit (memory).

[0029] For example, the database 20 is an electronic medical record storage device installed as part of an electronic medical record system introduced in a hospital or the like. For another example, the database 20 is an integrated management device that obtains various medical information from various medical information systems such as an electronic medical record system, a PACS (Picture Archiving and Communication System), a radiology department system, and a physical examination system, and manages the obtained medical information as a platform. In addition, Figure 1 Although one database 20 is shown in FIG. 1 , the database 20 may be composed of a combination of physically separated multiple storage devices. Furthermore, these multiple storage devices may be installed in different facilities.

[0030] Here, the medical information about the patient includes, for example, patient information about the patient's condition and intervention information about the patient's intervention. For example, the database 20 stores various medical information in association with information on the date and time when the medical information was acquired and the patient ID.

[0031] For example, patient information includes examination information collected through examination. Specifically, examples of examination information include vital data measured for patients (e.g., pulse rate, heart rate, respiratory rate, blood pressure, body temperature, transcutaneous arterial oxygen saturation (SpO2), etc.), examination data of specimens (blood, etc.) collected from patients, medical image data collected from patients, and measurement data of medical image data (blood vessel diameter, blood flow, etc.). In addition, patient information also includes various information related to the patient's condition, such as the patient's food intake, urine output, and symptoms reported by the patient (pain, itching, etc.).

[0032] In addition, the intervention information includes, for example, treatment information related to the treatment performed on the patient. Specifically, as examples of treatment information, information on medication administered to the patient (such as the type and amount of the administered medicine), surgery, radiotherapy, massage and other treatment measures can be cited. In addition, the intervention information also includes various information related to the intervention of the patient. For example, the database 20 stores the event of changing the plaster to a bandage as the patient recovers as intervention information, and associates it with the information of the changed date and time and the patient ID. For another example, the database 20 stores the event of sticking a wet cloth on the part where the patient complained of pain as intervention information, and associates it with the information of the changed date and time and the patient ID.

[0033] In addition, medical information is not limited to information recorded in hospitals, but may also be information recorded in medical facilities such as clinics, nursing facilities, and examination facilities. Alternatively, medical information may also be information recorded outside medical facilities such as the patient's home. For example, medical information may also be information recorded by a wearable device worn by a patient at any location.

[0034] The diagnosis support device is a device that supports the user's diagnosis by displaying various information based on the medical information obtained from the database 20. For example, the diagnosis support device 10 is Figure 1 As shown, there are an input interface 101 , a display 102 , a storage circuit 103 and a processing circuit 104 .

[0035] The input interface 101 receives various input operations from the user, converts the received input operations into electrical signals, and then outputs them to the processing circuit 104. For example, the input interface 101 is realized by a mouse or keyboard, a trackball, a switch, a button, a joystick, a touch panel for input operation by touching the operation surface, a touch screen integrating a display screen and a touch panel, a non-contact input circuit using an optical sensor, a sound input circuit, etc. In addition, the input interface 101 can also be composed of a tablet terminal that can wirelessly communicate with the main body of the diagnosis support device 10. In addition, the input interface 101 is not limited to a structure with physical operation components such as a mouse or a keyboard. For example, a processing circuit for receiving an electrical signal corresponding to an input operation from an external input device that is separately provided from the diagnosis support device 10 and outputting the electrical signal to the processing circuit 104 is also included in the example of the input interface 101.

[0036] The display 102 displays various information. For example, the display 102 displays a GUI (Graphical User Interface) for accepting various instructions, various settings, etc. from the user via the input interface 101. In addition, the display 102 performs a display based on the first change amount and the second change amount described later. For example, the display 102 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 102 can be a desktop type, or can be composed of a tablet terminal that can wirelessly communicate with the main body of the diagnosis support device 10.

[0037] The storage circuit 103 is implemented, for example, by a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc. For example, the storage circuit 103 stores a program for causing the circuit included in the diagnosis support device 10 to implement its function. Alternatively, the storage circuit 103 may be implemented by a server group (cloud) connected to the diagnosis support device 10 via the network NW.

[0038] The processing circuit 104 controls the overall operation of the diagnosis support device 10 by executing the extraction function 104a, the calculation function 104b, the display control function 104c, and the control function 104d. Here, the extraction function 104a is an example of an extraction unit. In addition, the calculation function 104b is an example of a calculation unit. In addition, the display control function 104c is an example of a display control unit.

[0039] For example, the processing circuit 104 reads out and executes a program corresponding to the extraction function 104a from the storage circuit 103, thereby obtaining medical information from the database 20 via the network NW and extracting a change point of the obtained medical information. In addition, for example, the processing circuit 104 reads out and executes a program corresponding to the calculation function 104b from the storage circuit 103, thereby calculating a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point. In addition, for example, the processing circuit 104 reads out and executes a program corresponding to the display control function 104c from the storage circuit 103, thereby performing a display based on the first change amount and the second change amount on the display 102. In addition, for example, the processing circuit 104 reads out and executes a program corresponding to the control function 104d from the storage circuit 103, thereby controlling the overall operation of the diagnosis support device 10 based on an operation received from a user via the input interface 101. In addition, the processing of each function of the processing circuit 104 will be described in detail later.

[0040] exist Figure 1In the diagnosis support device 10 shown, each processing function is stored in the form of a program executable by a computer in the storage circuit 103. The processing circuit 104 is a processor that realizes the function corresponding to each program by reading and executing the program from the storage circuit 103. In other words, the processing circuit 104 in the state of reading each program has the function corresponding to the read program.

[0041] In addition, Figure 1 In the embodiment, it is described that the extraction function 104a, the calculation function 104b, the display control function 104c and the control function 104d are realized by a single processing circuit 104, but a plurality of independent processors may be combined to form the processing circuit 104, and each processor executes a program to realize the function. In addition, each processing function of the processing circuit 104 may be appropriately dispersed or unified into a single or multiple processing circuits for realization.

[0042] The diagnosis support device 10 and the database 20 may be installed in any location as long as they can be connected via the network NW. For example, the diagnosis support device 10 may be installed in a hospital different from the database 20. That is, the network NW may be composed of a closed local area network within the hospital or a network via the Internet.

[0043] The term "processor" used in the above description refers to, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an Application Specific Integrated Circuit (ASIC), a programmable logic device (for example, a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor realizes its functions by reading out and executing the program stored in the storage circuit 103.

[0044] in addition, Figure 1In the embodiment, a case where a single storage circuit 103 stores programs corresponding to each processing function is described. However, the embodiment is not limited to this. For example, a plurality of storage circuits 103 may be dispersedly configured, and the processing circuit 104 reads the corresponding program from a single storage circuit 103. In addition, instead of storing the program in the storage circuit 103, the program may be directly implanted in the circuit of the processor. In this case, the processor realizes the function by reading and executing the program implanted in the circuit.

[0045] In addition, the processing circuit 104 may also use a processor of an external device connected via the network NW to realize functions. For example, the processing circuit 104 reads and executes a program corresponding to each function from the storage circuit 103, and uses a server group (cloud) connected to the diagnosis support device 10 via the network NW as a computing resource, thereby realizing Figure 1 The functions shown.

[0046] The above describes the diagnosis support system 1 including the diagnosis support device 10 and the database 20. Based on this structure, the diagnosis support device 10 in the diagnosis support system 1 makes it easy to grasp the medical information about the patient through the processing of the processing circuit 104. In addition, the following describes a process for the user to grasp the medical information of the patient P1 as an example.

[0047] First, use Figure 2 , describes the outline of the processing of the diagnosis support device 10. For example, the diagnosis support device 10 first obtains medical information about the patient P1 from the database 20. For example, based on the patient ID of the patient P1, the diagnosis support device 10 obtains medical information including patient information and intervention information about the patient P1. In addition, Figure 2 It is a diagram showing an outline of the processing of the diagnosis support device 10 according to the first embodiment.

[0048] Next, the diagnosis supporting device 10 extracts the change point of the medical information. Next, the diagnosis supporting device 10 calculates a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point. Next, the diagnosis supporting device 10 sets a display mode corresponding to at least one of the first change amount and the second change amount. And, the diagnosis supporting device 10 performs display based on the first change amount and the second change amount in the set display mode.

[0049] Next, the processing of the diagnosis support device 10 will be described in more detail. Figure 3 , explaining the medical information obtained from the database 20. Figure 3 is a diagram showing an example of medical information according to the first embodiment. For example, the extraction function 104a obtains Figure 3For example, the extraction function 104a uses the fact that the user receives an instruction to perform a diagnosis on the patient P1 as a trigger to obtain the intervention information and patient information from the database 20. Figure 3 Intervention information and patient information shown.

[0050] Figure 3 The intervention information shown is the record of diuretics and cardiotonic drugs administered on each day from "2019.5.17" to "2019.5.24". In other words, Figure 3 The intervention information shown is time series information related to the administration of diuretics and cardiotonic drugs.

[0051] Specifically, Figure 3 In "2019.5.17", it was stated that diuretics and cardiotonic drugs were not administered. In addition, Figure 3 In "2019.5.18", it is indicated that diuretics were administered "10", and cardiotonic drugs were not administered. In addition, Figure 3 In "2019.5.19", it means that the diuretic was administered "10" and the cardiotonic was administered "20". In addition, Figure 3 In "2019.5.20", it is indicated that diuretics and cardiotonic drugs were not administered. In addition, Figure 3 In "2019.5.21", it is indicated that diuretics were administered "10" and cardiotonic drugs were not administered. In addition, Figure 3 "2019.5.22", "2019.5.23" and "2019.5.24" indicate that diuretics and cardiotonic drugs were not administered.

[0052] also, Figure 3 The patient information shown is a record of NT-proBNP and the number of breaths per unit time for each day from "2019.5.17" to "2019.5.24". In other words, Figure 3 The patient information shown is time series information about NT-proBNP and respiratory rate.

[0053] Specifically, Figure 3 In "2019.5.17", it is indicated that NT-proBNP is "192" and the respiratory rate is "8". In addition, Figure 3 In "2019.5.18", it is indicated that NT-proBNP is "185" and the respiratory rate is "12". In addition, Figure 3 In "2019.5.19", it is indicated that NT-proBNP is "200" and the respiratory rate is "14". In addition, Figure 3 In "2019.5.20", it is indicated that NT-proBNP is "180" and the respiratory rate is "13". In addition, Figure 3In "2019.5.21", it is indicated that NT-proBNP is "92" and respiratory rate is "23". In addition, Figure 3 In "2019.5.22", it means that NT-proBNP is "100" and the respiratory rate is "21". In addition, Figure 3 In "2019.5.23", it is indicated that NT-proBNP is "89" and respiratory rate is "20". In addition, Figure 3 In "2019.5.24", it is indicated that NT-proBNP is "102" and the respiratory rate is "19".

[0054] Next, the extraction function 104a extracts the change points of the medical information. For example, the extraction function 104a is for Figure 3 The time series information shown sets a plurality of time points, compares the characteristics of the medical information before and after each time point, and extracts the time point with a large degree of change as a change point.

[0055] For example, the extraction function 104a may be Figure 4 As shown, the three days "2019.5.17 to 2019.5.19" or the period xt and the three days "2019.5.20 to 2019.5.22" or the period xt+1 are set. Figure 4 It is a diagram showing an example of the extraction process of the change point according to the first embodiment.

[0056] Next, the extraction function 104a is as follows Figure 5 As shown, the degree of change is evaluated by comparing the feature vector representing the medical information of period xt with the feature vector representing the medical information of period xt+1. For example, the extraction function 104a can perform subtraction using the feature vector representing the medical information of period xt and the feature vector representing the medical information of period xt+1, and use the magnitude (norm) of the calculated vector as the degree of change to evaluate. In addition, Figure 5 FIG. 1 is a diagram showing an example of the extraction process of the change point in the first embodiment. Figure 5 , but each eigenvector is a four-dimensional vector having a dimension corresponding to the diuretic, the cardiotonic, the NT-proBNP, and the respiratory rate.

[0057] Furthermore, the extraction function 104a is as follows Figure 4As shown, the degree of change is evaluated between the three days of "2019.5.18-2019.5.20" or the period xu and the three days of "2019.5.21-2019.5.23" or the period xu+1. Similarly, the extraction function 104a evaluates the degree of change between the three days of "2019.5.19-2019.5.21" and the three days of "2019.5.22-2019.5.24". That is, the extraction function 104a evaluates the degree of change at each moment while shifting the moment of the object. And the extraction function 104a extracts the moment with a large degree of change as a change point. For example, the extraction function 104a extracts the moment with a degree of change greater than a threshold as a change point.

[0058] In addition, Figure 4 , a case where a 3-day period (period xt, period xt+1, period xu, period xu+1, etc.) is set when evaluating the degree of change is described, but the length of the set period is not particularly limited. In addition, the extraction function 104a may also evaluate the degree of change while changing the length of the set period.

[0059] For example, the extraction function 104a evaluates the degree of change between the three days of "May 17, 2019 to May 19, 2019" or the period xt and the three days of "May 20, 2019 to May 22, 2019" or the period xt+1. In addition, the extraction function 104a evaluates the degree of change between the two days of "May 18, 2019 to May 19, 2019" and the two days of "May 20, 2019 to May 21, 2019". In addition, the extraction function 104a evaluates the degree of change between the one day of "May 19, 2019" and the one day of "May 20, 2019". And, based on these three evaluations, the extraction function 104a determines whether to extract the time between "May 19, 2019" and "May 20, 2019" as a change point. For example, the extraction function 104a calculates the average value for these three evaluations, and determines whether to extract the time between "2019.5.19" and "2019.5.20" as a change point based on whether the average value exceeds the threshold. In addition, the extraction function 104a also similarly determines whether to extract the other time points between "2019.5.17 to 2019.5.24" as change points. In this way, by changing the length of the set period while evaluating the degree of change, the extraction function 104a can extract both instantaneous changes and smooth changes as change points.

[0060] Alternatively, the extraction function 104a may also extract the change points using the subspace method. For example, the extraction function 104a first obtains a plurality of feature vectors from the period xt. Specifically, the extraction function 104a obtains from the period xt a plurality of feature vectors including a feature vector based on the three days of "2019.5.17 to 2019.5.19", a feature vector based on the two days of "2019.5.17 to 2019.5.18", a feature vector based on the two days of "2019.5.18 to 2019.5.19", a feature vector based on the one day of "2019.5.17", a feature vector based on the one day of "2019.5.18", and a feature vector based on the one day of "2019.5.19". In addition, the extraction function 104a calculates a subspace based on these plurality of feature vectors.

[0061] Similarly, the extraction function 104a obtains multiple feature vectors from the period xt+1, and calculates the subspace based on the obtained multiple feature vectors. In addition, the extraction function 104a compares the subspace based on the period xt and the subspace based on the period xt+1 to determine whether to extract the time between "2019.5.19" and "2019.5.20" as a change point. In addition, the extraction function 104a also determines whether to extract the other time points in "2019.5.17 to 2019.5.24" as change points.

[0062] Alternatively, the extraction function 104a may also extract change points using density ratios. For example, the extraction function 104a compares the distribution of medical information during period xt with the distribution of medical information during period xt+1. For example, the extraction function 104a compares the statistical values ​​regarding the offset or shape of the distribution between the distribution of medical information during period xt and the distribution of medical information during period xt+1. In addition, the statistical value regarding the offset of the distribution is, for example, the average value of each distribution. In addition, the statistical value regarding the shape of the distribution is, for example, the variance value of each distribution. And, the extraction function 104a determines whether to extract the moment between "2019.5.19" and "2019.5.20" as a change point based on the comparison results between the distributions. In addition, the extraction function 104a also determines whether to extract the other moments in "2019.5.17 to 2019.5.24" as change points.

[0063] In addition, the extraction function 104a may also extract the change points by machine learning. For example, the extraction function 104a extracts the change points using a learned model M1 having a function of extracting the change points of medical information. In addition, the learned model M1 is generated in advance by the extraction function 104a and stored in the storage circuit 103.

[0064] For example, the extraction function 104a first collects learning data used in generating the learned model M1. For example, the extraction function 104a obtains pairs of medical information and change points as learning data. In addition, the medical information used as learning data may be medical information about the patient P1 or medical information about other patients.

[0065] In addition, the change point used as learning data can be set by a user such as a physician. Alternatively, the extraction function 104a can also automatically set the change point used as learning data. For example, the extraction function 104a can set the time that the user pays attention to when referring to medical information as the change point. For example, the extraction function 104a can set the time corresponding to the medical information displayed for a long time and the time corresponding to the medical information specified by the user to display detailed information as the change point.

[0066] Next, the extraction function 104a performs machine learning based on the collected learning data to generate a learned model M1. Here, the learned model M1 can be composed of a neural network, for example. A neural network is a network with a structure in which adjacent layers arranged in layers are coupled and information is transmitted from the input layer side to the output layer side. For example, the extraction function 104a generates the learned model M1 by performing deep learning on a multi-layer neural network using the above-mentioned learning data. In addition, a multi-layer neural network includes, for example, an input layer, a plurality of intermediate layers (hidden layers), and an output layer.

[0067] For example, the extraction function 104a inputs medical information about a patient as input side data to the neural network. In this case, in the neural network, information is transmitted while coupling only between adjacent layers from the input layer side to the output layer side, and the estimated result of the change point of the input medical information is output from the output layer. In addition, a neural network that transmits information unidirectionally from the input layer side to the output layer side is also called a convolutional neural network (CNN). In addition, although CNN is used as an example, the extraction function 104a can also use a neural network of a different type from CNN. For example, the extraction function 104a can also form a learned model M1 through neural networks such as a fully coupled neural network and a recurrent neural network (RNN: Recurrent Neural Network).

[0068] The extraction function 104a adjusts the parameters of the neural network so that the neural network can output a preferred result when the input side data is input, thereby generating a learned model M1. For example, the extraction function 104a repeats the process while adjusting the parameters of the neural network until the difference between the estimated result of the change point output from the neural network and the change point input as the learning data of the output side is lower than the threshold value. Thus, the extraction function 104a generates a learned model M1 that is endowed with the function of accepting the input of medical information and extracting the change point of the input medical information. In addition, the extraction function 104a stores the generated learned model M1 in the storage circuit 103.

[0069] Furthermore, the extraction function 104a inputs the medical information about the patient P1 to the learned model M1. For example, the extraction function 104a uses the case where the user receives an instruction indicating that a diagnosis about the patient P1 is to be made as a trigger, obtains the medical information from the database 20, reads the learned model M1 from the storage circuit 103, and inputs the medical information to the learned model M1. In addition, the learned model M1 outputs the change point of the input medical information. That is, the extraction function 104a can extract the change point of the medical information about the patient P1 using the learned model M1.

[0070] In addition, the case where the input side data of the learned model M1 is medical information is described, but the embodiment is not limited to this. For example, the extraction function 104a can also generate the learned model M1 by using the subspace or density ratio calculated based on the medical information as input side data instead of the medical information. In this case, the extraction function 104a calculates the subspace or density ratio from the medical information about the patient P1, and inputs the calculated subspace or density ratio into the learned model M1, thereby being able to extract the change point of the medical information about the patient P1.

[0071] In addition, although the learned model M1 is constituted by a neural network, the extraction function 104a may generate the learned model M1 by a machine learning method other than a neural network. In addition, although the extraction function 104a generates the learned model M1, the learned model M1 may be generated by another device different from the diagnosis support device 10.

[0072] Next, the extraction function 104a extracts the first period before the change point and the second period after the change point. For example, the extraction function 104a extracts the periods of preset length before and after the change point as the first period and the second period. The following is an example in which "3 days" is preset as the length of the period. For example, Figure 6As shown, when the change point V1 between "2019.5.19" and "2019.5.20" is extracted as the change point, the extraction function 104a extracts the period R11, which is 3 days before the change point V1, as the first period, and extracts the period R12, which is 3 days after the change point V1, as the second period. Figure 6 This is a diagram showing an example of the extraction process of the first period and the second period according to the first embodiment.

[0073] Alternatively, the extraction function 104a may extract the periods between the extracted multiple change points as the first period and the second period. For example, when a change point V1 between "2019.5.19" and "2019.5.20", a change point V2 between "2019.5.16" and "2019.5.17", and a change point V3 between "2019.5.23" and "2019.5.24" are extracted as change points, the extraction function 104a extracts the change point V1 as follows: Figure 7 The extraction function 104a extracts the period R21 between the change point V1 and the change point V2 as the first period before the change point V1. In addition, the extraction function 104a extracts the period R22 between the change point V1 and the change point V3 as the second period after the change point V1. Figure 7 1 is a diagram showing an example of the extraction process of the first period and the second period in the first embodiment. The extraction function 104a may set an upper limit value for the length of the period, and extract the first period and the second period based on the upper limit value when the period between the change points is long.

[0074] Next, the calculation function 104b calculates a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point. For example, the calculation function 104b calculates the first change amount based on the patient information of the first period and the patient information of the second period, and calculates the second change amount based on the intervention information of the first period and the intervention information of the second period. Figure 6 The period R11 and the period R12 shown in FIG. 1 are described as the case where the periods are before and after the change point V1.

[0075] First, regarding the calculation of the first change, use Figure 8 Provide explanation. Figure 8 FIG. 4 is a diagram showing an example of a calculation process of the first variation according to the first embodiment. Figure 8 As shown, the calculation function 104b calculates the condition of the period R11 and the condition of the period R12, respectively, and calculates the difference between the condition of the period R11 and the condition of the period R12 as the first variation.

[0076] Specifically, the calculation function 104b calculates the condition of the period R11 based on the patient information of the period R11. That is, the calculation function 104b calculates the condition of the period R11 based on the NT-proBNP and respiratory rate of the patient P1 from "2019.5.17 to 2019.5.19". In addition, the calculation function 104b can also calculate a numerical value representing the condition as the condition of the period R11, and can also allocate it to multiple ranks representing the condition. In addition, the setting method of these numerical values ​​and ranks representing the condition is not particularly limited. For example, it can be set by the user, it can be set for each hospital, or it can be set according to generally used guidelines.

[0077] For example, the calculation function 104b calculates the condition of the period R11 according to a predetermined table in which the values ​​of NT-proBNP and respiratory rate are associated with the condition of the period R11. Also, for example, the calculation function 104b calculates the condition of the period R11 according to a predetermined table in which the combination of NT-proBNP and respiratory rate is associated with the condition of the period R11.

[0078] In addition, for example, the calculation function 104b may calculate the condition of the period R11 according to a predetermined formula. For example, the calculation function 104b calculates the variance value of NT-proBNP and respiratory rate in the period R11 as the condition of the period R11.

[0079] Alternatively, the calculation function 104b may calculate the condition of the period R11 by a machine learning method. For example, the calculation function 104b calculates the condition of the period R11 using a learned model M2 having a function of accepting input of patient information and outputting the condition of the patient. The learned model M2 is generated in advance by the calculation function 104b, for example, and stored in the storage circuit 103.

[0080] For example, the calculation function 104b first collects learning data used in the generation of the learned model M2. For example, the calculation function 104b obtains a pair of patient information and a disease condition as learning data. In addition, the patient information used as learning data may be patient information about the patient P1 or patient information about other patients. In addition, the disease condition used as learning data can be set by a user such as a physician, for example. For example, the user can set the disease condition based on the result of a medical consultation or the like conducted at the time when the paired patient information is obtained.

[0081] Next, the computing function 104b performs machine learning based on the collected learning data to generate a learned model M2. Here, the learned model M2 can be composed of a neural network, for example. For example, the computing function 104b generates the learned model M2 by performing deep learning on a multi-layer neural network using the above-mentioned learning data. In addition, the computing function 104b stores the generated learned model M2 in the storage circuit 103.

[0082] The calculation function 104b inputs the patient information about the patient P1 to the learned model M2 read from the storage circuit 103. The learned model M2 outputs the condition of the period R11 based on the input patient information. That is, the calculation function 104b can calculate the condition of the period R11 using the learned model M2.

[0083] In addition, although the case where the learned model M2 is composed of a neural network is described, the computing function 104b may generate the learned model M2 by a machine learning method other than a neural network. In addition, although the case where the computing function 104b generates the learned model M2 is described, the learned model M2 may be generated in another device different from the diagnosis support device 10.

[0084] As described above, the calculation function 104b calculates the condition of the period R11 based on the patient information of the period R11. Similarly, the calculation function 104b calculates the condition of the period R12 based on the patient information of the period R12. Furthermore, the calculation function 104b calculates the first change amount based on the condition of the period R11 and the condition of the period R12. For example, in the case where a numerical value is calculated as the condition, the calculation function 104b calculates the difference in numerical value between the condition of the period R11 and the condition of the period R12 as the first change amount. In addition, for example, in the case where a level is calculated as the condition, the calculation function 104b calculates the difference in level number between the condition of the period R11 and the condition of the period R12 as the first change amount.

[0085] In addition, the calculation function 104b may also standardize the first change amount. For example, the calculation function 104b may standardize the first change amount calculated in an arbitrary numerical range or the first change amount calculated as a level to a numerical value of "0 to 1". Alternatively, the calculation function 104b may also standardize the first change amount so that the sum of the first change amount and the second change amount described later becomes "1". In addition, hereinafter, the standardized first change amount is also recorded as the disease condition change degree.

[0086] Next, regarding the calculation of the second variation, we use Fig. 9 Provide explanation. Fig. 9 FIG. 4 is a diagram showing an example of a calculation process of the second variation according to the first embodiment. Fig. 9As shown, the calculation function 104 b converts the intervention information of the period R11 and the intervention information of the period R12 into vectors, compares the vectors, and calculates the second variation.

[0087] For example, the calculation function 104b can vectorize the intervention information of the period R11 and the period R12 using the representative value of each period (e.g., the maximum value, the value near the change point, etc.). Specifically, the intervention information of the period R11 can be represented by a vector of (diuretic, cardiotonic) = (10, 20). In addition, the intervention information of the period R12 can be represented by a vector of (diuretic, cardiotonic) = (10, 0). And the calculation function 104b calculates the second change amount by comparing these two vectors. For example, the calculation function 104b can calculate the second change amount by comparing these two vectors. Fig. 9 As shown, the second variation is calculated by subtracting these two vectors.

[0088] In addition, the calculation function 104b can also vectorize the intervention information of the period R11 and the period R12 using other statistical values ​​(such as average values) of each period. In addition, when the presence or absence of the implementation of the operation or the like is used as the intervention information, the calculation function 104b can, for example, numerically convert "presence" into "1" and "absence" into "0", thereby Fig. 9 The second variation is calculated in the same manner as in the case shown. Fig. 9 For the sake of convenience, a two-dimensional vector is described in detail, but the dimension of the vector is not particularly limited.

[0089] In addition, the calculation function 104b may also standardize the second variation. For example, the calculation function 104b may also standardize the second variation calculated in an arbitrary numerical range to a value of "0 to 1". Alternatively, the calculation function 104b may also standardize the second variation so that the sum of the first variation and the second variation becomes "1". In addition, the standardized second variation is also recorded as the intervention variation degree below.

[0090] Next, the display control function 104c sets the display mode in accordance with at least one of the first change amount and the second change amount. For example, the display control function 104c sets the display mode in accordance with the ratio of the first change amount to the second change amount. Fig. 10A and Fig. 10B Provide explanation. Fig. 10A and Fig. 10B This is a diagram showing an example of the display mode setting process according to the first embodiment.

[0091] First, the calculation function 104b is as follows Fig. 10AAs shown, the first change amount and the second change amount are normalized so that the sum of the first change amount and the second change amount becomes "1". Fig. 10A This indicates that the first normalized change amount, that is, the degree of change in the condition, is "0.6", and the second normalized change amount, that is, the degree of change in the intervention, is "0.4".

[0092] Next, the display control function 104c sets a mixed color obtained by mixing the two colors at a ratio corresponding to the ratio of the first change amount to the second change amount. Fig. 10B As shown in FIG. 1 , a mixed color obtained by mixing red and blue is set so that "red: blue = 6: 4" corresponding to the ratio of the disease state change degree "0.6" and the intervention change degree "0.4". That is, the display control function 104c sets the mixed color to be closer to red when the disease state changes more than the intervention change, sets the mixed color to be closer to blue when the intervention change more than the disease state change, and sets the mixed color to be an intermediate color (purple) when both the disease state and the intervention change to the same degree.

[0093] The display control function 104c then displays the first and second changes using the set mixed color. For example, the display control function 104c displays the time series of at least one of the patient information and the intervention information, and displays the change points using the set mixed color.

[0094] For example, the display control function 104c is as follows: Fig.11A As shown, NT-proBNP and respiratory rate are displayed in time series in the graph, and the change points are shown as bar charts. Fig.11A The bar lines representing the change points in the graph are displayed in a mixed color set according to the ratio of the first change amount to the second change amount. More specifically, Fig.11A In the case where the change in the condition is greater than the change in the intervention, the change point is illustrated using a mixed color set to be closer to red. Fig.11A It is a diagram showing a display example of the first embodiment.

[0095] Reference Fig.11AThe user of the display can understand at a glance that there is a change point between "2019.5.20" and "2019.5.21", and that the change in the condition is greater than the change in the intervention. That is, the user can easily understand that the condition of the patient P1 has changed significantly due to the intervention performed between "2019.5.20" and "2019.5.21". In other words, the user can easily understand that the intervention performed between "2019.5.20" and "2019.5.21" has a greater impact on the condition of the patient P1.

[0096] In addition, for example, the display control function 104c may be Fig. 11B As shown in FIG. 1 , NT-proBNP and respiratory rate are displayed in time series in a graph, and the change points are shown as bar graphs. Fig. 11B The bar lines representing the change points in the graph are displayed in a mixed color set according to the ratio of the first change amount to the second change amount. More specifically, Fig. 11B In the case where the change in the condition is small relative to the change in the intervention, the change point is illustrated using a mixed color set to be closer to blue. Fig. 11B It is a diagram showing a display example of the first embodiment.

[0097] Reference Fig. 11B The user of the display can easily understand at a glance that there is a change point between "2019.5.20" and "2019.5.21", and that the change in the condition is small compared to the change in the intervention. In other words, the user can easily understand that the condition of patient P1 has not changed much due to the intervention performed between "2019.5.20" and "2019.5.21". In other words, the user can easily understand that the intervention performed between "2019.5.20" and "2019.5.21" has little effect on the condition of patient P1.

[0098] In addition, for example, the display control function 104c may be Fig. 11C As shown, NT-proBNP and respiratory rate are displayed in time series in the graph, and the change points are shown as bar charts. Fig. 11C The bar lines representing the change points in the graph are displayed in a mixed color set according to the ratio of the first change amount to the second change amount. More specifically, Fig. 11C In the case where the condition and intervention both change to the same extent, the change point is illustrated using a mixed color set to an intermediate color (purple). Fig. 11C It is a diagram showing a display example of the first embodiment.

[0099] Reference Fig. 11CThe user of the display can understand at a glance that there is a change point between "2019.5.20" and "2019.5.21", and that the change in the condition is the same as the change in the intervention. That is, the user can easily understand that the condition of the patient P1 has changed to the normally expected degree relative to the magnitude of the intervention due to the intervention performed between "2019.5.20" and "2019.5.21".

[0100] In addition, Fig.11A , Fig. 11B as well as Fig. 11C In the example of time series display, the case where NT-proBNP and respiratory rate are displayed in a graph is shown. However, the embodiment is not limited to this. For example, the display control function 104c may display patient information other than NT-proBNP and respiratory rate in a graph, or may display intervention information in a graph.

[0101] In addition, the display control function 104c can also be Fig.11D As shown, a time series display is performed using a table. In this case, the display control function 104c can illustrate the change point by displaying a bar chart on the table. In addition, Fig.11D In the case where the condition and intervention both change to the same extent, the change point is illustrated using a mixed color set to an intermediate color (purple). Fig.11D It is a diagram showing a display example of the first embodiment.

[0102] In addition, the display control function 104c may also Fig.11A , Fig. 11B , Fig. 11C as well as Fig.11D Additional information is displayed in addition to the bar lines shown. For example, the display control function 104c may also add text annotations near the bar lines to display patient information, intervention information, etc. other than the first change amount or the second change amount, NT-proBNP, and respiratory rate. For example, the display control function 104c may also change the display content each time the user clicks on the bar lines, while displaying the first change amount or the second change amount, patient information, intervention information, etc. in sequence. In addition, for example, the display control function 104c may also display additional more detailed information each time the user clicks on the bar lines.

[0103] Next, use Fig.12 An example of a processing procedure of the diagnosis support system 1 will be described. Fig.121 is a flowchart for explaining a series of processes of the diagnosis support system 1 according to the first embodiment. Step S101, step S102, and step S103 are steps corresponding to the extraction function 104a. In addition, step S104 is a step corresponding to the calculation function 104b. In addition, step S105 and step S106 are steps corresponding to the display control function 104c.

[0104] First, the processing circuit 104 obtains medical information including patient information related to the state of the patient P1 and intervention information related to the intervention on the patient P1 from the database 20 (step S101). Next, the processing circuit 104 extracts a change point of the medical information (step S102). Next, the processing circuit 104 extracts a first period before the change point and a second period after the change point (step S103).

[0105] Next, the processing circuit 104 calculates the first variation based on the patient information of the first period and the patient information of the second period, and calculates the second variation based on the intervention information of the first period and the intervention information of the second period (step S104). Next, the processing circuit 104 sets a display mode corresponding to at least one of the first variation and the second variation (step S105). Furthermore, the processing circuit 104 performs display based on the first variation and the second variation in the set display mode (step S106), and ends the processing.

[0106] As described above, according to the first embodiment, the database 20 stores medical information including patient information related to the state of the patient P1 and intervention information related to the intervention on the patient P1. In addition, the extraction function 104a extracts the change point of the medical information. In addition, the calculation function 104b calculates a first change amount representing the change of the patient information before and after the change point and a second change amount representing the change of the intervention information before and after the change point. In addition, the display control function 104c performs a display based on the first change amount and the second change amount in a display form set corresponding to at least one of the first change amount and the second change amount. Therefore, the diagnosis support system 1 of the first embodiment can make it easy to grasp the medical information about the patient P1.

[0107] For example, refer to FIG. 11A to FIG. 11D The user of the display can grasp the change point of the medical information according to the position of the bar line, and can grasp the first change amount and the second change amount of each change point according to the color of the bar line. That is, according to the diagnosis support system 1 of the first embodiment, the user can easily grasp the summary of the medical information even when a large number of change points are extracted. Furthermore, by grasping the summary first, the user can efficiently expand the detailed medical information corresponding to the purpose and grasp it in sequence.

[0108] Another method for users to grasp medical information is that users can refer to the medical information by accessing the database 20. However, since there may be a large amount of medical information stored in the database 20, the user has to select and refer to the medical information, which is a burden for the user.

[0109] In addition, as another method for users to grasp medical information, it is possible to consider referring to summaries. A summary is a document made based on past cases, for example, made at the time of transfer or discharge. That is, a summary is a summary of the medical information about the patient during the period before the transfer or discharge. However, depending on the patient, there are cases where a large amount of information is recorded in the summary, and it is not easy for the user to grasp all the recorded medical information. In addition, for example, in the case of wanting to grasp subtle changes in medical information, there are also cases where the information recorded in the summary is insufficient.

[0110] In addition, as another method for the user to grasp the medical information, it is possible to automatically extract a part of the medical information stored in the database 20 and present it to the user. For example, it is possible to automatically extract the change points of the medical information and only extract the medical information about the change points, thereby reducing the amount of medical information presented to the user. However, depending on the patient, there are cases where a large number of change points are determined. In this case, a large amount of medical information is extracted, and the burden on the user who grasps the medical information is also increased.

[0111] In contrast, the diagnosis support system 1 calculates the first change amount and the second change amount based on the extracted change point, and displays the first change amount and the second change amount in a display mode set corresponding to at least one of the first change amount and the second change amount. That is, the diagnosis support system 1 not only extracts medical information corresponding to the change point, but also abstracts it and displays it. As a result, the user can intuitively and easily grasp the change of medical information.

[0112] The first embodiment has been described above, but the present invention can be implemented in various different forms other than the above-described embodiment.

[0113] For example, in FIG. 11A to FIG. 11D In the embodiment, the display mode is set according to the ratio of the first change amount to the second change amount, but the embodiment is not limited thereto. For example, the display control function 104c may set the display mode according to the size of the first change amount and the second change amount in addition to the ratio of the first change amount to the second change amount.

[0114] For example, the calculation function 104b normalizes the first change amount and the second change amount so that they are in the numerical range of "0 to 1". That is, the calculation function 104b calculates the disease state change degree in the numerical range of "0 to 1" and calculates the intervention change degree in the numerical range of "0 to 1". Next, the display control function 104c sets the display form according to the ratio of the disease state change degree to the intervention change degree and the size of each of the disease state change degree and the intervention change degree.

[0115] For example, the display control function 104c sets a mixed color that is a mixture of two colors in a ratio corresponding to the ratio of the disease state change degree to the intervention change degree, and sets the chromaticity in accordance with the magnitude of the sum of the disease state change degree and the intervention change degree. For example, the display control function 104c sets the mixed color so that the greater the sum of the disease state change degree and the intervention change degree, the higher the chromaticity.

[0116] Furthermore, the display control function 104c is connected to FIG. 11A to FIG. 11D Similarly to the case shown in FIG. 1 , a time series display based on at least one of the patient information and the intervention information is performed, and the change point is illustrated by a bar graph with a set mixed color. In this case, the user can understand the change point of the medical information based on the position of the bar graph, and can understand the ratio of the first change amount to the second change amount and the size of each based on the color of the bar graph. For example, in the case of Fig. 10B Similarly, when red and blue are used, if the bar line is bright red, the user can understand that the change in the condition is large relative to the change in intervention and the change in the condition is a large change. In addition, for example, if the bar line is dark red, the user can understand that the change in the condition is large relative to the change in intervention, but the change in the condition is a small change.

[0117] As another example, the display control function 104c sets a mixed color of two colors in a ratio corresponding to the ratio of the disease state change degree to the intervention change degree, and sets the shape of the bar corresponding to the size of the sum of the disease state change degree and the intervention change degree. For example, the display control function 104c sets the thickness of the bar so that the greater the sum of the disease state change degree and the intervention change degree, the thicker the bar. In addition, for example, the display control function 104c sets the bar as a dotted line, and sets the interval so that the greater the sum of the disease state change degree and the intervention change degree, the narrower the interval between points. In addition, for example, the display control function 104c sets the transparency of the bar so that the greater the sum of the disease state change degree and the intervention change degree, the lower the transparency.

[0118] Furthermore, the display control function 104c is FIG. 11A to FIG. 11DSimilarly, in the case shown, a time series display based on at least one of the patient information and the interventional information is performed, and the change point is illustrated by a bar graph of a set mixed color and shape. In this case, the user can grasp the change point of the medical information based on the position of the bar, grasp the ratio of the first change amount to the second change amount based on the color of the bar, and further grasp the size of each of the first change amount and the second change amount based on the shape of the bar.

[0119] In addition, FIG. 11A to FIG. 11D In the embodiment, the change point is illustrated by a bar line, but the embodiment is not limited thereto. For example, the display control function 104c may also illustrate the change point by using other graphics other than a bar line, or by using text. For example, the display control function 104c may illustrate the change point by using a graphic such as an arrow or a triangle with a mixed color corresponding to the ratio of the first change amount and the second change amount. For another example, the display control function 104c may illustrate the change point by using text with a mixed color corresponding to the ratio of the first change amount and the second change amount as the character color.

[0120] In addition, the case where a mixed color in which two colors are mixed at a ratio corresponding to the ratio of the first change amount and the second change amount is set has been described so far, but the embodiment is not limited to this. That is, setting the color as the display form is just an example, and the display control function 104c can also set other display forms other than color. For example, the display control function 104c can also set the order of displaying the first change amount and the second change amount as the display form, and display the first change amount and the second change amount in the set order.

[0121] For example, the calculation function 104b normalizes the first change amount and the second change amount respectively so as to be in the numerical range of "0 to 1". That is, the calculation function 104b calculates the disease state change degree in the numerical range of "0 to 1" and calculates the intervention change degree in the numerical range of "0 to 1". Next, the display control function 104c sets the order of displaying the disease state change degree and the intervention change degree according to the disease state change degree and the intervention change degree.

[0122] For example, the display control function 104c may be Fig.13 As shown, the order of displaying the disease state change degree and the intervention change degree is set so that the greater the intervention change degree and the smaller the disease state change degree, the higher the priority. That is, the display control function 104c determines the order so that the intervention change degree is in descending order and the disease state change degree is in ascending order. In addition, Fig.13 It is a diagram showing a display example of the second embodiment.

[0123] Specifically, in Fig.13In the case shown, the priority of the degree of change of the condition "0.9" indicating the change of the patient information and the degree of change of the intervention "0.1" indicating the change of the intervention information around the change point "2019.5.20 23.34.24" is "1". In addition, the priority of the degree of change of the condition "0.8" indicating the change of the patient information and the degree of change of the intervention "0.2" indicating the change of the intervention information around the change point "2019.5.19 08.54.23" is "2". In addition, the priority of the degree of change of the condition "0.2" indicating the change of the patient information and the degree of change of the intervention "0.2" indicating the change of the intervention information around the change point "2019.5.2312.12.11" is "3". In addition, the priority of the degree of change of the condition "0.1" indicating the change of the patient information and the degree of change of the intervention "0.6" indicating the change of the intervention information around the change point "2019.5.20 11.12.11" is "4". In addition, the priority of the disease change degree "0.1" indicating the change of patient information and the intervention change degree "0.7" indicating the change of intervention information before and after the change point "2019.5.21 14.43.21" is "5". In addition, the priority of the disease change degree "0.2" indicating the change of patient information and the intervention change degree "0.9" indicating the change of intervention information before and after the change point "2019.5.1711.32.43" is "6".

[0124] Fig.13 In the case shown in FIG. 1 , the display control function 104c can display the treatment in a manner that the smaller the change in the condition is relative to the change in the intervention, the higher the priority. Thus, the user can easily find, for example, a treatment that has no effect or a treatment that has a small effect. In addition, the display control function 104c can also be as follows: Fig.13 As shown, the patient information and intervention information before and after the change point are also displayed.

[0125] Alternatively, the display control function 104c may set the order of displaying the disease state change degree and the intervention change degree so that the smaller the intervention change degree and the larger the disease state change degree, the higher the priority. In this case, the display control function 104c may display the disease state change degree in a manner that the larger the change in the disease state is relative to the change in the intervention, the higher the priority. Thus, the user can easily find, for example, an effective treatment.

[0126] Alternatively, the display control function 104c may set the order in which the disease state change degree and the intervention change degree are displayed, so that the greater the disease state change degree is, the higher the priority is. That is, the display control function 104c may also set the order in which the disease state change degree and the intervention change degree are displayed based only on the first change amount. Alternatively, the display control function 104c may set the order in which the disease state change degree and the intervention change degree are displayed, so that the greater the intervention change degree is, the higher the priority is. That is, the display control function 104c may also set the order in which the disease state change degree and the intervention change degree are displayed based only on the second change amount.

[0127] In addition, in the above-mentioned embodiment, the following case is described: the first period before the change point and the second period after the change point are extracted, the first change amount is calculated based on the patient information of the first period and the patient information of the second period, and the second change amount is calculated based on the intervention information of the first period and the intervention information of the second period. However, the embodiment is not limited to this, and the extraction of the first period and the second period by the extraction function 104a can also be omitted. In this case, the calculation function 104b can calculate the first change amount based on the patient information before the change point and the patient information after the change point. In addition, the calculation function 104b can calculate the second change amount based on the intervention information before the change point and the intervention information after the change point.

[0128] Alternatively, after calculating the first change amount and the second change amount for the patient P1, the extraction function 104a may extract another patient P2 whose first change amount and second change amount are similar. For example, the extraction function 104a extracts the distribution of change points on the time axis and the patient P2 whose values ​​of the first change amount and the second change amount calculated for each change point are similar to those of the patient P1, and obtains the medical information about the patient P2 from the database 20.

[0129] Here, for example, when patient P1 is hospitalized and patient P2 has been discharged, the extraction function 104a extracts the medical information about patient P2 for the entire period until discharge. In addition, the display control function 104c causes the display 102 to display the medical information about patient P2. Thus, the user can refer to the patient information about the state of patient P2 to predict the future changes in the condition of patient P1, or refer to the intervention information about the intervention for patient P2 to make an intervention plan for patient P1.

[0130] In the above embodiment, the database 20 stores medical information including patient information and intervention information. However, the embodiment is not limited thereto. For example, the medical information may be stored in the storage circuit 103 of the diagnosis support device 10.

[0131] The components of each device in the above-mentioned embodiments are conceptual functions and do not necessarily need to be physically configured as shown in the figure. That is, the specific form of the distribution and integration of each device is not limited to the content of the figure, and all or part of it can be functionally or physically distributed and integrated in any unit corresponding to various loads and usage conditions. Furthermore, all or any part of each processing function performed in each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware based on wired logic.

[0132] In addition, the diagnostic assistance method described in the above-mentioned embodiment can be implemented by executing a pre-prepared program by a computer such as a personal computer and a workstation. The program can be distributed via a network such as the Internet. In addition, the program can also be recorded in a non-temporary recording medium readable by a computer such as a hard disk, a floppy disk (FD), a CD-ROM, an MO, or a DVD, and read from the recording medium by a computer to be executed.

[0133] According to at least one embodiment described above, medical information about a patient can be easily grasped.

[0134] Several embodiments of the present invention have been described, but these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the scope of the subject matter of the invention. These embodiments and their variations are included in the scope and subject matter of the invention, and are also included in the invention described in the claims and their equivalents.

Claims

1. A diagnostic assistance system, wherein: have: A storage unit storing medical information, the medical information including patient information related to the patient's condition and intervention information related to the intervention on the patient; An extraction unit, extracting the change points of the medical information; a calculation unit, calculating a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point, and standardizing the first change amount and the second change amount respectively; as well as The display control unit performs display based on the first change amount and the second change amount in a display mode set according to a ratio of the normalized first change amount to the normalized second change amount.

2. The diagnostic assistance system according to claim 1, wherein: The extracting unit extracts a first period before the change point and a second period after the change point; The calculation unit calculates the first change amount based on the patient information of the first period and the patient information of the second period, and calculates the second change amount based on the intervention information of the first period and the intervention information of the second period.

3. The diagnostic assistance system according to claim 2, wherein: The calculation unit calculates the second change amount based on a vector indicating the intervention information in the first period and a vector indicating the intervention information in the second period.

4. The diagnostic assistance system according to claim 2, wherein: The calculation unit calculates the first change amount based on the condition of the patient based on the patient information in the first period and the condition of the patient based on the patient information in the second period.

5. The diagnostic assistance system according to claim 1, wherein: The display control unit performs display based on the first change amount and the second change amount using a mixed color obtained by mixing two colors at the ratio.

6. The diagnosis assistance system according to claim 5, wherein: The display control unit performs a time series display based on at least one of the patient information and the intervention information, and graphically displays the change point using the mixed color.

7. A diagnostic aid, wherein: have: an extracting unit for extracting a change point of medical information, wherein the medical information includes patient information related to a patient's condition and intervention information related to an intervention on the patient; a calculation unit, calculating a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point, and standardizing the first change amount and the second change amount respectively; as well as The display control unit performs display based on the first change amount and the second change amount in a display mode set according to a ratio of the normalized first change amount to the normalized second change amount.

8. A diagnostic aid method, wherein: include: extracting change points of medical information, the medical information including patient information related to a patient's condition and intervention information related to an intervention on the patient; calculating a first change amount indicating a change in the patient information before and after the change point and a second change amount indicating a change in the intervention information before and after the change point, and standardizing the first change amount and the second change amount respectively; and Display based on the first change amount and the second change amount is performed in a display mode set according to a ratio of the normalized first change amount to the normalized second change amount.

Images