Medical support device and medical support method

Abstract

To appropriately and efficiently support medical activity for a subject to whom external force is applied.SOLUTION: A medical support device includes an external force information acquisition unit, a medical support information generation unit, and an output control unit. The external information acquisition unit acquires external force information on external force applied to a subject. The medical support information generation unit generates medical support information for supporting medical activity for the subject on the basis of the external force information acquired by the external force information acquisition unit. The output control unit controls output from an output part of the medical support information generated by the medical support information generation unit.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The embodiments disclosed in the present specification and drawings relate to a medical support device and a medical support method. [Background technology]

[0002] Emergency medical care for traffic injuries faces two challenges: early treatment and accurate diagnosis. The challenge of early treatment is to begin treatment as quickly as possible after an accident, which can be achieved using conventional emergency notification systems. However, it is extremely difficult to accurately determine the extent of injuries sustained by injured people brought to emergency hospitals in a short amount of time. In particular, if an injured person is unconscious, doctors may not be able to interview the patient about the history of their injuries or their pain. Apart from visible injuries such as incision bleeding, fractures, and dislocations, doctors must infer the extent of their injuries based on basic medical information such as pulse, blood pressure, heart rate, electrocardiogram, and electroencephalogram.

[0003] If the injured person is conscious, they may not feel any pain or other abnormalities immediately after the injury, such as internal organ or cranial nerve damage, and unless they report a bruise, detailed examinations such as CT scans or MRIs may not be performed. This can delay treatment for serious hidden injuries. In light of these issues, an emergency notification system has been disclosed that detects and reports the stress on occupants caused by a collision, enabling accurate diagnosis.

[0004] Furthermore, in cases of high-energy trauma (such as falls from a height or car accidents at a certain speed or higher), diagnosis is made with reference to the trauma initial medical treatment guidelines. In the second stage (secondary survey) of the three-stage interpretation of the trauma initial medical treatment guidelines, diagnosis is made by systematically searching for damage throughout the subject's body. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2002-127857 A Summary of the Invention [Problem to be solved by the invention]

[0006] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to appropriately and efficiently support medical procedures on subjects to whom external forces have been applied. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of the configurations shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0007] A medical support device according to an embodiment includes an external force information acquisition unit, a medical support information generation unit, and an output control unit. The external force information acquisition unit acquires external force information related to an external force applied to a subject. The medical support information generation unit generates medical support information for supporting medical treatment on the subject based on the external force information acquired by the external force information acquisition unit. The output control unit controls output from the output unit of the medical support information generated by the medical support information generation unit. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic diagram showing the configuration of a medical support device according to a first embodiment. [Figure 2] FIG. 2 is a schematic diagram showing the configuration of a medical support system equipped with a medical support device according to the first embodiment. [Figure 3] FIG. 3 is a block diagram showing an example of functions of the medical support device according to the first embodiment. [Figure 4] FIG. 4 is a flowchart showing a medical support method in the medical support device according to the first embodiment. [Figure 5] FIG. 5 is a diagram showing a first display example of superimposition information in the medical support device according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing a second display example of superimposition information in the medical support device according to the first embodiment. [Figure 7] FIG. 7 is a diagram showing a third display example of superimposition information in the medical support device according to the first embodiment. [Figure 8] FIG. 8 is a diagram showing a fourth display example of superimposition information in the medical support device according to the first embodiment. [Figure 9] FIG. 9 is a diagram showing a fifth display example of superimposition information in the medical support device according to the first embodiment. [Figure 10] FIG. 10 is a diagram showing a sixth display example of superimposition information in the medical support device according to the first embodiment. [Figure 11] FIG. 11 is a schematic diagram showing the configuration of a medical support system equipped with a medical support device according to a second embodiment. [Figure 12] FIG. 12 is an explanatory diagram showing an example of a data flow during learning in the medical support device according to the second embodiment. [Figure 13] FIG. 13 is an explanatory diagram showing an example of a data flow during operation of the medical support device according to the second embodiment. [Figure 14] FIG. 14 is an explanatory diagram showing an example of a data flow during learning in the medical support device according to the third embodiment. [Figure 15] FIG. 15 is an explanatory diagram showing an example of a data flow during operation of the medical support device according to the third embodiment. [Figure 16] FIG. 16 is an explanatory diagram showing an example of a data flow during learning in the medical support device according to the fourth embodiment. [Figure 17] FIG. 17 is an explanatory diagram showing an example of a data flow during operation of the medical support device according to the fourth embodiment. [Figure 18] FIG. 18 is a diagram showing an example of display of damaged site information in the medical support device according to the fourth embodiment. [Figure 19] FIG. 19 is an explanatory diagram showing an example of a data flow during learning in the medical support device according to the fifth embodiment. [Figure 20]FIG. 20 is an explanatory diagram showing an example of a data flow during operation of the medical support device according to the fifth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of a medical support device and a medical support method will be described in detail with reference to the drawings.

[0010] (First embodiment) FIG. 1 is a schematic diagram showing the configuration of a medical support device according to the first embodiment.

[0011] 1 shows a medical support device 10 according to the first embodiment. The medical support device 10 is an imaging diagnostic device (data server), a workstation, an image interpretation terminal, or the like, and is provided on a medical image system connected via a network N (shown in FIG. 2). The medical support device 10 may also be an offline device.

[0012] The medical support device 10 includes a processing circuit 11, a memory circuit 12, an input interface 13, a display 14, and a network interface 15.

[0013] The processing circuitry 11 controls the operation of the medical support device 10 in response to an input operation received from an operator via the input interface 13. For example, the processing circuitry 11 is realized by a processor. The function of the processing circuitry 11 will be described later with reference to FIG. 3.

[0014] The memory circuitry 12 is configured with semiconductor memory elements such as RAM (Random Access Memory) and flash memory, a hard disk, an optical disk, etc. The memory circuitry 12 may be configured with portable media such as USB (Universal Serial Bus) memory and DVD (Digital Video Disk). The memory circuitry 12 stores various processing programs (including application programs and an OS (Operating System)) used in the processing circuitry 11, data required for executing the programs, etc. The OS may also include a GUI (Graphical User Interface) that makes extensive use of graphics to display information to the operator on the display 14 and allows basic operations to be performed via the input interface 13. The memory circuitry 12 is an example of a memory unit.

[0015] The input interface 13 includes an input device operable by an operator and an input circuit that inputs signals from the input device. The input device may be a trackball, a switch, a mouse, a keyboard, a touchpad that performs input operations by touching the operation surface, a touchscreen that combines the display screen and touchpad, a non-contact input device using an optical sensor, or a voice input device. When the operator operates the input device, the input circuit generates a signal corresponding to the operation and outputs it to the processing circuit 11. The medical support device 10 may also include a touch panel in which the input device is integrated with the display 14. The input device is not limited to devices that include physical operating components such as a mouse and a keyboard. For example, the input interface 13 may include a configuration in which the input circuit receives an electrical signal corresponding to an input operation from an external input device provided separately from the medical support device 10 and outputs the electrical signal to the processing circuit 11. The input interface 13 is an example of an input unit.

[0016] The display 14 is a display device such as a liquid crystal display panel, a plasma display panel, or an organic EL (Electro Luminescence) panel. The display 14 is connected to the processing circuitry 11 and displays various information and images generated under the control of the processing circuitry 11. The display 14 is an example of an output unit. The medical support device 10 may also include, as another output unit, a speaker (not shown) that converts an electrical signal representing sound (hereinafter referred to as an "acoustic signal") into physical sound, that is, air vibrations.

[0017] The network interface 15 is configured with connectors conforming to parallel connection specifications or serial connection specifications. The network interface 15 has the function of performing communication control according to each standard and connecting to a network N (shown in FIG. 2) via a telephone line, thereby connecting the medical support device 10 to the network. The network interface 15 is an example of a communication unit.

[0018] Next, a system to which the medical support device 10 is applied will be described. FIG. 2 is a schematic diagram showing the configuration of a medical support system in which the medical support device 10 is provided.

[0019] FIG. 2 shows a medical support system 1 equipped with a medical support device 10. The medical support system 1 includes the medical support device 10 shown in FIG. 1, one or more diagnostic imaging devices 20, one or more image servers 30, and one or more information acquisition systems 40. The medical support device 10, diagnostic imaging devices 20, image servers 30, and information acquisition systems 40 are connected to each other via a network N so that they can communicate with each other. This connection can be an electrical connection via an electronic network. Here, the electronic network refers to a general information and communication network using electrical communication technology, and includes wireless / wired hospital-based local area networks (LANs) and the Internet, as well as telephone communication lines, optical fiber communication networks, cable communication networks, satellite communication networks, and the like.

[0020] The imaging diagnostic device 20 is a device for generating medical images, and includes an X-ray diagnostic device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, a nuclear medicine diagnostic device, an ultrasound diagnostic device, etc. The imaging diagnostic device 20 includes an imaging device 21 and an image processing device 22. The imaging device 21 acquires data that forms the basis of medical image data. The image processing device 22 has a general computer configuration and controls the operation of the imaging device 21 to acquire data from the imaging device 21, process the data, and generate medical image data. When the imaging diagnostic device 20 is an X-ray diagnostic device, the imaging device 21 includes an X-ray tube, an X-ray detector, etc. When the imaging diagnostic device 20 is an X-ray CT device, an MRI device, or a nuclear medicine diagnostic device, the imaging device 21 is a so-called gantry device. When the imaging diagnostic device 20 is an ultrasound diagnostic device, the imaging device 21 is a so-called ultrasound probe.

[0021] The image processing device 22 has a general configuration of a computer, for example, the image processing device 22 includes a processing circuit, a memory, an input interface, a display, and a network interface (not shown).

[0022] The image server 30 has a general computer configuration. For example, the image server 30 has a processing circuit, a memory circuit, an input interface, a display, and a network interface (all not shown). The configurations of the processing circuit, memory circuit, input interface, display, and network interface of the image server 30 are the same as the configurations of the processing circuit 11, memory circuit 12, input interface 13, display 14, and network interface 15 shown in FIG. 1, so a description thereof will be omitted.

[0023] The image server 30 is, for example, a DICOM (Digital Imaging and Communications in Medicine) server, and is connected to devices such as the image diagnostic apparatus 20 so as to be able to transmit and receive data via a network N. The image server 30 manages medical image data such as CT image data generated by the image diagnostic apparatus 20 as a DICOM file.

[0024] The information acquisition system 40 acquires information about the interior and exterior of a vehicle and / or information about detection of a person's fall. The information about the interior and exterior of a vehicle is collected by sensors mounted on vehicles such as ordinary vehicles involved in traffic accidents, self-driving cars, and connected cars. The information about detection of a person's fall is collected by a smart home, a monitoring system, etc.

[0025] FIG. 3 is a block diagram showing an example of the functions of the medical support device 10. As shown in FIG.

[0026] 3, the processing circuitry 11 realizes an external force information acquisition function F1, a medical support information generation function F2, and an output control function F3 by reading and executing a computer program stored in the storage circuitry 12 or directly incorporated in the processing circuitry 11. Hereinafter, a case where the functions F1 to F3 function as software by executing a computer program will be described as an example, but all or part of the functions F1 to F3 may be realized by a circuit such as an ASIC. Furthermore, all or part of the functions F1 to F3 may be realized by the image processing device 22 of the image diagnostic device 20 or the image server 30.

[0027] The external force information acquisition function F1 includes a function for acquiring external force information related to an external force applied to the subject. Note that the external force information acquisition function F1 is an example of an external force information acquisition unit.

[0028] In the case of high-energy trauma (e.g., a fall from a height or a car accident at a certain speed or higher), diagnosis is made with reference to the trauma initial care guidelines. In the second stage (secondary survey) of the three-stage interpretation of the trauma initial care guidelines, diagnosis is made by systematically searching for damage throughout the body. Based on the mechanism of injury (the cause and circumstances leading to the injury), damage is searched for by estimating the input position, input direction, and strength (magnitude) of the energy (i.e., external force) applied to the subject, who is the subject of medical treatment, as external force information. However, in situations where it is difficult for the subject to describe the mechanism of injury, this can easily lead to pitfalls. Therefore, the external force information acquisition function F1 acquires vehicle interior and exterior information and fall detection information from the information acquisition system 40, and acquires external force information based on the interior and exterior information and fall detection information.

[0029] First, the external force information acquisition function F1 can acquire information about the interior and exterior of a vehicle at the time of a traffic accident as image data from the information acquisition system 40 using an event data recorder (EDR) for traffic accident analysis. The EDR's technical requirements also include the physical size and position classification of the vehicle's occupants. For example, image data of the accident vehicle in an undamaged state is pre-stored in the memory circuit 12 for each vehicle model, and the image data of the accident vehicle and external force information of the occupants of the accident vehicle (i.e., the input position, input direction, and strength (magnitude) of the external force) are associated and pre-stored in the memory circuit 12. The external force information acquisition function F1 compares the image data of the accident vehicle that is the target of medical treatment with pre-stored image data of an undamaged vehicle of the same vehicle model to acquire external force information on the occupants of the accident vehicle.

[0030] Second, the external force information acquisition function F1 can acquire fall detection information in the event of an indoor fall accident from the information acquisition system 40 as image data indicating the posture of the fall when a fall is detected by an AI camera (fall detection camera) installed indoors. For example, image data for each fall posture of the person who has fallen is pre-stored in the memory circuitry 12, and the image data for each fall posture is associated with the external force information of the person who has fallen (i.e., the input position, input direction, and strength (magnitude) of the external force) and pre-stored in the memory circuitry 12. The external force information acquisition function F1 compares the image data of the person who is the target of medical treatment with the pre-stored image data for each fall posture to acquire external force information for the person who has fallen. Note that the external force information acquisition function F1 may also acquire fall detection information in the event of an indoor fall accident as information from a mobile terminal or wearable device.

[0031] The medical support information generation function F2 includes a function for generating medical support information for supporting medical treatment on a subject based on the external force information acquired by the external force information acquisition function F1. The medical support information generation function F2 includes a superimposition information generation function F21, an examination information generation function F22, a damaged site information generation function F23, and a treatment information generation function F24.

[0032] The superimposition information generation function F21 includes a function of generating superimposition information in which external force information is added to medical image data, that is, superimposition information in which symbols and / or characters indicating the external force information are superimposed on the medical image data, as medical support information. Also, for example, the superimposition information generation function F21 generates superimposition information that is an acoustic signal representing the external force information.

[0033] The examination information generation function F22 includes a function for generating examination information related to at least one of the necessity of an examination and the examination order (imaging plan, imaging range, etc.) as medical support information based on the external force information acquired by the external force information acquisition function F1 (third embodiment, described later). The damaged site information generation function F23 includes a function for generating damaged site information for identifying a damaged site in the subject as medical support information based on the external force information acquired by the external force information acquisition function F1 (fourth embodiment, described later). The treatment information generation function F24 generates treatment information representing a treatment plan (treatment plan, treatment or rehabilitation period) for the subject as medical support information based on the external force information acquired by the external force information acquisition function F1 (fifth embodiment, described later). The medical support information generation function F2 is an example of a medical support information generation unit.

[0034] The output control function F3 includes a function for controlling the output of medical support information generated by the medical support information generation function F2 to an output unit. Specifically, the output control function F3 controls the output of medical support information from the display 14 and the output of medical support information from a speaker (not shown). The output control function F3 is an example of an output control unit.

[0035] Next, a medical support method in the medical support device 10 will be described.

[0036] Figure 4 is a flowchart showing a method for processing a medical image file. In Figure 4, the symbols "ST" followed by numbers indicate each step in the flowchart. Here, in Figure 4, the medical support information will be described as superimposed information in which external force information is added to CT image data as medical image data.

[0037] The external force information acquisition function F1 acquires CT image data of a subject who is the subject of medical treatment from the image processing device 22 of the image diagnostic device 20 (step ST1). In step ST1, the external force information acquisition function F1 acquires CT image data including multiple CT images obtained by whole-body imaging of the subject. The external force information acquisition function F1 acquires external force information related to an external force applied to the subject (step ST2).

[0038] The superimposition information generation function F21 of the medical support information generation function F2 generates medical support information for supporting medical treatment on the subject based on the external force information acquired in step ST2 (step ST3). In step ST3, the superimposition information generation function F21 acquires CT image data for display from the CT image data acquired in step ST1 (step ST31), and adds external force information to the CT image data for display, i.e., generates superimposition information in which symbols and / or characters indicating the external force information are superimposed on the medical image data as medical support information (step ST32).

[0039] The output control function F3 controls the output of the superimposition information generated in step ST32 from the display 14 (step ST4).

[0040] 5 to 11 are diagrams showing first to sixth display examples of the superimposition information in step ST4, respectively.

[0041] Figure 5(A) shows superimposed information in which external force information is superimposed on 2D CT image data. The external force information is represented by an arrow symbol. The position of the arrowhead indicates the input position (body surface) of the external force. The direction of the arrow indicates the input direction of the external force at the input position. The color within the arrow frame (corresponding to the gradient bar in Figure 5(A)) indicates the strength (absolute value) of the external force at the input position. The length of the arrow indicates the strength of the external force at the input position (relative value compared to that at other input positions). The color consists of hue, saturation, and brightness, and the difference in absolute value of the external force strength can be expressed using at least one of the hue, saturation, and brightness within the arrow frame. In this way, because the arrowhead is located near the subject's body surface, external force information indicating the external force applied to the subject can be superimposed on the 2D CT image without interfering with the operator's damage detection. In the display of FIG. 5(A), the cross-sectional position of the displayed superimposed information can be changed in accordance with an operation performed via the input interface 13.

[0042] According to the display in Figure 5(A), the external force applied to the subject is superimposed on the two-dimensional CT image as external force information (arrows), allowing an operator such as a doctor to search for damage while visually checking the external force information as a guide.

[0043] FIG. 5(B) shows superimposed information in which external force information is superimposed on three-dimensional CT image data. The external force information is represented by an arrow symbol. The position of the arrowhead, the direction of the arrow, the color, and the length of the arrow have the same meanings as those shown in FIG. 5(A). In this way, external force information indicating the external force applied to the subject can be superimposed on the three-dimensional CT image so as not to interfere with the operator's damage search. In the display of FIG. 5(B), the projection direction of the displayed superimposed information can also be changed in accordance with operations performed via the input interface 13.

[0044] According to the display in Figure 5(B), the external force applied to the subject is superimposed on the three-dimensional CT image as external force information (arrows), allowing the operator to visually check the external force information as a guide while searching for damage.

[0045] Figure 6 shows superimposed information in which external force information is superimposed on three-dimensional CT image data. Figure 6 also shows an example in which, in addition to the display of the first external force information shown in Figure 5(B), second external force information is also superimposed on the image of the subject included in the CT image. The second external force information can be obtained from the external force propagation distribution inside the body estimated using a simulation or a human phantom based on the first external force information. In the display shown in Figure 6, the second external force information superimposed on the three-dimensional CT image may interfere with the operator's damage detection. Therefore, by setting the area where the operator is performing damage detection as a hidden area F on the CT image data image, the output control function F3 can hide the display of external force information in the hidden area F. Alternatively, the display of external force information (i.e., arrows) can be hidden in the area surrounding the displayed mouse pointer (a circular or square area of ​​a certain length centered on the mouse pointer), or the display of external force information can be hidden only near the center of the medical image data. In the display of Fig. 6, the projection direction of the displayed superimposed information can also be changed following an operation via the input interface 13. Also, Fig. 6 is based on three-dimensional CT image data, but the same applies to the case where two-dimensional CT image data is used as the basis.

[0046] According to the display in Figure 6, the external forces applied to the subject are superimposed on the CT image as first and second external force information (arrows), allowing the operator to search for damage while visually checking the first and second external force information as a guide.

[0047] FIG. 7 shows superimposed information in which external force information is superimposed on three-dimensional CT image data. The external force information is represented by an arrow symbol and the letters "(1)" and "(2)." The position of the arrowhead, the direction of the arrow, the color, and the length of the arrow are the same as those shown in FIG. 5(A). The display example in FIG. 7 differs from the display example in FIG. 5(B) in that it shows a display example in which multiple external forces, for example, two external forces, are applied to the subject. In other words, FIG. 7 illustrates a case in which multiple collisions are performed, and the external force information of each collision is displayed so that it can be distinguished. The letter "(1)" represents the first external force information, and the letter "(2)" represents the second external force information. In the display in FIG. 7, the projection direction of the displayed superimposed information can also be changed following operations via the input interface 13. Furthermore, while FIG. 7 is based on three-dimensional CT image data, the same applies to a case in which two-dimensional CT image data is used as the basis.

[0048] According to the display in Figure 7, multiple external forces applied to the subject multiple times are superimposed on the CT image as multiple distinguishable external force information (arrows), allowing the operator to perform damage detection while visually checking the multiple external force information as a guide.

[0049] FIG. 8 shows superimposed information in which external force information is superimposed on three-dimensional CT image data. The external force information is expressed as a symbolic frame around the text and the text "A strong external force may have been applied to the seventh rib on the right side of the chest from a lower position." In the display of FIG. 8, the projection direction of the CT image data in the displayed superimposed information can also be changed in accordance with an operation via the input interface 13. Also, while FIG. 8 is based on three-dimensional CT image data, the same applies when two-dimensional CT image data is used as the basis.

[0050] According to the display in Figure 8, the external force applied to the subject is superimposed on the CT image as external force information (character string), allowing the operator to perform damage search while visually checking the external force information as a guide.

[0051] The right side of FIG. 9 shows the order (priority) of diagnoses performed by an operator such as a doctor at the multiple imaging regions A1 to A4 on the left side. The left side of FIG. 9 shows the positions of the multiple imaging regions A1 to A4. The superimposition information generation function F21 can determine the order of diagnoses (1 to 4) in addition to the above-mentioned superimposition information based on the external force information. For example, the superimposition information generation function F21 can set a high priority for the imaging region A1 that is close to the external force input position, and a low priority for the imaging region A4 that is far from the external force input position. Note that for imaging positions that include important organs (e.g., the brain, the heart), the order of diagnoses can also be determined by adding weights in addition to distance.

[0052] The display in FIG. 9 can provide the order of diagnosis, allowing the operator to search for damage starting from imaging regions that have a high priority for life support.

[0053] FIG. 10(A) shows superimposed information in which external force information is superimposed on two-dimensional CT image data. The external force information is represented by arrows as symbols and deformation lines (dashed lines) that represent the shape after deformation. The outer deformation lines indicate the deformation of the skin based on the external force information. The inner deformation lines indicate damage to organs based on the external force information as internal reaching energy (estimated value). Note that in the display of FIG. 10, the cross-sectional position of the displayed superimposed information can also be changed following operations via the input interface 13. Also, while FIG. 10(A) is based on two-dimensional CT image data, the same applies when three-dimensional CT image data is used as the basis.

[0054] The display in Fig. 10(A) visualizes the deformation of the skin and organs based on external force information, allowing the operator to search for damage while visually checking the dashed lines after deformation as a guide. The superimposition information generation function F21 may also generate image data in which the CT image data has been corrected according to the deformation lines, as shown in Fig. 10(B).

[0055] As described above, the medical support device 10 according to the first embodiment of the medical support system 1 uses internal and external information and fall detection information from the information acquisition system 40 (shown in FIG. 2 ) to allow the operator to visually (or audibly) confirm the external force applied to the subject while performing damage detection. This improves efficiency compared to damage detection that requires the operator to repeatedly confirm the mechanism of injury with the subject. Furthermore, the medical support device 10 according to the first embodiment does not require the subject to explain the mechanism of injury, thereby avoiding potential pitfalls. In other words, the medical support device 10 according to the first embodiment can generate superimposed information based on external force information on the subject based on the internal and external information and fall detection information from the information acquisition system 40 and output it as shown in FIGS. 5 to 10 . This allows the operator performing the diagnosis (including damage detection) of the subject to receive effective medical support information for medical treatment of the subject.

[0056] (Second embodiment) The method for generating superimposition information by the superimposition information generation function F21 shown in Fig. 3 is not limited to the method described above. For example, the superimposition information generation function F21 can generate superimposition information based on medical image data. This case will be described below.

[0057] FIG. 11 is a schematic diagram showing the configuration of a medical support system equipped with a medical support device according to the second embodiment.

[0058] Fig. 11 shows a medical support system 1A equipped with a medical support device 10. The medical support system 1A includes the medical support device 10 shown in Fig. 1, one or more diagnostic imaging devices 20, and one or more image servers 30. The medical support device 10, diagnostic imaging devices 20, and image servers 30 are connected to each other via a network N so as to be able to communicate with each other. This connection can be an electrical connection via an electronic network, or the like.

[0059] The medical support system 1A shown in Fig. 11 has a configuration in which the information acquisition system 40 is removed from the medical support system 1 shown in Fig. 2. In Fig. 11, the same members as those shown in Fig. 2 are denoted by the same reference numerals and their description will be omitted.

[0060] In the medical support system 1A shown in Fig. 11, the superimposition information generation function F21 shown in Fig. 3 performs processing to generate superimposition information based on medical image data, for example, CT image data. This processing may use, for example, a lookup table (LUT) that associates CT image data with superimposition information. This processing may also use machine learning. Furthermore, deep learning using a multilayer neural network such as a convolutional neural network (CNN) or a convolutional deep belief network (CDBN) may also be used as the machine learning.

[0061] Here, an example is shown in which the superimposition information generation function F21 includes a neural network Na and generates superimposition information based on CT image data using deep learning. That is, the superimposition information generation function F21 generates superimposition information for a subject who is the subject of a medical procedure by inputting CT image data of the subject to a trained model for generating superimposition information based on CT image data.

[0062] FIG. 12 is an explanatory diagram showing an example of data flow during learning.

[0063] The superimposition information generation function F21 sequentially updates the parameter data Pb by learning from a large amount of input training data. The training data consists of a combination of CT image data Q1, Q2, Q3, ... as training input data and superimposition information S1, S2, S3, .... The CT image data Q1, Q2, Q3, ... constitute a training input data group Q. The superimposition information S1, S2, S3, ... constitute a training output data group S. The superimposition information S1, S2, S3, ... may be superimposition information relating to the corresponding CT image data Q1, Q2, Q3, ....

[0064] The superimposition information generation function F21 performs so-called learning, updating the parameter data Pb each time training data is input so that the results of processing CT image data Q1, Q2, Q3, ... using the neural network Nb approach the superimposition information S1, S2, S3, .... Generally, when the rate of change in the parameter data Pb converges within a threshold, learning is determined to be complete. Hereinafter, the parameter data Pb after learning will be referred to as learned parameter data Pb'.

[0065] It should be noted that the type of training input data should match the type of input data during operation shown in Fig. 12. For example, if the input data during operation is CT image data of the subject's head, the training input data group Q during learning should also be CT image data of the head.

[0066] Furthermore, the "image data" includes raw data generated by the diagnostic imaging device 20 (shown in FIG. 11). That is, the input data to the neural network Nb may be raw data before scan conversion.

[0067] FIG. 13 is an explanatory diagram showing an example of data flow during operation.

[0068] During operation, the superimposition information generation function F21 inputs CT image data Q' of a subject who is the subject of medical treatment, and outputs superimposition information S' of the subject using the learned parameter data Pb'.

[0069] The neural network Nb and the trained parameter data Pb' constitute a trained model 11b. The neural network Nb is stored in the memory circuitry 12 in the form of a program. The trained parameter data Pb' may be stored in the memory circuitry 12, or in a storage medium connected to the medical support device 10 via the network N. In this case, the superimposition information generation function F21 realized by the processor of the processing circuitry 11 reads and executes the trained model 11b from the memory circuitry 12, thereby generating superimposition information as medical support information based on the CT image data. The trained model 11b may be constructed using an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0070] In addition, in order to improve the accuracy of the superposition information S' output by the superposition information generation function F21, in addition to the CT image data, external image (optical image) data showing trauma to the subject, and identification information including at least one of the subject's height, weight, medical history, and medical history of blood relatives may be used as input data.

[0071] In this case, during learning, the CT image data Q1, Q2, Q3, ..., as training input data, including the appearance image data and identification information of each subject, are also input to the neural network Nb as training input data. During operation, the superimposition information generation function F21 inputs the CT image data Q' of the subject who is the subject of medical treatment together with the appearance image data and identification information of the subject to the trained model 11b read from the storage circuitry 12, and outputs superimposition information S' regarding the subject. Using the CT image data and the appearance image data and identification information of the subject as input data makes it possible to generate trained parameter data Pb' that has been trained according to the injury and type of the subject, thereby improving the accuracy of medical treatment such as diagnosis compared to when only CT image data is used as input data.

[0072] As described above, the medical support device 10 according to the second embodiment of the medical support system 1A allows the operator to visually (or audibly) check the external force applied to the subject and perform damage detection without relying on internal and external information or fall detection information from the information acquisition system 40 (shown in FIG. 2). This improves efficiency compared to damage detection that requires repeated confirmation of the mechanism of injury with the subject. Furthermore, the medical support device 10 according to the second embodiment can generate superimposed information based on external force information derived from medical image data of the subject, without relying on internal and external information or fall detection information from the information acquisition system 40, and output the information as shown in FIGS. 5 to 10. This allows the operator diagnosing the subject to be provided with effective medical support information for medical treatment of the subject. Furthermore, the medical support device 10 according to the second embodiment can generate superimposed information from medical image data with higher accuracy, and can generate superimposed information easily because there is no need to acquire external force information during diagnosis.

[0073] (Third embodiment) The examination information generation function F22 of the medical support information generation function F2 shown in Figure 3 will be described. The examination information generation function F22 generates examination information related to at least one of the necessity of an examination and the examination order (imaging plan, imaging range, etc.) as medical support information based on the external force information (including the external force information included in the superimposed information) acquired by the external force information acquisition function F1. Head CT examinations involve a relatively high amount of radiation exposure among radiological examinations, and it has been reported that radiation exposure from head CT examinations poses a risk of cancer in infants. Therefore, unnecessary head CT examinations should be avoided as much as possible. Therefore, examination information indicating the necessity of an examination such as a CT examination is generated based on the age of the patient who is the subject of medical treatment (e.g., whether the patient is under 2 years old or between 2 and 18 years old), the subject's level of consciousness, whether or not there is loss of consciousness, and the mechanism of injury. The examination information generation function F22 also generates examination information indicating the examination order using information about the external force applied to the subject.

[0074] In the medical support system 1A shown in FIG. 11, the examination information generation function F22 shown in FIG. 3 performs processing to generate an examination order from the examination information based on external force information (including external force information included in the superimposed information). This processing may use, for example, a lookup table that associates the superimposed information with the examination order. Machine learning may also be used for this processing. Furthermore, deep learning using a multilayer neural network such as a CNN or a convolutional deep belief network may also be used as the machine learning.

[0075] Here, an example is shown in which the test information generation function F22 includes a neural network Nc and generates an examination order based on superimposed information including external force information using deep learning. That is, the test information generation function F22 generates an examination order for a subject by inputting external force information of the subject who is the subject of medical treatment into a trained model for generating an examination order based on superimposed information including external force information.

[0076] FIG. 14 is an explanatory diagram showing an example of data flow during learning.

[0077] The examination information generation function F22 sequentially updates the parameter data Pc by learning from a large amount of input training data. The training data consists of a combination of superimposed information S1, S2, S3, ... as training input data and examination orders T1, T2, T3, .... The superimposed information S1, S2, S3, ... constitutes a training input data group S. The examination orders T1, T2, T3, ... constitute a training output data group T. It is preferable that the examination orders T1, T2, T3, ... are examination orders related to the corresponding superimposed information S1, S2, S3, ....

[0078] The test information generation function F22 performs so-called learning, updating the parameter data Pc each time training data is input, so that the result of processing the superposition information S1, S2, S3, ... by the neural network Nc approaches the test orders T1, T2, T3, .... Generally, when the rate of change in the parameter data Pc converges within a threshold, learning is determined to be complete. Hereinafter, the parameter data Pc after learning will be referred to as learned parameter data Pc'.

[0079] It should be noted that the type of training input data should match the type of input data during operation shown in Fig. 14. For example, if the input data during operation is superimposed information including CT image data of the subject's head, the training input data group S during learning should also be superimposed information including CT image data of the head.

[0080] Furthermore, the "image data" includes raw data generated by the diagnostic imaging device 20 (shown in FIG. 11). That is, the input data to the neural network Nc may be raw data before scan conversion.

[0081] FIG. 15 is an explanatory diagram showing an example of data flow during operation.

[0082] During operation, the examination information generation function F22 inputs superimposed information S' of a subject who is the subject of medical treatment, and outputs an examination order T' for the subject using the learned parameter data Pc'.

[0083] The neural network Nc and the trained parameter data Pc' constitute a trained model 11c. The neural network Nc is stored in the memory circuitry 12 in the form of a program. The trained parameter data Pc' may be stored in the memory circuitry 12, or in a storage medium connected to the medical support device 10 via the network N. In this case, the examination information generation function F22 realized by the processor of the processing circuitry 11 reads and executes the trained model 11c from the memory circuitry 12 to generate an examination order based on the superimposed information. The trained model 11c may be constructed using an integrated circuit such as an ASIC or FPGA.

[0084] In addition, in order to improve the accuracy of the test order T' output by the test information generation function F22, in addition to the superimposition information, external image (optical image) data showing trauma on the subject, and identification information including at least one of the subject's height, weight, medical history, and medical history of blood relatives may be used as input data.

[0085] In this case, during learning, the superimposition information S1, S2, S3, ..., the appearance image data and identification information of each subject are also input as training input data to the neural network Nc. During operation, the examination information generation function F22 inputs the superimposition information S' of the subject who is the subject of medical treatment together with the appearance image data and identification information of the subject to the trained model 11c read from the storage circuitry 12, and outputs an examination order T' for the subject. Using the superimposition information and the appearance image data and identification information of the subject as input data makes it possible to generate trained parameter data Pc' that has been trained according to the injury and type of the subject, thereby improving the accuracy of medical treatment such as diagnosis compared to when only superimposition information is used as input data.

[0086] As described above, according to the medical support device 10 of the third embodiment in the medical support system 1A, in addition to the effects of the medical support device 10 of the second embodiment, by generating and outputting examination information (necessity of examination and examination order) based on external force information of the subject, it is possible to provide an operator who diagnoses the subject with medical support information that is effective for medical treatment of the subject.

[0087] (Fourth embodiment) The damaged site information generation function F23 of the medical support information generation function F2 shown in Fig. 3 will be described. The damaged site information generation function F23 generates, as medical support information, damaged site information that identifies a damaged site in the subject, based on the external force information (including the external force information included in the superimposition information) acquired by the external force information acquisition function F1.

[0088] In the medical support system 1A shown in FIG. 11, the damaged site information generation function F23 shown in FIG. 3 performs processing to generate damaged site information based on external force information (including external force information included in superimposition information). This processing may use, for example, a lookup table that associates superimposition information with damaged site information. Machine learning may also be used for this processing. Furthermore, deep learning using a multilayer neural network such as a CNN or a convolutional deep belief network may be used as the machine learning.

[0089] Here, an example is shown in which the damaged site information generation function F23 includes a neural network Nd and generates damaged site information based on superimposed information including external force information using deep learning. That is, the damaged site information generation function F23 generates damaged site information of a subject who is the target of medical treatment by inputting external force information of the subject to a trained model for generating damaged site information based on superimposed information including external force information.

[0090] FIG. 16 is an explanatory diagram showing an example of data flow during learning.

[0091] The damaged part information generation function F23 sequentially updates the parameter data Pd by learning from a large amount of input training data. The training data consists of a combination of superimposed information S1, S2, S3, ... as training input data and damaged part information U1, U2, U3, .... The superimposed information S1, S2, S3, ... constitutes a training input data group S. The damaged part information U1, U2, U3, ... constitutes a training output data group U. It is preferable that the damaged part information U1, U2, U3, ... is damaged part information related to the corresponding superimposed information S1, S2, S3, ....

[0092] The damaged part information generation function F23 performs so-called learning, updating the parameter data Pd each time training data is input so that the results of processing the superposition information S1, S2, S3, ... by the neural network Nd approach the damaged part information U1, U2, U3, .... Generally, when the rate of change in the parameter data Pd converges within a threshold, learning is determined to be complete. Hereinafter, the parameter data Pd after learning will be referred to specifically as learned parameter data Pd'.

[0093] It should be noted that the type of training input data should match the type of input data during operation shown in Fig. 16. For example, if the input data during operation is superimposed information including CT image data of the subject's head, the training input data group S during learning should also be superimposed information including CT image data of the head.

[0094] Furthermore, the "image data" includes raw data generated by the diagnostic imaging device 20 (shown in FIG. 11). That is, the input data to the neural network Nd may be raw data before scan conversion.

[0095] FIG. 17 is an explanatory diagram showing an example of data flow during operation.

[0096] During operation, the damaged site information generation function F23 inputs superimposed information S' of a subject who is the subject of medical treatment, and outputs damaged site information U' of the subject using the learned parameter data Pd'.

[0097] The neural network Nd and the learned parameter data Pd' constitute a learned model 11d. The neural network Nd is stored in the memory circuitry 12 in the form of a program. The learned parameter data Pd' may be stored in the memory circuitry 12, or may be stored in a storage medium connected to the medical support device 10 via the network N. In this case, the damaged site information generation function F23 realized by the processor of the processing circuitry 11 reads and executes the learned model 11d from the memory circuitry 12, thereby generating damaged site information based on the superimposed information. The learned model 11d may be constructed using an integrated circuit such as an ASIC or FPGA.

[0098] In addition, in order to improve the accuracy of the damaged area information U' output by the damaged area information generation function F23, in addition to the superimposition information, external image (optical image) data showing the subject's injury and identification information including at least one of the subject's height, weight, medical history, and medical history of blood relatives may be used as input data.

[0099] In this case, during learning, the superimposition information S1, S2, S3, ..., and the appearance image data and identification information of each subject are also input as training input data to the neural network Nd. During operation, the damaged site information generation function F23 inputs the superimposition information S' of the subject who is the target of medical treatment together with the appearance image data and identification information of the subject to the trained model 11d read from the storage circuitry 12, and outputs damaged site information U' regarding the subject. Using the superimposition information and the appearance image data and identification information of the subject as input data makes it possible to generate trained parameter data Pd' that has been trained according to the injury and type of the subject, thereby improving the accuracy of medical treatment such as diagnosis compared to when only superimposition information is used as input data.

[0100] FIG. 18 is a diagram showing an example of display of damaged site information in step ST4. FIG. 18 shows superimposed information in which external force information is superimposed on three-dimensional CT image data. FIG. 18 shows an example in which damaged area information (broken line) is superimposed in addition to the display of external force information shown in FIG. 5(B). In the display of FIG. 18, the projection direction of the displayed superimposed information can also be changed following an operation via the input interface 13. Also, while FIG. 18 is based on three-dimensional CT image data, the same applies to the case where two-dimensional CT image data is used as the basis.

[0101] According to the display of FIG. 18, by superimposing the damaged site information on the CT image, the operator can perform damage search while visually checking the damaged site information as a guide.

[0102] As described above, according to the medical support device 10 of the fourth embodiment in the medical support system 1A, in addition to the effects of the medical support device 10 of the second embodiment, by generating and outputting damaged area information based on external force information of the subject, it is possible to provide an operator who diagnoses the subject with medical support information that is effective for medical treatment of the subject.

[0103] (Fifth embodiment) The following describes the treatment information generation function F24 of the medical support information generation function F2 shown in Fig. 3. The treatment information generation function F24 generates, as medical support information, treatment information representing a treatment plan (treatment plan, treatment / rehabilitation period) for the subject based on the external force information (including external force information included in the superimposition information) acquired by the external force information acquisition function F1.

[0104] In the medical support system 1A shown in FIG. 11, the treatment information generation function F24 shown in FIG. 3 performs processing to generate treatment information based on external force information (including external force information included in superimposition information). This processing may use, for example, a lookup table that associates superimposition information with treatment information. Machine learning may also be used for this processing. Furthermore, deep learning using a multilayer neural network such as a CNN or a convolutional deep belief network may also be used as the machine learning.

[0105] Here, an example is shown in which the treatment information generation function F24 includes a neural network Ne and generates treatment information based on superimposed information including external force information using deep learning. That is, the treatment information generation function F24 generates treatment information for a subject by inputting external force information of the subject who is the target of medical treatment into a trained model for generating treatment information based on superimposed information including external force information.

[0106] FIG. 19 is an explanatory diagram showing an example of data flow during learning.

[0107] The treatment information generation function F24 sequentially updates the parameter data Pe by learning from a large amount of input training data. The training data consists of a combination of superimposed information S1, S2, S3, ... as training input data and treatment information V1, V2, V3, .... The superimposed information S1, S2, S3, ... constitutes a training input data group S. The treatment information V1, V2, V3, ... constitutes a training output data group V. It is preferable that the treatment information V1, V2, V3, ... be treatment information related to the corresponding superimposed information S1, S2, S3, ....

[0108] The treatment information generation function F24 performs so-called learning, updating the parameter data Pe each time training data is input so that the result of processing the superposition information S1, S2, S3, ... using the neural network Ne approaches the treatment information V1, V2, V3, .... Generally, when the rate of change in the parameter data Pe converges within a threshold, learning is determined to be complete. Hereinafter, the parameter data Pe after learning will be referred to as learned parameter data Pe'.

[0109] It should be noted that the type of training input data should match the type of input data during operation shown in Fig. 19. For example, if the input data during operation is superimposed information including CT image data of the subject's head, the training input data group S during learning should also be superimposed information including CT image data of the head.

[0110] Furthermore, the "image data" includes raw data generated by the diagnostic imaging device 20 (shown in FIG. 11). That is, the input data to the neural network Ne may be raw data before scan conversion.

[0111] FIG. 20 is an explanatory diagram showing an example of data flow during operation.

[0112] During operation, the treatment information generation function F24 inputs superimposed information S' of a subject who is the subject of medical treatment, and outputs treatment information V' of the subject using the learned parameter data Pe'.

[0113] The neural network Ne and the learned parameter data Pe' constitute a learned model 11e. The neural network Ne is stored in the memory circuitry 12 in the form of a program. The learned parameter data Pe' may be stored in the memory circuitry 12, or may be stored in a storage medium connected to the medical support device 10 via the network N. In this case, the treatment information generation function F24 realized by the processor of the processing circuitry 11 reads and executes the learned model 11e from the memory circuitry 12, thereby generating treatment information based on the superimposed information. The learned model 11e may be constructed using an integrated circuit such as an ASIC or FPGA.

[0114] In addition, in order to improve the accuracy of the treatment information V' output by the treatment information generation function F24, in addition to the superimposition information, external image (optical image) data showing the subject's trauma and identification information including at least one of the subject's height, weight, medical history, and medical history of blood relatives may be used as input data.

[0115] In this case, during learning, the superimposed information S1, S2, S3, ..., as training input data, including the appearance image data and identification information of each subject, is also input to the neural network Ne as training input data. During operation, the treatment information generation function F24 inputs the superimposed information S' of the subject who is the target of medical treatment, along with the appearance image data and identification information of the subject, to the trained model 11e read from the storage circuitry 12, and outputs treatment information V' regarding the subject. Using the superimposed information and the appearance image data and identification information of the subject as input data makes it possible to generate trained parameter data Pe' that has been trained according to the injury and type of the subject, thereby improving the accuracy of medical treatment such as diagnosis compared to when only superimposed information is used as input data.

[0116] As described above, according to the medical support device 10 of the fifth embodiment in the medical support system 1A, in addition to the effects of the medical support device 10 of the second embodiment, by generating and outputting treatment information based on external force information of the subject, it is possible to provide an operator who diagnoses the subject with medical support information that is effective for medical treatment of the subject.

[0117] The external force information acquisition function F1 is an example of an external force information acquisition unit. The medical support information generation function F2 is an example of a medical support information generation unit. The output control function F3 is an example of an output control unit. The superimposition information generation function F21 is an example of a superimposition information generation unit. The examination information generation function F22 is an example of an examination information generation unit. The damaged site information generation function F23 is an example of a damaged site information generation unit. The treatment information generation function F24 is an example of a treatment information generation unit.

[0118] According to at least one of the embodiments described above, it is possible to appropriately and efficiently support medical treatment for a subject to which an external force has been applied.

[0119] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0120] 1,1A Medical Support System 10 Medical support equipment 11 Processing circuit 20 Diagnostic imaging equipment 30 Image Server F1 External force information acquisition function F2 Medical support information generation function F21 Overlay information generation function F22 Inspection information generation function F23 Damage site information generation function F24 Treatment information generation function F3 Output control function

Claims

1. a fall detection unit that detects a fall of the subject; an external force information acquiring unit that acquires external force information related to an external force applied to the subject by analyzing the fall detection information obtained by the fall detection unit; a medical image acquisition unit that acquires a medical image of the subject; a medical support information generating unit that generates medical support information that supports medical treatment on the subject, the medical support information being information in which the acquired external force information is added to the medical image; an output control unit that displays the generated medical support information on a display; A medical support device comprising:

2. an external force information acquisition unit that acquires external force information related to an external force applied to the subject; a medical support information generating unit that generates medical support information that supports medical treatment on the subject based on the acquired external force information; an output control unit that displays the generated medical support information on a display; Equipped with the medical support information generation unit generates the medical support information by combining the medical image of the subject and the external force information; when the output control unit superimposes the external force information on the medical image and displays the medical support information, the output control unit hides the display of the external force information in a non-display area set on the medical image or in a surrounding area of ​​a displayed mouse pointer. Medical support equipment.

3. the medical support information generation unit generates the medical support information by combining the medical image of the subject and the external force information. The medical support device according to claim 1 or 2.

4. the external force information acquisition unit acquires external force information related to an external force applied to the subject by inputting a medical image of the subject to a trained model trained based on a data set including a medical image and the external force information; The medical support device according to claim 2 .

5. the output control unit causes the generated medical support information to be displayed on a display. The medical support device according to claim 1 or 2.

6. the medical support information generation unit generates the medical support information by combining the medical image of the subject and the external force information; when the output control unit superimposes the external force information on the medical image and displays the medical support information, the output control unit hides the display of the external force information in a non-display area set on the medical image or in a surrounding area of ​​a displayed mouse pointer. The medical support device according to claim 1 .

7. the medical support information generation unit generates, as the medical support information, examination information related to at least one of whether an examination is necessary and an examination order, based on the acquired external force information. The medical support device according to claim 1 or 2.

8. the medical support information generation unit generates an examination order for the subject by inputting the external force information of the subject to a trained model trained based on a dataset including the external force information and an examination order among the examination information. The medical support device according to claim 7.

9. the medical support information generation unit generates, as the medical support information, damaged site information for identifying a damaged site in the subject, based on the acquired external force information. The medical support device according to any one of claims 1 to 8.

10. the medical support information generation unit generates information on the damaged part of the subject by inputting the external force information of the subject to a trained model trained based on a data set having the external force information and the damaged part information. The medical support device according to claim 9.

11. the medical support information generation unit generates, as the medical support information, treatment information representing a treatment plan for the subject based on the acquired external force information. The medical support device according to any one of claims 1 to 10.

12. the medical support information generation unit generates treatment information of the subject by inputting the external force information of the subject to a trained model trained based on a dataset having the external force information and the treatment information. The medical support device according to claim 11.

13. Computer, a fall detection unit that detects a fall of the subject; an external force information acquisition unit that acquires external force information related to an external force applied to the subject by analyzing the fall detection information obtained by the fall detection unit; a medical image acquisition unit that acquires a medical image of the subject; a medical support information generating unit that generates medical support information that supports medical treatment on the subject, the medical support information being information in which the acquired external force information is added to the medical image; an output control unit that outputs the generated medical support information; A program that functions as a

14. Computer, an external force information acquisition unit that acquires external force information related to an external force applied to the subject; a medical support information generating unit that generates medical support information that supports medical treatment on the subject based on the acquired external force information; an output control unit that displays the generated medical support information on a display; A program that functions as the medical support information generation unit generates the medical support information by combining the medical image of the subject and the external force information; when the output control unit superimposes the external force information on the medical image and displays the medical support information, the output control unit hides the display of the external force information in a non-display area set on the medical image or in a surrounding area of ​​a displayed mouse pointer. program.

Images