Control method for inspection support system and medical device

Abstract

This inspection assistance system assists an inspection that uses a medical image and has one or more processors having hardware, wherein the processor acquires a video signal that includes information about the medical image, and is able to output inspection assistance information for assisting an inspection by carrying out a recognition process on the basis of the video signal, and determines, on the basis of the video signal, whether the video signal is a video signal used for recognition process.

Description

Technical Field

[0001] The present invention relates to an inspection support system suitable for a device that performs AI determination. and control methods for medical devices It relates to.

Background Art

[0002] In recent years, in the medical field as well, medical systems that perform various determinations using AI (artificial intelligence) have been developed. For example, in colonoscopy, an endoscopic AI operation support system may be adopted in order to perform a high-quality endoscopic examination with less burden on the patient. It is a system that determines the current endoscopic examination situation by AI and presents support information suitable for that situation.

[0003] In order to enable highly accurate AI determination, it is necessary to use an endoscopic image suitable for AI determination. However, even if an endoscopic image suitable for AI determination has been captured, if a wiring error or the like occurs between the video processor device of the endoscopic system and a device that outputs operation support information or the like by AI based on the endoscopic image (hereinafter referred to as an inspection support device), AI determination using an appropriate endoscopic image will not be carried out.

[0004] As a technique for preventing such wiring errors, Japanese Patent Application Laid-Open No. 2010-268063 discloses a technique for authenticating a device of a connection partner in order to manage whether to permit information communication between electronic devices. In this proposal, when an image forming apparatus detects the connection of an external device, it acquires the external device ID of the external device, and permits the use of the external device when the externally stored external device ID and the acquired external device ID match.

[0005] By applying this technique to a video processor device and an inspection support device, it is possible to confirm that the destination device of the communication connection is correct.

Prior Art Documents

Patent Documents

[0006] [Patent Document 1] Japanese Patent Publication No. 2010-268063 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] However, two types of connections are used between the video processor and the inspection support device: a communication connection for transmitting identification and control signals, and a video connection for transmitting video signals of endoscopic images. Therefore, even if the technology described in Japanese Patent Application Publication No. 2010-268063 is applied to the connection between the video processor and the inspection support device to confirm that the destination device for the communication connection is correct, this does not guarantee that the destination device for the video signal is correct. Furthermore, the video processor may be configured to output video from multiple output ports for each image transmission format. In this case, even if the destination device for the video signal is correct, a video signal not intended for AI judgment may be transmitted to the inspection support device due to miswiring. This invention relates to an inspection support system that can check whether the video signal input to the inspection support device is an appropriate video signal. and control methods for medical devices The purpose is to provide. [Disclosure of the Invention] [Means for solving the problem]

[0008] An examination support system according to one aspect of the present invention has one or more processors having hardware, the processor acquires source device information relating to a source device via a first communication route connected to a source device which is the source of medical images obtained by photographing a subject, determines whether the connected source device is appropriate based on the source device information, acquires a video signal via a second communication route connected to the source device, and determines whether the video signal is Approval It is determined whether or not the video signal is for recognition processing, and recognition processing is performed based on the video signal. Inspect Outputs inspection support information to assist with the inspection.

[0009] A medical device control method according to one aspect of the present invention involves: an equipment information acquisition unit acquiring source device information related to a source device, which is the source of a medical image obtained by photographing a subject, via a first communication route connected to the source device; an input information confirmation unit determining whether the connected source device is appropriate based on the source device information; the input information acquisition unit acquiring a video signal via a second communication route connected to the source device; and the input information confirmation unit determining whether the video signal is appropriate based on the video signal. Approval It is determined whether or not the video signal is for recognition processing, and recognition processing is performed based on the video signal. Inspect Outputs inspection support information to assist with the inspection. [Effects of the Invention]

[0011] According to the present invention, it is possible to check whether the video signal input to the inspection support device is an appropriate video signal. [Brief explanation of the drawing]

[0012] [Figure 1] This is a configuration diagram showing an inspection support system according to the first embodiment of the present invention. [Figure 2] This is an explanatory diagram illustrating an example of the input qualification determination method used by the AI ​​information verification unit 13b. [Figure 3] This is an explanatory diagram illustrating an example of the input qualification determination method used by the AI ​​information verification unit 13b. [Figure 4] This flowchart shows the AI ​​judgment control by the AI ​​information verification unit 13b. [Figure 5] This flowchart shows the device legitimacy determination by the connection destination confirmation unit 13a. [Figure 6] This flowchart shows the device legitimacy determination by the connection destination confirmation unit 13a. [Figure 7] This is a block diagram showing a variation. [Modes for carrying out the invention]

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0014] (First Embodiment) FIG. 1 is a configuration diagram showing an inspection support system according to a first embodiment of the present invention. This embodiment enables checking whether a medical image suitable for AI determination is input to the inspection support device.

[0015] This embodiment applies the inspection support system to an endoscopic examination, but it is applicable not only to endoscopic examinations but also to various AI processes that employ AI determination. For example, it is applicable to AI processes that output endoscopic operation guidance information based on endoscopic images, AI processes that output control information for an automatically inserted endoscope based on endoscopic images, AI processes that output identification information for lesions or organ sites based on endoscopic images, and the like.

[0016] FIG. 1 shows the configuration of an inspection support device 10 that constitutes an inspection support system according to a first embodiment of the present invention, and a video processor device 20 that outputs information used for AI determination to the inspection support device.

[0017] The video processor device 20 includes a device information output unit 21 and an input information output unit 22. For example, an endoscope (not shown) that is inserted into a subject and captures an endoscopic image related to the body cavity may be connected to the video processor device 20, and an endoscopic image such as an endoscopic examination image may be supplied from this endoscope. The video processor device 20 can acquire not only endoscopic images but also various types of information. For example, the video processor device 20 may be configured to acquire information on whether an endoscope is connected to the video processor device 20 and whether an examination using the endoscope has been started. Further, the video processor device 20 may be configured to acquire information regarding switching of various observation modes of the endoscope, such as an NBI (narrow band observation) mode or an RDI (infrared light observation) mode.

[0018] The device information output unit 21 transmits the device information of the device that is the source of the input information of the inspection support device 10 (source device), that is, the video processor device 20 which is the connection destination of the inspection support device 10, via the first communication cable C1 that constitutes the first communication route. The source device information transmitted by the device information output unit 21 may include, for example, in addition to the ID information that identifies the video processor device 20, information regarding the operation of the video processor device 20 (operation information), information regarding the operation of the video processor device 20 (operation information), and information regarding the video format of the video signal output from the video processor device 20. The operation information and operation information in the source device information include information on whether an endoscope is connected to the video processor device 20, information on whether the inspection by the endoscope has been started, information on the switching of the observation mode of the endoscope, and the like.

[0019] The input information output unit 22 outputs the input information of the inspection support device 10 via the second communication cable C2 that constitutes the second communication route. That is, the input information output unit 22 outputs a video signal of an image including an image for AI determination, for example, an endoscope image such as an endoscope inspection image, as the input information of the inspection support device 10. The image based on the video signal output by the input information output unit 22 includes an endoscope image, which is a medical image, in a part of the image.

[0020] Although the first and second communication routes have been described as being constituted by the first communication cable C1 and the second communication cable C2 respectively, it is not limited thereto, and the first and second communication routes may be constituted by a wireless route.

[0021] The inspection support device 10 includes an equipment information acquisition unit 11, an input information acquisition unit 12, an input information confirmation unit 13, and an AI judgment unit 14. The entire inspection support device 10, or at least one of the equipment information acquisition unit 11, input information acquisition unit 12, input information confirmation unit 13, and AI judgment unit 14, may be composed of one or more processors using a CPU (Central Processing Unit) or FPGA (Field Programmable Gate Array), or they may operate according to a program stored in memory (not shown) to control each part, or some or all of the functions may be realized by hardware electronic circuits.

[0022] The device information acquisition unit 11 acquires source device information transmitted from the device information output unit 21 via the first communication route and outputs the acquired source device information to the input information confirmation unit 13. The input information acquisition unit 12 acquires input information transmitted from the input information output unit 22 via the second communication route and outputs the acquired input information to the input information confirmation unit 13.

[0023] The input information confirmation unit 13 includes a connection destination confirmation unit 13a and an AI information confirmation unit 13b. The connection destination confirmation unit 13a is provided with source device information from the device information output unit 21. Using the source device information, the connection destination confirmation unit 13a determines whether the connected device is the correct device as the source of the endoscopic image, which is input information for determining the content to be supported by the inspection support device 10, and confirms the legitimacy of the device.

[0024] The AI ​​information verification unit 13b receives input information from the input information output unit 22. The AI ​​information verification unit 13b performs an input suitability determination to determine whether the input information is suitable for a predetermined recognition process, such as AI judgment, and outputs the input suitability determination result to the AI ​​judgment unit 14.

[0025] The AI ​​determination unit 14 is provided with input information and the input qualification determination result from the AI ​​information confirmation unit 13b. Based on the input qualification determination result, the AI ​​determination unit 14 decides whether or not to use the input information for AI determination. That is, if the AI ​​determination unit 14 receives an input qualification determination result from the AI ​​information confirmation unit 13b indicating that the input information is suitable for recognition processing, it performs recognition processing such as AI determination based on the input information. The AI ​​determination unit 14 outputs the determination result of the AI ​​determination, which is the recognition result of the recognition processing, to the connection destination confirmation unit 13a. Also, if the AI ​​determination unit 14 receives an input qualification determination result from the AI ​​information confirmation unit 13b indicating that the input information is not suitable for recognition processing, it does not perform recognition processing such as AI determination based on the input information.

[0026] The AI ​​information verification unit 13b may also be configured to supply the input information to the AI ​​determination unit 14 only when it is given an input suitability determination result indicating that the input information is suitable for recognition processing.

[0027] When the connection destination confirmation unit 13a receives an AI judgment result from the AI ​​judgment unit 14, it decides whether or not to adopt the AI ​​judgment result based on the device legitimacy judgment result for the device that transmitted the input information used in the AI ​​judgment. That is, if the connection destination confirmation unit 13a confirms that the device is legitimate based on the device legitimacy judgment result, it adopts the AI ​​judgment result and outputs the AI ​​judgment result to the display unit 30. If the connection destination confirmation unit 13a confirms that the device is not legitimate based on the device legitimacy judgment result, it does not output the AI ​​judgment result to the display unit 30.

[0028] The display unit 30 displays information from the input information confirmation unit 13 on the display screen. When an AI judgment result is input, the display unit 30 displays the input AI judgment result on the display screen. The input information confirmation unit 13 may also provide the display unit 30 with the judgment result of equipment legitimacy and the judgment result of input eligibility and display them.

[0029] (Input eligibility check) Figures 2 and 3 are explanatory diagrams illustrating an example of the input qualification determination method used by the AI ​​information verification unit 13b.

[0030] The video processor device 20 generates a video signal in a predetermined video format, with video parameters such as resolution, color gamut, and frame rate appropriately set, by performing predetermined video signal processing on the imaging signal obtained by imaging with an endoscope (not shown). The video processor device 20 encodes the generated video signal using a predetermined codec such as MPEG-4 or H.265 / HEVC, containerizes the encoded video signal in various file formats such as AVI, MOV, WMV, and FLV, and transmits it via an output terminal that conforms to a predetermined transmission standard such as DVI or SDI. Note that the video format of the video signal that can be transmitted is determined according to the transmission standard.

[0031] If a video format unsuitable for AI judgment is used, reliable AI judgment results cannot be obtained. Furthermore, even if an image in a video format suitable for AI judgment is input to the video processor device 20, an image unsuitable for AI judgment may be transmitted to the inspection support device 10 due to wiring errors or other reasons. In this case, reliable AI judgment results cannot be obtained either.

[0032] Image P1 in Figure 2 shows an example of an image suitable for AI judgment by the AI ​​judgment unit 14, and is obtained by imaging the video signal output from the input information output unit 22. Image P1 has a display area P1R that displays the endoscopic image over a relatively wide area of ​​the entire image. Within this display area P1R, there is an endoscopic image P1Ra in which the shape of the effective image area is octagonal due to a mask area (filled area) with triangular corners. This endoscopic image P1Ra has a relatively high resolution. The AI ​​judgment unit 14 obtains the AI ​​judgment result by performing recognition processing on this endoscopic image P1Ra.

[0033] In contrast, image P2 in Figure 2 shows an example of an image that does not conform to the AI ​​judgment of the AI ​​judgment unit 14. For example, if the type of endoscope connected to the video processor device 20 is not appropriate, or if the settings of the video processor device 20 are incorrect, it is expected that image P2 will be obtained when the video signal from the input information output unit 22 is imaged. That is, image P2 includes an endoscopic image P1Rb in which the shape of the effective image area is octagonal due to a triangular mask area (filled area) at the four corners within a relatively wide display area P1R of the entire image. The aspect ratio of this endoscopic image P1Rb is different from that of the endoscopic image P1Ra. That is, the positions in the image based on the video signal, which is the input information, are different for endoscopic image P1Ra and endoscopic image P1Rb, and even if the AI ​​judgment unit 14 performs recognition processing on this endoscopic image P1Rb, it is not guaranteed that it will obtain a correct AI judgment result.

[0034] Furthermore, if the wrong endoscope is connected to the video processor device 20, or if there is a setting error, or if an image not intended for AI judgment, such as an image for a printer, is input to the examination support device 10, the resolution of the endoscope image P1Rb input to the examination support device 10 may be lower than the resolution of the endoscope image P1Ra. In this case, even if the AI ​​judgment unit 14 performs recognition processing on this endoscope image P1Rb, it is not guaranteed that a correct AI judgment result will be obtained. Moreover, if an image that should be output to another device (not shown) is mistakenly provided to the examination support device 10, another image P2a may be superimposed on a part of the endoscope image P1Rb by PIP (Picture-in-Picture). In this case as well, the AI ​​judgment unit 14 may not be able to obtain a correct AI judgment result.

[0035] The AI ​​information verification unit 13b may determine that the video signal is not the video signal for the recognition process if at least one of the following cases is different: the aspect ratio and image quality of the training medical image used to generate the inference model of the AI ​​determination unit 14 is different from the aspect ratio and image quality of the input medical image; or if there is superimposition of other images on the medical image.

[0036] For such determination, the AI ​​information verification unit 13b may have, for example, a memory (not shown) that stores information about image features suitable for AI determination. The AI ​​information verification unit 13b determines the suitability of the input information by comparing the information in the memory with the input information regarding features such as image quality format, frame rate, aspect ratio, and superposition of other information, through image analysis of the input information and analysis of metadata associated with the input information, and obtains an input suitability determination result. For example, the AI ​​information verification unit 13b may determine the position of a medical image in an image based on a video signal, which is the input information, and obtain an input suitability determination result according to the degree of agreement between that position and the position of a medical image in an image suitable for AI determination.

[0037] Figure 2 illustrates an example of determining the validity of input information through image analysis of endoscopic images, which are the input information. In contrast, Figure 3 shows an example where the input information includes information that directly indicates the validity of the input information.

[0038] In Figure 3, image P3 has identification information P3a superimposed within the display area P1R, indicating that the endoscopic image P1Ra is suitable for AI analysis. In the example in Figure 3, the identification information P3a is image information. The identification information P3a is placed in the masked areas at the four corners of the display area P1R, and depending on the identification information P3a, almost no information about the valid image area of ​​the endoscopic image P1Ra is lost.

[0039] For example, barcodes or QR codes (registered trademarks) may be used as identification information P3a. Note that while Figure 3 shows an example of identification information P3a that is recognizable by sight, identification information P3a may also be of a type that is not visible to the naked eye.

[0040] Furthermore, the identification information P3a may be included not only as an image within the image based on the video signal, but also in the header information of the video signal, and metadata indicating that the endoscopic image P1Ra is an image suitable for AI judgment may be included in the input information.

[0041] The AI ​​information verification unit 13b can determine the validity of the input information by verifying the identification information P3a. In this way, the AI ​​information verification unit 13b determines, based on the video signal, whether or not the video signal is suitable for recognition processing such as AI judgment.

[0042] The AI ​​information verification unit 13b performs input qualification determination at appropriate timings. For example, the AI ​​information verification unit 13b may perform input qualification determination when the inspection support device 10 is started up, or when an endoscope (not shown) is connected to the video processor device 20. The AI ​​information verification unit 13b may also perform input qualification determination when the settings of the video processor device 20 that supplies input information to the inspection support device 10 are changed, or it may be configured to perform input qualification determination at predetermined intervals.

[0043] If the AI ​​information verification unit 13b determines, based on the results of the input qualification judgment, that the input information is not suitable for AI judgment, it may generate display information to display a warning message (alert) on the display screen of the display unit 30, such as "This is not the target endoscope" or "The settings are incorrect," and supply this information to the display unit 30.

[0044] (Device validity judgment) The connection destination verification unit 13a may have a memory that stores ID information about a legitimate device in order to determine the legitimacy of the device. The connection destination verification unit 13a may determine the legitimacy of the device by comparing the information stored in the memory with the source device information. For example, when endoscopic images are acquired in multiple examination rooms, it is conceivable that, due to wiring errors or the like, endoscopic images acquired in an examination room different from the target examination room may be input as input information from the video processor device 20 to the AI ​​information verification unit 13b. Even in this case, since the AI ​​judgment on the input image is obtained correctly, if the source device information is not checked, operation support will be provided based on endoscopic images from an examination room different from the target examination room. In this embodiment, the connection destination verification unit 13a determines the legitimacy of the device by comparing the information stored in the memory with the source device information. If the ID information stored in the memory and the ID information from the video processor device 20 are different, the AI ​​information verification unit 13b determines that the source device is invalid, thereby preventing the user from receiving incorrect operation support.

[0045] Furthermore, the source device information received by the connection destination confirmation unit 13a includes not only the ID information of the video processor device 20, as described above, but also information about the operation and control of the video processor device 20, including information about the connected endoscope, information about the examination, and information about the video format. The connection destination confirmation unit 13a may determine the legitimacy of the device based on this operation information, control information, and video format information. In addition, the connection destination confirmation unit 13a may determine the legitimacy of the device using information from the AI ​​information confirmation unit 13b and the results of the AI ​​determination.

[0046] For example, the connection destination verification unit 13a may determine that the device is not legitimate if the AI ​​information verification unit 13b determines that the same image has been continuously input when the operation of the video processor device 20 should change the endoscopic image of the input information, and determine that the device is legitimate if the AI ​​information verification unit 13b determines that the image has changed. In addition, the connection destination verification unit 13a may determine that the device is legitimate if the video format information matches the video format information determined by the AI ​​information verification unit 13b, and that the device is not legitimate if they do not match. In this way, the connection destination verification unit 13a can determine the legitimacy of the device based on the source device information from the video processor device 20 and the image recognition result from the AI ​​information verification unit 13b.

[0047] Furthermore, the connection confirmation unit 13a may, for example, determine that the device is legitimate if it receives operational information indicating that an endoscopic examination has been started, and if the AI ​​determination unit 14 determines that the input image has changed to an image of the inside of the body, i.e., that the examination has started. Alternatively, the connection confirmation unit 13a may determine the legitimacy of the device by checking whether the operational information related to the observation mode, such as NBI or RDI, matches the observation mode determined by the AI ​​determination result.

[0048] In other words, the video processor device 20, which is a medical image generation device, can set an observation mode for the endoscope. The AI ​​judgment result determines whether or not the endoscopic image obtained in the observation mode set by the video processor device 20 has been supplied to the examination support device 10 as input information. That is, the legitimacy of the device is determined by determining whether or not the medical image generation device connected to the first communication route and the medical image generation device connected to the second communication route are the same.

[0049] Furthermore, if the connection destination verification unit 13a determines that the device is invalid based on the device legitimacy determination, it may output a signal to the AI ​​determination unit 14 to stop the operation of the AI ​​determination unit 14.

[0050] (Various judgments) The AI ​​judgment unit 14 may be capable of performing various AI judgments. For example, the AI ​​judgment unit 14 may be capable of outputting endoscope operation guide information as examination support information. The AI ​​judgment unit 14 may also be capable of outputting control information for an automated endoscope insertion system. Furthermore, the AI ​​judgment unit 14 may be capable of outputting identification information for lesions and organ locations.

[0051] The input information confirmation unit 13 may control the AI ​​judgment unit 14 so as not to output inspection support information if it obtains an equipment legitimacy judgment indicating that the transmitting equipment is invalid, or an input eligibility judgment indicating that the input information is unsuitable for AI judgment. Furthermore, if the endoscope connected to the video processor device 20 is an automatically controlled endoscope capable of automatic insertion, the input information confirmation unit 13 may be configured to activate an emergency stop device (not shown) to stop the automatic insertion of the automatically controlled endoscope if it obtains an equipment legitimacy judgment indicating that the transmitting equipment is invalid, or an input eligibility judgment indicating that the input information is unsuitable for AI judgment.

[0052] Specifically, the input information confirmation unit 13 outputs inspection support information when it determines that the source device at the destination of the first communication route and the source device at the destination of the second communication route are the same. Furthermore, the input information confirmation unit 13 outputs alert information related to the destinations of the first communication route and the second communication route when it determines that the source devices at the destination of the first communication route and the source devices at the destination of the second communication route are different.

[0053] (operation) Next, the operation of the embodiment configured in this way will be described with reference to Figures 4 to 6. Figure 4 is a flowchart showing the AI ​​judgment control by the AI ​​information confirmation unit 13b. Figures 5 and 6 are flowcharts showing the device legitimacy determination by the connection destination confirmation unit 13a.

[0054] An endoscope is connected to the video processor device 20, and the video processor device 20 drives the endoscope to obtain endoscopic images. For example, to provide operational support for endoscopic examinations, the input information output unit 22 of the video processor device 20 outputs the endoscopic images obtained by the endoscope to the examination support device 10 via a second communication route. In addition, the device information output unit 21 of the video processor device 20 outputs source device information, including operation information of the video processor device 20, operation information, and information such as the video format of the endoscopic images, to the examination support device 10 via a first communication route.

[0055] In S1 of Figure 4, the AI ​​information verification unit 13b receives input information from the input information output unit 22 of the video processor device 20 in order to determine the input qualification of the input information. The AI ​​information verification unit 13b determines whether the input information is suitable for AI judgment by performing image analysis on the received input information, which is a medical image, and by analyzing the metadata of the input information (S2). If the AI ​​information verification unit 13b determines that the input information is suitable for AI judgment (YES judgment in S2), it outputs an input qualification determination result to the AI ​​judgment unit 14 to allow the AI ​​judgment unit 14 to perform AI judgment (S3). If the AI ​​information verification unit 13b determines that the input information is not suitable for AI judgment (NO judgment in S2), it outputs an input qualification determination result to the AI ​​judgment unit 14 to prevent the AI ​​judgment unit 14 from performing AI judgment (S4).

[0056] In this case, the AI ​​judgment unit 14 does not perform an AI judgment. The input information confirmation unit 13 also displays a warning message on the display screen of the display unit 30, such as "This is not the target endoscope" or "The settings are incorrect."

[0057] If the AI ​​information verification unit 13b determines that the input information is suitable for AI judgment, it provides the input information to the AI ​​judgment unit 14. The AI ​​judgment unit 14 obtains an AI judgment result through recognition processing of the input information. The AI ​​judgment unit 14 outputs this AI judgment result to the input information verification unit 13.

[0058] In S11 of Figure 5, the connection destination confirmation unit 13a of the input information confirmation unit 13 receives source device information from the device information output unit 21 of the video processor device 20 in order to determine the legitimacy of the source device. The connection destination confirmation unit 13a compares the received source device information with the information stored in memory to determine whether the source device that sent the source device information is the target device (S12). If the connection destination confirmation unit 13a determines that the source device is legitimate (YES determination in S12), it displays the AI ​​determination from the AI ​​determination unit 14 on the display unit 30 (S13). If the connection destination confirmation unit 13a determines that the source device is invalid (NO determination in S12), it does not display the AI ​​determination result from the AI ​​determination unit 14 on the display unit 30 (S14).

[0059] The example in Figure 5 shows that the connection destination verification unit 13a determines the legitimacy of the source device, for example, by the ID information of the source device, and decides whether to display or hide the AI ​​judgment result. However, the connection destination verification unit 13a may also use the outputs of the AI ​​information verification unit 13b and the AI ​​judgment unit 14 to determine the legitimacy of the source device and decide whether to display or hide the AI ​​judgment result.

[0060] Figure 6 shows an example of this case. The AI ​​information confirmation unit 13b of the input information confirmation unit 13 determines whether the input information is suitable for AI judgment in S31 of Figure 6. If the input information is not suitable for AI judgment, the AI ​​judgment is not performed, as described above. If the input information is suitable for AI judgment, the AI ​​judgment is performed, and the AI ​​judgment result is provided to the input information confirmation unit 13.

[0061] In S32, the input information confirmation unit 13 determines the legitimacy of the source device based on ID information, etc. (source determination). If the input information confirmation unit 13 determines that the source device is legitimate (YES determination in S33), it determines to start the inspection in the following S34. That is, when the input information confirmation unit 13 detects that the inspection has started based on the source device information, it obtains an AI determination result from the AI ​​determination unit 14 indicating whether or not the inspection has started. If both the source device information and the AI ​​determination result indicate that the inspection has started (YES determination in S35), the input information confirmation unit 13 determines the observation mode in the following S36.

[0062] When the input information confirmation unit 13 detects the observation mode based on the source device information, it obtains an AI judgment result regarding the observation mode from the AI ​​judgment unit 14. If both the source device information and the AI ​​judgment result indicate the same observation mode (YES judgment in S37), the input information confirmation unit 13 determines in the next S38 that the source is a legitimate device. In this case, the input information confirmation unit 13 displays the AI ​​judgment result from the AI ​​judgment unit 14 on the display unit 30.

[0063] On the other hand, if the input information confirmation unit 13 determines NO in S33, S35, and S37, the input information confirmation unit 13 determines in S39 that the source is not a legitimate device. In this case, the input information confirmation unit 13 does not display the AI ​​judgment result from the AI ​​judgment unit 14 on the display unit 30. In this case, the input information confirmation unit 13 displays a warning message such as "This is not the target endoscope" or "The settings are incorrect" on the display screen of the display unit 30.

[0064] In this embodiment, not only is the legitimacy of the transmitting device determined, but the suitability of the input information is also determined. Therefore, it is possible to reliably determine that input information suitable for AI judgment has been input from the target device. This makes it possible to prevent the display of inappropriate AI judgment results in cases such as when an inappropriate image is input for AI processing due to a connection error, or when AI judgment is performed on input information from a device that is not the target.

[0065] (modified version) Figure 7 is a block diagram showing a modified configuration. In Figure 1, an example was described in which the video processor device 20 transmits source device information and input information to the inspection support device 10. In the modified configuration shown in Figure 7, when identification information is superimposed on the input information and transmitted, information similar to this identification information is transmitted to the inspection support device 10 via a third communication cable C3 that constitutes a third communication route.

[0066] The video processor device 20A is a video processor device 20 with an identification information output unit 23 added. When the device information output unit 21 adds identification information to the input information and transmits it, the identification information output unit 23 outputs identification information having the same content as the identification information added to the input information to the inspection support device 10 via the third communication cable C3. The device information acquisition unit 11 of the inspection support device 10 receives the identification information output from the identification information output unit 23 and provides it to the input information confirmation unit 13.

[0067] In the example shown, the identification information output unit 23 outputs identification information to the inspection support device 10 via the third communication cable C3. However, the third communication cable C3 may be omitted, and the identification information may be supplied to the inspection support device 10 via the first communication cable C1.

[0068] The equipment information acquisition unit 11 of the inspection support device 10 provides the received identification information to the AI ​​information confirmation unit 13b. The AI ​​information confirmation unit 13b compares the identification information attached to the input information with the identification information received via the third communication cable C3. If it determines that these pieces of information are of the same content, it may determine that the source of the input information and the source of the transmitting device match, or, in some cases, it may further determine that the input information is suitable for AI determination.

[0069] The other configurations and functions are the same as those of the embodiment shown in Figure 1.

[0070] With this modified configuration, since the same identification information is transmitted via the first communication route and the second (or third) communication route, it is relatively easy to determine the legitimacy of the source of the input information.

[0071] The present invention is not limited to the embodiments described above, and the components can be modified and implemented in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the embodiments described above. For example, some components of all the components shown in the embodiments may be deleted. Moreover, components from different embodiments may be appropriately combined.

Claims

1. It is an inspection support system, Having one or more processors with hardware, The aforementioned processor, Information about the source device is obtained from the source device, which is the source of the medical images obtained by photographing the subject, via a first communication route connected to the source device. Based on the aforementioned source device information, it is determined whether the destination source device is appropriate or not. A video signal is acquired via a second communication route connected to the aforementioned transmitting device, and based on the video signal, it is determined whether or not the video signal is a video signal for recognition processing. By performing recognition processing based on the aforementioned video signal, inspection support information is output to assist in the inspection. A testing support system characterized by the following features.

2. The processor determines whether the video signal is the video signal for recognition processing based on the position of the medical image on the image obtained by imaging the video signal. The inspection support system according to feature 1.

3. The processor determines, based on the aspect ratio of the medical image, whether the video signal is a video signal for the recognition process. The inspection support system according to feature 1.

4. The processor determines, based on the image quality of the medical image, whether the video signal is a video signal for the recognition process. The inspection support system according to feature 1.

5. The processor determines whether the video signal is a video signal for recognition processing based on whether identification information indicating that the video signal is a video signal for recognition processing is superimposed on the medical image when the video signal is imaged. The inspection support system according to feature 1.

6. The aforementioned processor, The recognition process is performed when the aforementioned video signal is the video signal for the recognition process. The inspection support system according to feature 1.

7. The aforementioned identification information is a QR code (registered trademark). The inspection support system according to feature 5.

8. The aforementioned processor, Information regarding the content of the identification information is obtained via a first communication route connected to the source device which is the source of the medical image. The identification information is obtained via a second communication route connected to the aforementioned source device. Based on the information regarding the content of the identification information obtained via the first communication route and the identification information obtained via the second communication route, it is determined whether the source device connected to the first communication route and the source device connected to the second communication route are the same. The inspection support system according to feature 5.

9. The transmitting device is a medical image generation device that sets an observation mode and generates the medical image by observing the subject in the set observation mode. The aforementioned processor, Information related to the setting of the observation mode is acquired via the first communication route. The video signal is acquired via the second communication route described above. Based on whether the video signal corresponds to the observation mode setting, it is determined whether the medical image generating device connected to the first communication route and the medical image generating device connected to the second communication route are the same. The inspection support system according to feature 1.

10. The processor outputs an alert when it determines that the video signal is not a video signal for the recognition process. The inspection support system according to feature 1.

11. The aforementioned processor, When it is determined that the aforementioned video signal is not the video signal for the recognition processing, the inspection support information is not output. When it is determined that the aforementioned video signal is the video signal for the recognition process, the inspection support information is output. The inspection support system according to feature 1.

12. The aforementioned processor, When it is determined that the source device to which the first communication route is connected and the source device to which the second communication route is connected are the same, the inspection support information is output. When it is determined that the source device to which the first communication route is connected and the source device to which the second communication route is connected are different, alert information relating to the destinations of the first and second communication routes is output. The inspection support system according to feature 8.

13. The aforementioned processor, When it is determined that the source device to which the first communication route is connected and the source device to which the second communication route is connected are the same, the inspection support information is output. When it is determined that the source device to which the first communication route is connected and the source device to which the second communication route is connected are different, alert information relating to the destinations of the first and second communication routes is output. The inspection support system according to feature 9.

14. The aforementioned processor, The recognition processing is performed by inputting the video signal to a recognition processing device that has been trained using medical images for training. The system determines that the video signal is not the video signal for recognition processing if at least one of the aspect ratio and image quality of the medical image differs from the aspect ratio and image quality of the training medical image, or if the medical image has other images superimposed on it. The inspection support system according to feature 1.

15. The processor outputs endoscope operation guide information as the inspection support information. The inspection support system according to feature 1.

16. The processor outputs control information for the automated endoscope as the inspection support information. The inspection support system according to feature 1.

17. The device information acquisition unit acquires source device information relating to the source device, which is the source of the medical image obtained by photographing the subject, via a first communication route connected to the source device. The input information verification unit determines whether the destination source device is appropriate based on the source device information. The input information acquisition unit acquires the video signal via a second communication route connected to the source device. The input information confirmation unit determines, based on the video signal, whether or not the video signal is a video signal for recognition processing. By performing recognition processing based on the aforementioned video signal, inspection support information is output to assist in the inspection. A method for controlling a medical device, characterized by the features described above.

Images