Medical information processing device, medical information processing method, and program
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- CANON KK
- Filing Date
- 2022-05-20
- Publication Date
- 2026-08-03
Smart Images
Figure 0007898932000001 
Figure 0007898932000002 
Figure 0007898932000003
Abstract
Description
Technical Field
[0004] , , , , ,
[0005] , , ,
[0003] , ,
[0001] The embodiments disclosed in this specification and the drawings relate to a medical information processing apparatus, a medical information processing method, and a program.
Background Art
[0002] Conventionally, as a technique for supporting medical decision-making by doctors and the like, there is known a technique for presenting doctors with various time-series medical data such as patients' test values and disease occurrence probabilities calculated using machine learning models and the like. With this technique, doctors can make treatment plans considering the situation of patients from the past to the future.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] There are various types of medical data, and the increase or decrease of the values of each medical data and the direction of the increase or decrease of the risks that can occur to patients are not uniquely determined. For example, in the case of disease occurrence probability data, the higher the value, the higher the risk. On the other hand, in the case of blood pressure data, both excessive increase and excessive decrease in the value lead to an increase in risk. When checking multiple types of time-series medical data such as medical decision-making, doctors need to carefully discriminate the increase or decrease and goodness or badness of risks for each of these medical data one by one.
[0005] The problem to be solved by the embodiments disclosed in this specification and the drawings is to facilitate the confirmation of risks of multiple types of medical data and support medical decision-making. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. It is also possible to position the problems corresponding to the respective effects of each configuration shown in the embodiments described later as other problems. [Means for solving the problem]
[0006] The medical information processing device of this embodiment includes an acquisition unit, a specification unit, and a generation unit. The acquisition unit acquires first time-series data relating to a first parameter and second time-series data relating to a second parameter different from the first parameter. The specification unit identifies a first risk range relating to the first parameter and a second risk range relating to the second parameter. The generation unit generates display information that shows the first time-series data and the second time-series data in association with each other on a display area where the first risk range and the second risk range are associated. [Brief explanation of the drawing]
[0007] [Figure 1] A diagram showing an example of the usage environment and functional blocks of the medical information processing device 1 according to the first embodiment. [Figure 2] A diagram showing an example of risk criterion information D1 according to the first embodiment. [Figure 3] A flowchart showing an example of the processing flow of the medical information processing device 1 according to the first embodiment. [Figure 4A] A figure showing the probability (vt) of heart failure occurrence according to the first embodiment. [Figure 4B] A figure showing the probability of heart failure occurrence (SR-t) according to the first embodiment. [Figure 5A] A figure showing a weight (vt) graph according to the first embodiment. [Figure 5B] A figure showing a weight (SR-t) graph according to the first embodiment. [Figure 6A] A figure showing a motor function level (vt) graph according to the first embodiment. [Figure 6B] A figure showing the motor function level (SR-t) graph according to the first embodiment. [Figure 7A] A figure showing the drug efficacy (vt) graph according to the first embodiment. [Figure 7B] A figure showing the drug efficacy (SR-t) graph according to the first embodiment. [Figure 8A]Figure showing the LDL-C (v-t) graph according to the first embodiment. [Figure 8B] Figure showing the LDL-C (SR-t) graph according to the first embodiment. [Figure 9A] Figure showing another example of the body weight (v-t) graph according to the first embodiment. [Figure 9B] Figure showing another example of the body weight (SR-t) graph according to the first embodiment. [Figure 10] Figure showing a superimposed graph that simultaneously displays multiple types of medical data according to the first embodiment. [Figure 11] Figure showing the risk change trend (SR-c) graph according to the second embodiment. [Figure 12A] Figure showing the multiple types of medical data (SR-t) graph according to the second embodiment. [Figure 12B] [[ID=二十一]]Figure showing an example of the risk change trend (SR-c) graph according to the second embodiment. [[ID=二十二]] [[ID=二十三]] [Figure 12C] [[ID=二十四]]Figure showing another example of the risk change trend (SR-c) graph according to the second embodiment. [[ID=二十五]] [[ID=二十六]] [Figure 12D] [[ID=二十七]]Figure showing yet another example of the risk change trend (SR-c) graph according to the second embodiment. [[ID=二十八]] [[ID=二十九]] [Figure 13A] [[ID=三十]]Figure for explaining the vertex interpolation of the risk change trend (SR-c) graph according to the second embodiment. [[ID=三十一]] [[ID=三十二]] [Figure 13B] [[ID=三十三]]Figure for explaining the vertex interpolation of the risk change trend (SR-c) graph according to the second embodiment. [[ID=三十四]] [[ID=三十五]] [[ID=三十六]]
Mode for Carrying Out the Invention
[0008] [[ID=四十]] [[ID=四十一]]Hereinafter, a medical information processing apparatus, a medical information processing method, and a program according to an embodiment will be described with reference to the drawings. [[ID=四十二]] [[ID=四十三]]
[0009] [[ID=四十四]] [[ID=四十五]]<First Embodiment>[[ID=四十六]] The medical information processing device of the first embodiment standardizes the meanings represented by the risk criteria (hereinafter referred to as "risk criteria") for multiple types of medical data and the direction of change in the quality of risk (hereinafter referred to as "risk change direction"), facilitating the confirmation of the risks of multiple types of medical data and assisting in medical decision-making. The risk criteria refer to the threshold ε, range θ, etc. predetermined for determining the risk for each type of medical data. The threshold ε refers to clinical judgment values such as diagnostic thresholds, treatment thresholds, and preventive medicine thresholds. The range θ refers to, for example, a reference range predetermined in accordance with a predetermined guideline. The risk change direction refers to the fluctuation direction of the value of the medical data with respect to the risk criteria. The multiple types of medical data include, for example, time-series data (examination values, etc.) of an arbitrary patient such as the probability of heart failure occurrence, body weight, blood pressure, motor function level, drug efficacy, LDL-C (bad cholesterol), TG (triglyceride), etc.
[0010] When the risk of medical data increases, it means that the value of the medical data fluctuates in the direction of becoming less healthy rather than the healthy state within the risk criteria. In the conventional method, when the risk criteria are determined by the threshold ε, if the value of the medical data is greater than the threshold ε (or less than the threshold ε), it is determined that the risk has increased. On the other hand, when the risk criteria are determined by the range θ (β≧θ≧α), if the value of the medical data is greater than the range θ (that is, the value of the medical data is greater than β) or the value of the medical data is less than the range θ (that is, the value of the medical data is less than α), it is determined that the risk has increased. Thus, in the conventional method, the risk change direction varies depending on the type of medical data. On the other hand, in this embodiment, by unifying the meaning of the risk change direction to that the fluctuation of the value of the medical data greater than the risk criteria is "risk increase", the comprehensibility for various types of medical data is improved.
[0011] [Configuration of Medical Information Processing Device] Figure 1 shows an example of the usage environment and functional blocks of the medical information processing device 1. The medical information processing device 1 is deployed, for example, in a medical institution such as a hospital. The medical information processing device 1 may be, for example, a workstation, a server, etc. The medical information processing device 1 is connected to at least one terminal device 3, at least one medical information database 5, etc., via a communication network NW, enabling data transmission and reception. The communication network NW refers to an information and communication network in general that utilizes telecommunications technology. The communication network NW includes wireless / wired LANs such as hospital backbone LANs (Local Area Networks) and the Internet network, as well as telephone communication lines, optical fiber communication networks, cable communication networks, and satellite communication networks.
[0012] The medical information processing device 1 includes, for example, a processing circuit 100, a communication interface 110, and a memory 120. The communication interface 110 communicates with external devices such as a terminal device 3 and a medical information database 5 via a communication network NW. The communication interface 110 includes, for example, a communication interface such as a NIC (Network Interface Card).
[0013] The processing circuit 100 includes, for example, an acquisition function 101, a specific function 102, a calculation function 103, a generation function 104, and a provision function 105. The processing circuit 100 realizes these functions, for example, by a hardware processor (computer) executing a program stored in the memory 120 (storage circuit).
[0014] A hardware processor refers to circuits such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), Application Specific Integrated Circuit (ASIC), and programmable logic devices (e.g., Simple Programmable Logic Device (SPLD) or Complex Programmable Logic Device (CPLD), Field Programmable Gate Array (FPGA)). Instead of storing the program in memory 120, the hardware processor can be configured to directly embed the program within its circuitry. In this case, the hardware processor functions by reading and executing the program embedded within its circuitry.
[0015] The above program may be stored in memory 120 in advance, or it may be stored in a non-temporary storage medium such as a DVD or CD-ROM, and installed from the non-temporary storage medium to memory 120 when the non-temporary storage medium is mounted in the drive device (not shown) of the medical information processing device 1. The hardware processor is not limited to being configured as a single circuit, but may be configured as a single hardware processor by combining multiple independent circuits to realize each function. Alternatively, multiple components may be integrated into a single hardware processor to realize each function.
[0016] The acquisition function 101 acquires multiple types of time-series medical data D2 from the medical information database 5 via the communication network NW and stores them in the memory 120. The acquisition function 101 is an example of the "acquisition unit" in the claims. Specifically, the acquisition function 101 (acquisition unit) acquires first time-series data (first medical data) relating to a first parameter and second time-series data (second medical data) relating to a second parameter different from the first parameter.
[0017] The identification function 102 identifies the risk range for each of the multiple types of acquired medical data D2 based on the risk criterion information D1 stored in the memory 120. Details of the risk range will be described later. The identification function 102 is an example of the "identification unit" in the claims. That is, the identification function 102 (identification unit) identifies a first risk range for a first parameter and a second risk range for a second parameter. The identification function 102 (identification unit) identifies the first risk range and the second risk range based on the risk criteria set in advance according to the type of parameter.
[0018] The calculation function 103 calculates the uniform risk SR and uniform threshold ST for each of the multiple types of acquired medical data D2 based on the risk criterion information D1 stored in the memory 120. The uniform risk SR is an index obtained by processing the value (v) of the medical data using a predetermined conversion formula according to the type (pattern) of the medical data. The uniform threshold ST is an index determined based on the risk criterion information D1. Details of the uniform risk SR and uniform threshold ST will be described later. The calculation function 103 is an example of the "calculation unit" in the claims. That is, the calculation function 103 (calculation unit) calculates the uniformized uniform risk and uniform threshold for each of the first time series data and the second time series data, based on the risk criteria predetermined according to the type of parameter.
[0019] The generation function 104 generates display information for displaying multiple types of acquired medical data D2 in correspondence with each other, using the calculated uniform risk SR and uniform threshold ST. The generation function 104 generates, for example, a vt graph with the value of the medical data (v) on the vertical axis and time (t) on the horizontal axis. Furthermore, the generation function 104 generates, for example, an SR-t graph with the uniform risk (SR) on the vertical axis, time (t) on the horizontal axis, and the origin (SR,t) as (ST,0). This makes it possible to represent different types of time-series medical data on a single type of SR-t graph. In this SR-t graph, fluctuations in values that exceed the standard (uniform threshold ST) are unified to mean "increased risk". The generation function 104 is an example of the "generation unit" in the claims. That is, the generation function 104 (generation unit) generates display information that shows the first time-series data and the second time-series data in correspondence on a display area where the first risk range and the second risk range are associated. The generation function 104 (generation unit) normalizes the first risk range and the second risk range to generate display information that represents a single, unified risk range. The generation function 104 (generation unit) associates the first risk range and the second risk range based on a uniform threshold.
[0020] The provided function 105 transmits the generated display information to the terminal device 3 via the communication network NW.
[0021] Memory 120 can be implemented by semiconductor memory elements such as RAM (Random Access Memory), flash memory, hard disks, or optical discs. These non-transient storage media may also be implemented by other storage devices connected via a communication network NW, such as NAS (Network Attached Storage) or external storage server devices. Memory 120 may also include non-transient storage media such as ROM (Read Only Memory) or registers. Memory 120 stores, for example, risk criterion information D1, medical data D2, etc. In addition, memory 120 stores programs, parameter data, and other data used by the processing circuit 100.
[0022] Figure 2 shows an example of risk criterion information D1. As shown in Figure 2, risk criterion information D1 associates risk criteria (thresholds, ranges) with types of medical data. For example, for "probability of heart failure," the risk criterion is associated with "threshold ε(≧v)." This threshold ε(≧v) means that if the value of the medical data (v) is less than or equal to threshold ε, it is within the reference range (a good condition). Also, for example, for "body weight," the risk criterion is associated with "range θ(β≧θ≧α)." This range θ(β≧θ≧α) means that if the value of the medical data (v) is within the range θ (β≧v≧α), it is within the reference range (a good condition). Also, for example, for "LDL-C," both "threshold ε(≧v)" and "range θ(β≧θ≧α)" are associated with the risk criteria.
[0023] [Configuration of Terminal Device 3] Terminal device 3 is a device for referencing display information (medical data) provided by the medical information processing device 1. Terminal device 3 is operated by an operator, such as a doctor or technician. Terminal device 3 can be, for example, a personal computer, a tablet, or a smartphone.
[0024] The terminal device 3 includes, for example, a communication interface 30, an input interface 32, a display 34, and a processing circuit 36. The communication interface 30 communicates with external devices such as the medical information processing device 1 and the medical information database 5 via a communication network NW.
[0025] The input interface 32 receives various input operations from the operator of the terminal device 3, converts the received input operations into electrical signals, and outputs them to the processing circuit 36. For example, the input interface 32 includes a mouse, keyboard, trackball, switch, button, joystick, touch panel, etc. The input interface 32 may also be a user interface that accepts audio input, such as from a microphone.
[0026] In this specification, the term "input interface" is not limited to those equipped with physical operating components such as a mouse or keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device located separately from the device and outputs this electrical signal to a control circuit is also included as an example of an input interface.
[0027] The display 34 displays various types of information. For example, the display 34 displays images generated by the processing circuit 36, or a GUI (Graphical User Interface) for receiving various input operations from the operator. For example, the display 34 may be an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, or an organic EL (Electro Luminescence) display.
[0028] The processing circuit 36 launches a dedicated application program or browser, etc., and displays the display information (medical data) provided by the medical information processing device 1 on the display 34. The processing circuit 36 also generates a GUI for receiving various input operations from the operator and displays it on the display 34. For example, the processing circuit 36 generates a GUI for receiving input operations for medical data requests from the operator and displays it on the display 34. If it receives an input operation for a request to acquire medical data, it sends the request to acquire medical data to the medical information processing device 1.
[0029] [Structure of Medical Information Database 5] The medical information database 5 stores patient-specific medical data acquired by various diagnostic and testing devices. The medical information database 5 stores arbitrary medical data such as the probability of heart failure, weight, blood pressure, exercise function level, drug efficacy, LDL-C (bad cholesterol), and TG (triglycerides), associated with the patient ID. The medical information database 5 can be implemented using semiconductor memory elements such as RAM and flash memory, hard disks, or optical discs. The medical information database 5 may also be integrated into the medical information processing device 1.
[0030] [Processing flow] Next, the processing of the medical information processing device 1 will be explained. Figure 3 is a flowchart showing an example of the processing flow of the medical information processing device 1. The processing shown in Figure 3 is executed when the medical information processing device 1 receives a processing request transmitted from the terminal device 3 based on an operation by a doctor or the like.
[0031] First, the acquisition function 101, in response to an acquisition request from the terminal device 3, acquires multiple types of time-series medical data of the patient to be acquired from the medical information database 5 via the communication interface 110 (step S101). The acquisition function 101 acquires medical data from the medical information database 5 based, for example, on the patient ID included in the acquisition request. The acquisition function 101 stores the acquired medical data in the memory 120.
[0032] Next, the specific function 102 obtains risk criterion information D1 from memory 120 (step S103). The specific function 102 obtains, for example, information on risk criteria (thresholds and / or ranges) corresponding to the type of medical data obtained by the acquisition function 101.
[0033] Next, the identification function 102 identifies the risk range for each type of medical data based on the risk criteria (step S105). Then, the calculation function 103 calculates the uniform risk SR and uniform threshold ST based on the risk criteria and medical data (step S107). The content of these risk range identification processes and uniform risk SR and uniform threshold ST calculation processes differs depending on the type of medical data (setting of risk criteria). These processes will be explained below separately for each setting of risk criteria.
[0034] (1) When a threshold ε(≧v) is set as the risk criterion Figures 4A and 4B show graphs of the probability of heart failure occurrence (vt) and (SR-t), respectively. As shown in Figure 2, for the probability of heart failure occurrence, only a threshold is defined as the risk criterion. Furthermore, this threshold is "ε(≧v)", meaning that fluctuations in the value of clinical data that exceed the standard threshold ε represent a pattern of increased risk. In this case, as shown in Figure 4A, if the horizontal axis is time (t) and the vertical axis is the probability of heart failure occurrence (v) (value of clinical data), the specific function 102 identifies the range above the threshold ε on the vertical axis as the risk range (range requiring attention).
[0035] Furthermore, as shown in Figure 4B, the calculation function 103 calculates the same value as the probability of heart failure (v) as the uniform risk SR (SR=v). The calculation function 103 also calculates the same value as the threshold ε as the uniform threshold ST (ST=ε). As a result, the heart failure probability (SR-t) graph shown in Figure 4B has the same waveform as the heart failure probability (vt) graph shown in Figure 4A. Note that the black circles in Figures 4A and 4B represent measured values, and the white circles represent predicted values. The same applies to other graphs below.
[0036] In other words, if a threshold is set as a risk criterion, the calculation function 103 (calculation unit) uses the threshold as a uniform threshold and calculates the values of the first time series data and the second time series data as a uniform risk.
[0037] (2) When a range (β≧θ≧α) is defined as the risk criterion Figures 5A and 5B show the weight (vt) graph and (SR-t) graph. As shown in Figure 2, for weight (1), only a range is defined as the risk criterion. Furthermore, this range is "β≧θ≧α", meaning that both fluctuations in the value of the medical data that are greater than the criterion (range θ) (v>θ) and fluctuations in the value of the medical data that are smaller than the risk criterion (v<θ) are patterns of increased risk. This is because both excessive increases and excessive decreases in weight lead to an increased risk. In this case, as shown in Figure 5A, when the horizontal axis is time (t) and the vertical axis is weight (v), the specific function 102 identifies both the range excluding the criterion range θ on the vertical axis, i.e., the range above β and the range below α, as the risk range (range requiring attention).
[0038] Furthermore, as shown in Figure 5B, the calculation function 103 calculates the uniform risk SR based on the following formula (1). The calculation function 103 also calculates the uniform threshold ST based on the following formula (2). As a result, in the weight (SR-t) graph shown in Figure 5B, any change in weight value that exceeds the standard (uniform threshold ST) is unified to mean "increased risk".
[0039] SR = |v - (α + β) / 2| ... Equation (1) ST = (β - α) / 2 ... Equation (2)
[0040] In other words, the calculation function 103 (calculation unit) calculates uniform risk based on the midpoint value between the upper and lower limits of a range when a range is set as the risk criterion. If the identification unit identifies two risk ranges for one parameter based on the risk criterion, the calculation function 103 (calculation unit) sets half the difference between the upper and lower limits of the range as the uniform threshold.
[0041] (3) When a threshold ε(≦v) is defined as a risk criterion Figures 6A and 6B show the motor function level (vt) graph and (SR-t) graph. As shown in Figure 2, for motor function level, only a threshold is defined as a risk criterion. Furthermore, this threshold is "ε(≦v)", meaning that fluctuations in the value of the clinical data that fall below the risk criterion represent a pattern of increased risk. In this case, as shown in Figure 6A, if the horizontal axis is time (t) and the vertical axis is motor function level (v), the specific function 102 identifies the range below the threshold ε on the vertical axis as the risk range (range requiring attention).
[0042] Furthermore, as shown in Figure 6B, the calculation function 103 calculates v', which is obtained by transforming the motor function level (v) so that it is symmetrical to v=ε, as the uniform risk SR (SR=v'). The calculation function 103 also calculates the same value as the threshold ε as the uniform threshold ST (ST=ε). As a result, the motor function level (SR-t) graph shown in Figure 6B is a waveform obtained by transforming the motor function level (vt) graph shown in Figure 6A so that it is symmetrical to v=ε, and fluctuations in values that are larger than the standard (uniform threshold ST) are unified to mean "increased risk".
[0043] In other words, if a threshold is set as a risk criterion, the calculation function 103 (calculation unit) uses the threshold as a uniform threshold and calculates a uniform risk by symmetrically transforming the values of the first time series data and the second time series data based on the threshold.
[0044] (4) When a range (v≧β, α≧v) is defined as the risk criterion Figures 7A and 7B show the drug efficacy (vt) graph and (SR-t) graph. As shown in Figure 2, for drug efficacy, only a range is defined as a risk criterion. Furthermore, this range is "v≧β, α≧v", meaning that the clinical data included in the specified range is a pattern of increased risk. This is because for drug efficacy, the range other than the range in which it can be judged as effective (v≧β) or ineffective (α≦v), that is, the range of unknown effect where it cannot be judged as either effective or ineffective, leads to an increase in risk. In this case, as shown in Figure 7A, if the horizontal axis is time (t) and the vertical axis is drug efficacy (v), the specific function 102 identifies the range on the vertical axis "β>v>α" as the risk range (range requiring attention).
[0045] Furthermore, as shown in Figure 7B, the calculation function 103 calculates the uniform risk SR based on the following equations (3) and (4). The calculation function 103 also calculates the uniform threshold ST based on the following equation (5). As a result, in the drug efficacy (SR-t) graph shown in Figure 7B, any change in weight value that exceeds the risk criterion (uniform threshold ST) is unified to mean "increased risk".
[0046] SR=2*(α+β) / 2-v=α+β-v···v>(α+β) / 2···Formula (3) SR=v···v≦(α+β) / 2···Formula (4) ST=α...Equation (5)
[0047] In other words, when the calculation function 103 (calculation unit) identifies one risk range for one parameter based on the risk criteria by the identification unit, it calculates the lower limit of the range as a uniform threshold.
[0048] (5) When both a threshold and a range are defined as risk criteria Figures 8A and 8B show the LDL-C(vt) graph and (SR-t) graph, respectively. As shown in Figure 2, both a threshold and a range are defined as risk criteria for LDL-C. Here, the threshold "ε(≧v)" is set within the range "β≧θ≧α". Thus, for LDL-C, a clinical judgment value (diagnostic threshold (ε)) lower than the upper limit of the reference range (β) may be set. In this case, as shown in Figure 8A, if the horizontal axis is time (t) and the vertical axis is LDL-C(v), the specific function 102 exceeds the threshold ε on the vertical axis, but the area within the range "β≧θ≧α" is defined as a gray zone (quasi-risk range), and the area outside the range "β≧θ≧α" is identified as a range requiring attention.
[0049] Furthermore, as shown in Figure 8B, the calculation function 103 calculates the same value as LDL-C(v) as the uniform risk SR (SR=v). The calculation function 103 also calculates the same value as the threshold ε as the uniform threshold ST (ST=ε). As a result, the LDL-C(SR-t) graph shown in Figure 8B has the same waveform as the LDL-C(vt) graph shown in Figure 8A, and fluctuations in values that exceed the standard (uniform threshold ST) are unified to mean "increased risk".
[0050] Another example of a risk criterion where both a threshold and a range are defined is triglycerides (TG). For TG, a clinical judgment value (diagnostic threshold) higher than the upper limit of the reference range may be set. In this case, for example, the lower of the diagnostic threshold and the upper limit of the reference range (in this case, the upper limit of the reference range) may be set as the uniform threshold ST. In this case, it may be indicated that a gray zone is included.
[0051] In other words, the identification function 102 (identification unit) identifies a quasi-risk range based on the threshold and range when both a threshold and a range are set for a single parameter as a risk criterion. The calculation function 103 (calculation unit) calculates the smaller of the threshold and the lower limit of the risk range for a single parameter as the uniform threshold.
[0052] (6) When a nonlinear threshold or range is defined as a risk criterion Figures 9A and 9B show other examples of weight (vt) graphs and (SR-t) graphs. As shown in Figure 2, for weight (2), only a range is defined as the risk criterion. Furthermore, this range is defined as "β(t)≧θ≧α(t)", meaning the definition of the risk criterion range is a function of time (t) (nonlinear). In this case, as shown in Figure 9A, when the horizontal axis is time (t) and the vertical axis is weight (v), the specific function 102 identifies both the range above β(t) and the range below α(t) as the risk range (range requiring attention), excluding the standard range θ on the vertical axis.
[0053] Furthermore, as shown in Figure 9B, the calculation function 103 calculates the uniform risk SR based on the following formula (6). The calculation function 103 also calculates the uniform threshold ST based on the following formula (7). As a result, in the weight (SR-t) graph shown in Figure 9B, any change in weight value that exceeds the standard (uniform threshold ST) is unified to mean "increased risk".
[0054] SR=|v-(α(t)+β(t)) / 2|...Equation (6) ST=(β(t)-α(t)) / 2...Equation (7)
[0055] In the above explanation, when determining the uniform threshold ST for a range, we assumed that the medical data followed a normal distribution and used the midpoint of the range as an example. If the medical data does not follow a normal distribution, the range threshold may be set to the midpoint by converting the data to a normal distribution using a parametric method.
[0056] Returning to Figure 3, the generation function 104 uses the uniform risk SR and uniform threshold ST calculated by the calculation function 103 to generate display information for displaying multiple types of medical data D2 acquired by the acquisition function 101 in correspondence with each other (step S109). Figure 10 shows a superimposed graph that displays multiple types of medical data simultaneously. The superimposed graph shown in Figure 10 displays the heart failure incidence probability (SR-t) graph shown in Figure 4B, the body weight (SR-t) graph shown in Figure 5B, and the exercise function level (SR-t) graph shown in Figure 6B, superimposed with the uniform threshold ST aligned. As described above, in the graphs shown in Figures 4B, 5B, and 6B, fluctuations in values that are larger than the standard (uniform threshold ST) are unified to mean "increased risk," making it possible to display a superimposed graph like the one shown in Figure 10. Physicians can refer to this superimposed graph to simultaneously and easily check the risk of multiple types of medical data. For example, doctors can focus on clinical data that penetrates deeper into areas of concern and make informed decisions about treatment.
[0057] In other words, the generation function 104 (generation unit) generates display information in the form of a graph with time on the horizontal axis and uniform risk on the vertical axis.
[0058] Furthermore, the SR-t graph may be normalized by setting the uniform threshold ST to 0.5 and ensuring that the uniform risk SR falls within the range of 0 to 1. This improves visibility when displaying multiple types of clinical data on the superimposed graph shown in Figure 10. Additionally, the vertical axis (uniform risk SR axis) may be displayed on a logarithmic scale.
[0059] Furthermore, on the superimposed graph shown in Figure 10, related data from multiple types of clinical data may be displayed in a way that indicates their relationship. For example, when displaying data for "left ventricular ejection fraction (LVEF)" and "brain natriuretic peptide (BNP)," which are related to the disease of heart failure, they may be displayed in similar colors, or when one data is selected, the other data may be highlighted. In addition, current risk and future risk may be displayed in a way that distinguishes them. For example, current risk may be displayed with a solid line, and future risk may be displayed with a dashed line.
[0060] Next, the providing function 105 transmits (provides) the display information generated by the generation function 104 to the terminal device 3 via the network NW (step S111). As a result, the display information is displayed on the display of the terminal device 3, and the doctor can confirm this display information. This completes the processing of this flowchart. In the above explanation, continuous data was used as an example, but it can also be applied to discrete data such as the number of times a drug is administered.
[0061] According to the first embodiment described above, it is possible to easily identify risks in multiple types of medical data and support medical decision-making. This makes it immediately clear which data to focus on among multiple types of medical data. In addition, it is possible to reduce the chances of overlooking various changes in a patient's risk.
[0062] <Second Embodiment> The second embodiment will now be described. The second embodiment differs from the first embodiment in that the medical information processing device 1 generates a risk change trend (SR-c) graph, which shows the amount of change in risk per unit time, instead of a superimposed graph (SR-t graph) as display information. The following description will focus on the differences from the first embodiment, and the points common to the first embodiment will not be explained. In the description of the second embodiment, the same reference numerals will be used for parts that are the same as in the first embodiment.
[0063] Figure 11 shows a graph of the risk change trend (SR-c). The graph shown in Figure 11 is generated based on the heart failure incidence probability (SR-t) graph shown in Figure 4B and the body weight (SR-t) graph shown in Figure 5B. Alternatively, the graph shown in Figure 11 is generated based on the heart failure incidence probability (vt) graph shown in Figure 4A and the body weight (vt) graph shown in Figure 5B. In the graph shown in Figure 11, the horizontal axis represents the risk change (c), which shows the amount of change in risk per unit time, and the vertical axis represents the uniform risk SR. The risk change (c) is calculated by the following equation (8).
[0064] Risk change c = v(t) - v(t-1) ... Equation (8)
[0065] In the graph shown in Figure 11, the first quadrant (D: caution required), the second quadrant (A: improving but high risk), the third quadrant (B: safe), and the fourth quadrant (C: worsening but still low risk) are defined. The origin (SR,c) is set to (ST,0). The marks MK attached to the uniform risk SR of each clinical data indicate that the time (t) of each peak (plot) corresponds to each other. By checking this risk change trend (SR-c) graph, physicians can easily see the amount of risk change per unit time. Physicians can focus particularly on clinical data showing larger risk changes and make informed decisions regarding treatment.
[0066] Figures 12A to 12D show other examples of risk change trend (SR-c) graphs. Figure 12A is a superimposed graph that simultaneously displays SR-c graphs for three clinical data sets A to C. Figure 12B is an SR-c graph generated based on all the plots in the superimposed graph shown in Figure 12A. In this SR-c graph, plots with small risk changes are concentrated near the vertical axis, making it difficult to see the risk change trend.
[0067] On the other hand, the SR-c graph shown in Figure 12C is an SR-c graph generated by applying a moving average to each clinical data point in the superimposed graph shown in Figure 12A. By performing such a moving average, the trend of risk changes becomes easier to understand, and the visibility is improved.
[0068] Furthermore, the SR-c graph shown in Figure 12D is an SR-c graph generated by applying moving average processing and decimation processing to each clinical data point of the superimposed graph shown in Figure 12A. By performing such moving average processing and decimation processing, the trend of risk changes becomes even clearer, and the visibility is improved.
[0069] Furthermore, while the SR-c graph can be made more visually appealing through arbitrary processing such as moving average processing and decimation, it will no longer be able to simply correspond to the vt graph or SR-t graph. For example, time information is lost from the SR-c graph by the amount of vertices that are decimated. As a result, physicians may misinterpret the time changes in risk. Therefore, as shown below, the vertices of the original SR-c graph are generated at corresponding locations on the curve of the processed SR-c graph.
[0070] Figures 13A and 13B illustrate the vertex interpolation process for the risk change trend (SR-c) graph. As shown in Figure 13A, when arbitrary processing (e.g., moving average processing and decimation processing) is applied to the original (A) vertices that indicate risk changes on the SR-c graph, a processed (A') is generated with a reduced number of vertices from the original (A). By performing vertex interpolation processing on this processed (A'), a processed (A'') is generated. When performing this vertex interpolation processing, as shown in Figure 13B, dynamic programming (DP) matching is performed between the original (A) and the processed (A') based on information such as the risk value, risk change amount, and time associated with each vertex, to achieve flexible mapping in one-dimensional space. Depending on the processing method, a unique mapping can be achieved (e.g., if time is not processed). Furthermore, vertices (AP1, AP2, AP3) are generated in places where there are no vertices in the processed (A'). Note that the spacing between vertices is not equal, as it represents the interval of risk change. If the number of vertices in the processed (A') is greater than that of the original (A), the vertices that could not be matched are deleted.
[0071] For example, if the vertices in the original graph (A) are spaced one day apart (a total of six vertices represent five days' worth of data changes), the processed graph (A') will have only three vertices, making it appear as if only two days' worth of data changes are represented. However, by performing the vertex interpolation process described above, it becomes possible to recognize that the processed graph (A'') still represents five days' worth of data changes. Furthermore, even in this processed graph (A''), it remains easier to track risk changes than in the original graph (A). This is because adding vertices to the processed graph (A') does not change the shape of the graph itself.
[0072] In other words, the generation function 104 (generation unit) generates display information in the form of a graph where the horizontal axis represents the amount of change of the first time series data and the second time series data, and the vertical axis represents uniform risk. The generation function 104 (generation unit) generates display information that shows processed data, which has undergone predetermined processing on the uniform risk, on the graph.
[0073] According to the second embodiment described above, it is possible to easily check the risks of multiple types of medical data and support medical decision-making. This makes it immediately clear which data to focus on among multiple types of medical data. It also reduces the chance of overlooking various changes in a patient's risk. Furthermore, by using the risk change trend (SR-c) graph, it becomes possible to easily grasp the risk change trend.
[0074] Furthermore, some or all of the functions of the medical information processing device 1 described above may be implemented in the terminal device 3. In this case, the terminal device 3 is an example of a "medical information processing device".
[0075] The embodiments described above can be expressed as follows. Equipped with processing circuitry, The aforementioned processing circuit is Obtain first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter. Identify a first risk range for the first parameter and a second risk range for the second parameter. Display information is generated that shows the first time series data and the second time series data in association on a display area where the first risk range and the second risk range are associated. Medical information processing device.
[0076] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]
[0077] 1…Medical information processing device, 3…Terminal device, 5…Medical information database, 30…Communication interface, 32…Input interface, 34…Display, 36…Processing circuit, 100…Processing circuit, 101…Acquisition function, 102…Identification function, 103…Calculation function, 104…Generation function, 105…Provision function, 110…Communication interface, 120…Memory
Claims
1. An acquisition unit that acquires first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter, A specification unit that identifies a first risk range for the first parameter and a second risk range for the second parameter based on a risk criterion set in advance according to the type of parameter, A generation unit generates display information that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated, A medical information processing device comprising, Furthermore, the system includes a calculation unit that calculates a standardized, uniform risk and a uniform threshold for each of the first and second time series data based on the risk criteria. The generation unit aligns the positions of the uniform thresholds to associate the first risk range with the second risk range, and normalizes the first risk range and the second risk range to generate the display information that represents a single uniform risk range. If a threshold is set as the risk criterion, the calculation unit sets the threshold as the uniform threshold and calculates the uniform risk as the values of the first time series data and the second time series data, or the values of the first time series data and the second time series data obtained by symmetrically transforming them based on the threshold. Medical information processing device.
2. The generation unit generates the display information which is a graph with time on the horizontal axis and the uniform risk on the vertical axis. The medical information processing device according to claim 1.
3. An acquisition unit that acquires first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter, A specification unit that identifies a first risk range for the first parameter and a second risk range for the second parameter based on a risk criterion set in advance according to the type of parameter, A generation unit generates display information that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated, A medical information processing device comprising, Furthermore, the system includes a calculation unit that calculates a standardized, uniform risk and a uniform threshold for each of the first and second time series data based on the risk criteria. The generation unit aligns the positions of the uniform thresholds to associate the first risk range with the second risk range, and normalizes the first risk range and the second risk range to generate the display information that represents a single uniform risk range. The calculation unit, when a reference range is set as the risk criterion, calculates the uniform risk based on the midpoint value between the upper and lower limits of the reference range. Medical information processing device.
4. When the calculation unit identifies two risk ranges for a single parameter based on the risk criteria for which the reference range is set by the identification unit, namely a risk range above the upper limit of the reference range and a risk range below the lower limit of the reference range, the calculation unit calculates half the difference between the upper limit and the lower limit as the uniform threshold. The medical information processing device according to claim 3.
5. An acquisition unit that acquires first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter, A specification unit that identifies a first risk range for the first parameter and a second risk range for the second parameter based on a risk criterion set in advance according to the type of parameter, A generation unit generates display information that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated, A medical information processing device comprising, Furthermore, the system includes a calculation unit that calculates a standardized, uniform risk and a uniform threshold for each of the first and second time series data based on the risk criteria. The generation unit aligns the positions of the uniform thresholds to associate the first risk range with the second risk range, and normalizes the first risk range and the second risk range to generate the display information that represents a single uniform risk range. The specified unit, when both a threshold and a reference range are set for one parameter as the risk criterion, identifies a quasi-risk range based on the threshold and the reference range. The calculation unit calculates the smaller of the threshold value and the lower limit of the risk range for one parameter as the uniform threshold value. The generation unit generates the display information indicating a standardized risk range and a quasi-risk range. Medical information processing device.
6. An acquisition unit that acquires first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter, A specification unit that identifies a first risk range for the first parameter and a second risk range for the second parameter based on a risk criterion set in advance according to the type of parameter, A generation unit generates display information that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated, A medical information processing device comprising, Furthermore, the system includes a calculation unit that calculates a standardized, uniform risk and a uniform threshold for each of the first and second time series data based on the risk criteria. The generation unit aligns the positions of the uniform thresholds to associate the first risk range with the second risk range, and normalizes the first risk range and the second risk range to generate the display information that represents a single uniform risk range. The generation unit generates the display information, which is a graph in which the horizontal axis represents the change in the first time series data and the second time series data, and the vertical axis represents the uniform risk. Medical information processing device.
7. The generation unit generates the display information shown on the graph from the processed data that has undergone predetermined processing for the uniform risk. The medical information processing device according to claim 6.
8. Computers Obtain first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter. Based on a risk criterion set in advance according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. A medical information processing method, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If a threshold is set as the risk criterion, the threshold is set as the uniform threshold, and the values of the first time series data and the second time series data, or the values of the first time series data and the second time series data obtained by symmetrically transforming them based on the threshold, are calculated as the uniform risk. Medical information processing method.
9. A computer, Obtain first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter. Based on a risk criterion set in advance according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. A medical information processing method, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If a reference range is set as the risk criterion, the uniform risk is calculated based on the midpoint value between the upper and lower limits of the reference range. Medical information processing method.
10. A computer, Obtain first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter. Based on a risk criterion set in advance according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. A medical information processing method, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If both a threshold and a reference range are set for one parameter as the risk criterion, a quasi-risk range is identified based on the threshold and the reference range. The smaller of the aforementioned threshold and the lower limit of the risk range for one parameter is calculated as the uniform threshold. The system generates the display information that shows a standardized risk range and a quasi-risk range. Medical information processing method.
11. A computer, Obtain first time series data relating to a first parameter and second time series data relating to a second parameter different from the first parameter. Based on a risk criterion set in advance according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. A medical information processing method, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. The system generates the display information, which is a graph in which the horizontal axis represents the change in the first time series data and the second time series data, and the vertical axis represents the uniform risk. Medical information processing method.
12. On the computer, The system obtains a first time series data relating to a first parameter and a second time series data relating to a second parameter different from the first parameter. Based on a preset risk criterion according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. It is a program, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If a threshold is set as the risk criterion, the threshold is set as the uniform threshold, and the values of the first time series data and the second time series data, or the values of the first time series data and the second time series data obtained by symmetrically transforming them based on the threshold, are used to calculate the uniform risk. program.
13. A computer, The system obtains a first time series data relating to a first parameter and a second time series data relating to a second parameter different from the first parameter. Based on a preset risk criterion according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. It is a program, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If a reference range is set as the risk criterion, the uniform risk is calculated based on the midpoint value between the upper and lower limits of the reference range. program.
14. A computer, The system obtains a first time series data relating to a first parameter and a second time series data relating to a second parameter different from the first parameter. Based on a preset risk criterion according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. It is a program, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. If both a threshold and a reference range are set for one parameter as the risk criterion, a quasi-risk range is identified based on the threshold and the reference range. The smaller of the aforementioned threshold and the lower limit of the risk range for one parameter is calculated as the uniform threshold. To generate the display information that shows a standardized risk range and a quasi-risk range, program.
15. A computer, The system obtains a first time series data relating to a first parameter and a second time series data relating to a second parameter different from the first parameter. Based on a preset risk criterion according to the type of parameter, a first risk range for the first parameter and a second risk range for the second parameter are identified. Display information is generated that shows the first time-series data and the second time-series data in association on a single display area where the first risk range and the second risk range are associated. It is a program, Furthermore, based on the risk criteria, a standardized, uniform risk and a uniform threshold are calculated for each of the first and second time series data. By aligning the positions of the uniform thresholds and associating the first risk range with the second risk range, and by normalizing the first risk range and the second risk range, the display information indicating a single uniform risk range is generated. The system generates the display information, which is a graph in which the horizontal axis represents the change in the first time series data and the second time series data, and the vertical axis represents the uniform risk. program.