Assistance device, assistance method, and recording medium

Abstract

Provided are an assistance device, an assistance method, and a recording medium that make it possible to perform appropriate control of an energy device. The assistance device inputs electrical information into a trained model, estimates intraoperative information, determines whether the accuracy of the intraoperative information is equal to or greater than a certain value, and performs control according to the intraoperative information if it is determined that the accuracy of the intraoperative information is equal to or greater than the certain value.

Description

Support device, support method, and recording medium

[0001] The present disclosure relates to an assistance device, an assistance method, and a recording medium.

[0002] Conventionally, there has been known an energy device that applies treatment energy to a portion of biological tissue that is to be treated (hereinafter referred to as the target portion) (see, for example, Patent Document 1). In Patent Document 1, energy and captured images of biological tissue are input to a trained model that has trained using an energy device or a plurality of training tissue images of biological tissue, and image recognition information that is at least one of tissue information regarding the biological tissue and processing information regarding treatment of the biological tissue is estimated, and an energy output instruction based on this estimated image recognition information is output to a generator that supplies energy to the energy device.

[0003] WO 2023 / 286334

[0004] However, in the above-mentioned Patent Document 1, data other than captured images is not utilized, and no consideration is given to cases where the accuracy of the estimation made by the trained model is low, so there is room for improvement in the control method for the energy device.

[0005] The present disclosure has been made in view of the above, and aims to provide an assistance apparatus, an assistance method, and a recording medium that are capable of performing appropriate control over an energy device.

[0006] In order to solve the above-mentioned problems and achieve the objectives, the assistance device of the present disclosure is an assistance device equipped with a processor, wherein the processor acquires electrical information related to an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue that is to be treated, and inputs the electrical information into a trained model trained using multiple learning data sets that use at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information related to the size, type, and state of the portion as input parameters, and links intraoperative information related to the operation on the portion as an output parameter, causes the electrical information to be estimated, determines whether the accuracy of the intraoperative information is equal to or greater than a certain value, and performs control according to the intraoperative information if it is determined that the accuracy is equal to or greater than the certain value.

[0007] In addition, in the support device according to the present disclosure, in the above disclosure, the processor calculates the accuracy of the intraoperative information, determines whether the accuracy is equal to or greater than a certain value, and if it determines that the accuracy is equal to or greater than the certain value, performs control according to the intraoperative information.

[0008] In addition, in the support device according to the present disclosure, the processor calculates the accuracy based on a plurality of pieces of intraoperative information.

[0009] In addition, in the support device according to the present disclosure, when the processor determines that the degree of certainty is equal to or greater than a certain value, the processor outputs the intraoperative information to the outside.

[0010] In addition, in the support device of the present disclosure, in the above disclosure, if the processor determines that the accuracy is a certain value or greater, it changes the control parameters for driving the energy device based on the intraoperative information, while if it determines that the accuracy is not a certain value or greater, it maintains the control parameters for driving the energy device.

[0011] In addition, in the support device according to the present disclosure, in the above disclosure, the electrical information is time-series data of at least one of an US signal and an HF signal for driving the energy device.

[0012] In addition, in the assistance device according to the present disclosure, in the above disclosure, the intraoperative information is at least one of control parameters for controlling the energy device when the energy device applies energy to the part and the part information.

[0013] In addition, in the support device according to the present disclosure, in the above disclosure, the processor inputs the electrical information to each of a plurality of trained models that have learned each of the plurality of learning data under different conditions, causes each of the plurality of trained models to estimate a plurality of pieces of intraoperative information, calculates the accuracy of the intraoperative information based on the plurality of pieces of intraoperative information, determines whether the accuracy is equal to or greater than a certain value, and if it determines that the accuracy is equal to or greater than the certain value, performs control according to the intraoperative information.

[0014] In addition, in the support device according to the present disclosure, in the above disclosure, the processor uses at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and part information regarding the size, type, and state of the part as input parameters, inputs the electrical information into an authenticity-trained model that has trained multiple authenticity training data linked to the accuracy of the degree of deviation between the electrical information and the multiple training data as an output parameter, estimates the accuracy of the electrical information as an estimation result, determines whether the accuracy of the electrical information is greater than or equal to a certain value, and if the accuracy of the electrical information is greater than or equal to the certain value, inputs the electrical information into the trained model and estimates the intraoperative information.

[0015] In addition, in the support device of the present disclosure, in the above disclosure, the processor uses at least the following as input parameters: driving information when the energy device is driven; device information indicating the state of the energy device; part information regarding the size, type, and state of the part; and correspondence information indicating whether the trained model is compatible; inputs the electrical information into a judgment-trained model that has trained a plurality of judgment-trained data linked to the correspondence information as an output parameter, estimates the correspondence information, determines whether the accuracy of the correspondence information is greater than or equal to a certain value, and if it determines that the accuracy of the correspondence information is greater than or equal to the certain value, inputs the electrical information into the trained model and estimates the intraoperative information.

[0016] In addition, in the support device according to the present disclosure, in the above disclosure, the processor uses at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and part information regarding the size, type, and state of the part as input parameters, inputs the electrical information into a situation discrimination trained model that has learned a plurality of situation learning data linked to situation information regarding the state of the part as output parameters, estimates the situation information, selects an optimal trained model from a plurality of trained models that have learned each of the plurality of learning data obtained under different situations based on the situation information, inputs the electrical information into the optimal trained model, and estimates the intraoperative information.

[0017] In addition, the support device according to the present disclosure, in the above disclosure, determines whether the accuracy of the situation information is equal to or greater than a certain value, and if it determines that the accuracy of the situation information is equal to or greater than a certain value, inputs the electrical information into the optimal trained model and estimates the intraoperative information.

[0018] In addition, in the above disclosure, the assistance device according to the present disclosure, when it is determined that the accuracy of the situation information is not a certain value or more, inputs the electrical information into the optimal trained model to estimate the intraoperative information, while when it is determined that the accuracy of the situation information is not a certain value or more, maintains the control parameters for driving the energy device.

[0019] In addition, in the assistance device according to the present disclosure, in the above disclosure, the processor uses at least the following as input parameters: driving information when the energy device is driven; device information indicating the state of the energy device; part information regarding the size, type, and state of the part; and grip information regarding the gripping state of the part by the energy device; inputs the electrical information into a grip discrimination trained model that has learned a plurality of grip learning data linked to the grip information as an output parameter, estimates the grip information, and further inputs the grip information into the trained model, and estimates the intraoperative information.

[0020] In addition, the assistance device according to the present disclosure, in the above disclosure, determines whether the accuracy of the grasping information is equal to or greater than a certain value, and if it determines that the accuracy of the grasping information is equal to or greater than a certain value, further inputs the grasping information into the trained model to estimate the intraoperative information.

[0021] In addition, in the support device of the present disclosure, in the above disclosure, the processor determines and outputs control parameters for controlling the energy device based on the electrical information, and after outputting the control parameters to the energy device, acquires the electrical information, inputs the electrical information into the trained model, and estimates the intraoperative information.

[0022] In addition, in the assistance device according to the present disclosure, in the above disclosure, the processor uses as input parameters at least any of driving information when the energy device is driven, device information indicating the state of the energy device, part information related to the size, type and state of the part, and secondary discrimination information indicating whether secondary discrimination is possible using the trained model, and inputs the electrical information into a secondary discrimination trained model that has trained a plurality of secondary discrimination learning data linked to the secondary discrimination information as an output parameter, estimates the secondary discrimination, and inputs the electrical information and the secondary discrimination information into the trained model, and estimates the intraoperative information.

[0023] In addition, in the assistance device according to the present disclosure, in the above disclosure, the secondary discrimination information is a pressure resistance value of a blood vessel in the region or a state of treatment by the energy device on the region.

[0024] Furthermore, an assistance method according to the present disclosure is an assistance method executed by an assistance device having a processor, wherein the processor acquires electrical information related to an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue that is to be treated, and inputs the electrical information into a trained model trained using a plurality of learning data to which at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information related to the size, type, and state of the portion is linked as an input parameter with intraoperative information related to the operation on the portion as an output parameter, estimates the intraoperative information, determines whether the accuracy of the intraoperative information is equal to or greater than a certain value, and performs control according to the intraoperative information if it is determined that the accuracy is equal to or greater than the certain value.

[0025] Furthermore, a recording medium according to the present disclosure is a non-transitory computer-readable recording medium having an executable program recorded thereon, the program acquiring electrical information relating to an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue that is to be treated, and using at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information relating to the size, type, and state of the portion as input parameters, inputting the electrical information into a trained model trained using multiple learning data linked to intraoperative information related to the operation on the portion as an output parameter, estimating the intraoperative information, determining whether the accuracy of the intraoperative information is equal to or greater than a certain value, and if it is determined that the accuracy is equal to or greater than the certain value, performing control according to the intraoperative information.

[0026] According to the present disclosure, it is possible to provide an effect of appropriately controlling an energy device.

[0027] FIG. 1 is a diagram illustrating a schematic configuration of a medical system according to a first embodiment of the present disclosure. FIG. 2 is a block diagram illustrating a functional configuration of a support device according to the first embodiment of the present disclosure. FIG. 3 is a flowchart illustrating an outline of processing performed by the support device according to the first embodiment of the present disclosure. FIG. 4 is a block diagram illustrating a functional configuration of a support device according to a second embodiment of the present disclosure. FIG. 5 is a flowchart illustrating an outline of processing performed by the support device according to the second embodiment of the present disclosure. FIG. 6 is a block diagram illustrating a functional configuration of a support device according to a third embodiment of the present disclosure. FIG. 7 is a flowchart illustrating an outline of processing performed by the support device according to the third embodiment of the present disclosure. FIG. 8 is a block diagram illustrating a functional configuration of a support device according to a fourth embodiment of the present disclosure. FIG. 9 is a flowchart illustrating an outline of processing performed by the support device according to the fourth embodiment of the present disclosure. FIG. 10 is a diagram schematically illustrating determination content determined by a determination unit of the support device according to the fourth embodiment of the present disclosure. FIG. 11 is a diagram illustrating an example of a determination result determined by a determination unit of the support device according to the fourth embodiment of the present disclosure. FIG. 12 is a block diagram illustrating a functional configuration of a support device according to a fifth embodiment of the present disclosure. FIG. 13 is a flowchart illustrating an outline of processing performed by the support device according to the fifth embodiment of the present disclosure. FIG. 14 is a block diagram showing a functional configuration of a support device according to a sixth embodiment of the present disclosure. FIG. 15 is a flowchart showing an outline of processing performed by the support device according to the sixth embodiment of the present disclosure. FIG. 16 is a block diagram showing a functional configuration of a support device according to a seventh embodiment of the present disclosure. FIG. 17 is a flowchart showing an outline of processing performed by the support device according to the seventh embodiment of the present disclosure. FIG. 18 is a block diagram showing a functional configuration of a support device according to an eighth embodiment of the present disclosure. FIG. 19 is a flowchart showing an outline of processing performed by the support device according to the eighth embodiment of the present disclosure. FIG. 20 is a block diagram showing a functional configuration of a support device according to a ninth embodiment of the present disclosure. FIG. 21 is a flowchart showing an outline of processing performed by the support device according to the ninth embodiment of the present disclosure. FIG. 22 is a flowchart showing an outline of processing performed by the support device according to a first modification of the ninth embodiment. FIG. 23 is a flowchart showing an outline of processing performed by the support device according to a second modification of the ninth embodiment.FIG. 24 is a block diagram showing the functional configuration of the treatment tool and the treatment tool control device according to the tenth embodiment.

[0028] A medical system including an assistance device according to the present disclosure will be described in detail below with reference to the drawings. Note that the invention of the present disclosure is not limited to the following embodiments. Furthermore, the drawings referred to in the following description merely show a schematic representation of the shape, size, and positional relationship to the extent that the contents of the present disclosure can be understood. In other words, the present disclosure is not limited to the shape, size, and positional relationship exemplified in each drawing. Furthermore, in the description of the drawings, identical parts are denoted by the same reference numerals. Furthermore, the configuration of the medical system according to the present disclosure will be described below, followed by the configuration of the endoscope system.

[0029] (Embodiment 1) [Configuration of Medical System] Fig. 1 is a diagram showing a schematic configuration of a medical system according to embodiment 1. The medical system 1 shown in Fig. 1 is used in the medical field, and is a system for observing biological tissue in a subject such as a living organism, and for applying energy to a site of the biological tissue that is to be treated, thereby performing treatment.

[0030] As shown in FIG. 1, the medical system 1 includes an endoscopic system 10 that displays an observation image based on image data captured inside a subject, a treatment system 100 that performs treatment by applying treatment energy to a target area of ​​biological tissue, and an assistance device 200 that assists the treatment system 100.

[0031] [Configuration of Endoscope System] First, we will explain the configuration of the endoscope system 10. The endoscope system 10 includes an insertion section 2, a light source device 3, a light guide 4, an endoscopic camera head 5 (medical imaging device), a display device 6, and an endoscope control device 7.

[0032] The insertion section 2 is rigid or at least partially flexible and has an elongated shape. The insertion section 2 is inserted into a subject such as a patient via a trocar. The insertion section 2 is provided with an optical system such as a lens that forms an observation image inside.

[0033] The light source device 3 is connected to one end of the light guide 4 and supplies illumination light to the end of the light guide 4 to irradiate the inside of the subject under the control of the endoscope control device 7. The light source device 3 is realized using at least a light source, a processor having hardware such as an FPGA (Field Programmable Gate Array) or a CPU (Central Processing Unit), and a memory that is a temporary storage area used by the processor. Here, the light source is configured using one or more of an LED (Light Emitting Diode) light source, a xenon lamp, and a semiconductor laser element such as an LD (Laser Diode).

[0034] One end of the light guide 4 is detachably connected to the light source device 3, and the other end is detachably connected to the insertion section 2. The light guide 4 guides the illumination light supplied from the light source device 3 to the insertion section 2.

[0035] The eyepiece 21 of the insertion portion 2 is detachably connected to the endoscopic camera head 5. Under the control of the endoscope control device 7, the endoscopic camera head 5 receives the observation image formed by the insertion portion 2 and performs photoelectric conversion to generate an image pickup signal (RAW data), and outputs this image pickup signal to the endoscope control device 7. In the following, the configuration including the insertion portion 2 and the endoscopic camera head 5 will be simply referred to as the endoscope 2.

[0036] Under the control of the endoscope control device 7, the display device 6 displays an observation image based on an imaging signal that has been subjected to image processing in the endoscope control device 7, and various information related to the medical system 1. The display device 6 also displays various information input from the support device 200 via the endoscope control device 7. The display device 6 is realized using a display monitor such as a liquid crystal or organic EL (Electro Luminescence) display.

[0037] The endoscope control device 7 is realized using a processor having hardware such as a GPU (Graphics Processing Unit), FPGA, or CPU, and a memory that is a temporary storage area used by the processor. The endoscope control device 7 comprehensively controls the operations of the light source device 3, the endoscopic camera head 5, and the display device 6 in accordance with programs stored in the memory. The endoscope control device 7 comprehensively controls the operations of the endoscope system 10 while cooperating with the treatment system 100 and the assistance device 200.

[0038] [Overview of Treatment System] Next, the configuration of the treatment system 100 will be described. The treatment system 100 applies at least one of ultrasonic energy and high-frequency energy to a region of biological tissue to be treated (hereinafter simply referred to as "target region"), thereby treating the target region. Treatments that can be performed by the treatment system 100 according to embodiment 1 include a treatment for coagulating and sealing the target region, a treatment for incising the target region, and a treatment for simultaneously performing coagulation and incision.

[0039] 1, the treatment system 100 includes a treatment tool 110 and a treatment tool control device 120. In FIG. 1, one side of the treatment tool 110 along the central axis Ax is referred to as a distal side Ar1, and the other side is referred to as a proximal side Ar2. In the following, the treatment system 100 will be described using an ultrasonic device capable of applying ultrasonic energy and high-frequency energy as an example of an energy device, but the treatment system 100 may be a monopolar device that applies high-frequency power between an electrode provided at the tip of the device and an electrode outside the body, or a bipolar device that applies high-frequency power between two jaws.

[0040] The treatment tool 110 is an ultrasonic treatment tool that applies ultrasonic energy and high-frequency energy to a target site to treat the target site, and corresponds to a surgical device. The treatment tool 110 includes a handpiece 111 and an ultrasonic transducer 112.

[0041] The handpiece 111 includes a holding case 113 , a movable handle 114 , a switch 115 , a rotating knob 116 , a pipe 117 , a jaw 118 , and a vibration transmission member 119 .

[0042] The ultrasonic transducer 112 includes a TD (transducer) case 112a and an ultrasonic vibrator 112b.

[0043] The TD case 112a supports the ultrasonic vibrator 112b and is detachably connected to the holding case main body 113a.

[0044] The ultrasonic transducer 112b generates ultrasonic vibrations under the control of the treatment tool control device 120. In the first embodiment, the ultrasonic transducer 112b is configured by a BLT (bolt-tightened Langevin type transducer).

[0045] The holding case 113 constitutes the external appearance of the treatment tool 110 and supports the entire treatment tool 110. The holding case 113 includes a substantially cylindrical holding case main body 113a that is coaxial with the central axis Ax, and a fixed handle 113b that extends downward in FIG. 1 from the main body of the holding case 113 and is held by an operator such as a surgeon.

[0046] The movable handle 114 accepts an opening / closing operation by an operator such as a surgeon. The opening / closing operation is an operation of opening and closing the jaw 118 relative to an end 119a on the distal end side Ar1 of the vibration transmission member 119. The switch 115 is provided in a state exposed to the outside from the side surface of the distal end side Ar1 of the fixed handle 113b. The switch 115 accepts a treatment operation by an operator such as a surgeon. The treatment operation is an operation of applying ultrasonic energy or high-frequency energy to a target area. If the switch 115 has multiple buttons, an operation instruction is assigned to each button.

[0047] The rotation knob 116 has a generally cylindrical shape coaxial with the central axis Ax and is provided on the distal end side Ar1 of the holding case main body 113a. The rotation knob 116 is rotated by an operator such as a surgeon. When rotated, the rotation knob 116 rotates around the central axis Ax relative to the holding case main body 113a. Furthermore, when the rotation knob 116 rotates, the pipe 117, the jaw 118, and the vibration transmission member 119 rotate around the central axis Ax.

[0048] The pipe 117 is a cylindrical pipe. A pin (not shown) that rotatably supports the jaw 118 is fixed to the end of the distal end side Ar1 of the pipe 117.

[0049] At least a portion of the jaw 118 is made of a conductive material. The jaw 118 opens and closes with respect to an end 119a on the distal end side Ar1 of the vibration transmission member 119 in response to an operator such as a surgeon gripping the movable handle 114, and grips the target area between the jaw 118 and the end 119a.

[0050] The vibration transmission member 119 is made of a conductive material and has an elongated shape that extends linearly along the central axis Ax. The vibration transmission member 119 is inserted into the pipe 117 with an end 119a on the distal end side Ar1 protruding outward. At this time, the end on the proximal end side Ar2 of the vibration transmission member 119 is mechanically connected to the ultrasonic transducer 112, although not specifically shown in the drawings. That is, the vibration transmission member 119 transmits ultrasonic vibrations generated by the ultrasonic transducer 112 from the end on the proximal end side Ar2 to the end 119a on the distal end side Ar1. In the first embodiment, the ultrasonic vibrations are longitudinal vibrations that vibrate in a direction along the central axis Ax.

[0051] The treatment tool control device 120, under the control of the assistance device 200, comprehensively controls the operation of the treatment tool 110 via the electric cable 130. Specifically, the treatment tool control device 120 detects a treatment operation on the switch 115 by an operator such as a surgeon. When the treatment tool control device 120 detects a treatment operation by the operator, it applies ultrasonic energy or high-frequency energy via the electric cable 130 to a target site grasped between the jaw 118 and the end 119a on the distal end side Ar1 of the vibration transmission member 119. In other words, the treatment tool control device 120 treats the target site via the treatment tool 110.

[0052] For example, when ultrasonic energy is applied to a target site, the treatment tool control device 120, under the control of the support device 200, supplies driving power to the ultrasonic transducer 112b via the electric cable 130. This causes the ultrasonic transducer 112b to generate longitudinal vibrations (ultrasonic vibrations) that vibrate in a direction along the central axis Ax. Furthermore, the end 119a on the distal end side Ar1 of the vibration transmission member 119 vibrates at a desired amplitude due to the longitudinal vibrations. Then, ultrasonic vibrations are applied from the end 119a to the target site grasped between the jaw 118 and the end 119a. In other words, ultrasonic energy is applied to the target site from the end 119a.

[0053] Furthermore, for example, when applying high-frequency energy to a target site, the treatment tool control device 120, under the control of the assistance device 200, supplies high-frequency power between the jaw 118 and the vibration transmission member 119 via the electric cable 130. As a result, high-frequency current flows through the target site grasped between the jaw 118 and the end 119a on the distal end side Ar1 of the vibration transmission member 119. In other words, high-frequency energy is applied to the target site.

[0054] The treatment tool control device 120 is communicatively connected to the endoscope control device 7 and the support device 200, and when the switch 115 is pressed, outputs a signal indicating the switch being pressed to the endoscope control device 7. Furthermore, the treatment tool control device 120 outputs electrical information related to an energy signal for driving the treatment tool 110 to the support device 200 and receives control parameters for driving the treatment tool 110 input from the support device 200. Here, the electrical information refers to time-series data of at least one of an US signal and a high-frequency (HF) signal for driving the treatment tool 110. Of course, the electrical information may include voltage, current, power, impedance, phase, etc. Furthermore, the control parameters include voltage, current, supplied power, phase, output time, etc.

[0055] [Configuration of the Support Device] Next, a description will be given of the functional configuration of the support device 200. The support device 200 outputs various information for supporting the endoscope system 10 and the treatment system 100 to the endoscope system 10 and the treatment system 100. Specifically, the support device 200 acquires various information from the endoscope system 10 and the treatment system 100, estimates control parameters for supporting the endoscope system 10 and the treatment system 100 based on the acquired various information, and outputs the estimated control parameters to the endoscope system 10 and the treatment system 100.

[0056] [Functional Configuration of the Support Device] Next, a description will be given of the functional configuration of the support device 200. FIG.

[0057] The support device 200 shown in FIG. 2 includes an I / O (Input / Output) device 210, an I / O device 220, an input unit 230, a display unit 240, an output unit 250, a communication unit 260, a recording unit 270, and a control unit 280.

[0058] The I / O device 210 receives image data and various information output from the endoscope control device 7, and outputs the received image data and various information to the control unit 280. The I / O device 210 also outputs various information input from the control unit 280 to the endoscope control device 7. The I / O device 210 is configured using, for example, an interface circuit or the like including a connector to which a cable capable of image transmission and communication is connected.

[0059] The I / O device 220 receives electrical information and various types of information output from the treatment tool control device 120, and outputs the received electrical information and various types of information to the control unit 280. The I / O device 220 also outputs various types of information, including control parameters input from the control unit 280, for the treatment tool 110 to treat the target site, to the treatment tool control device 120. The I / O device 220 is configured using, for example, an interface circuit or the like including a connector to which a communication cable is connected.

[0060] The input unit 230 receives input operations from an operator such as a surgeon, and outputs various signals corresponding to the received input operations to the control unit 280. The input unit 230 is configured using input devices such as a touch panel, switches, a keyboard, a mouse, and buttons.

[0061] The display unit 240 displays various information related to the support device 200 under the control of the control unit 280. The display unit 240 is configured using, for example, an organic EL or liquid crystal display monitor.

[0062] The output unit 250 outputs various information related to the support device 200 under the control of the control unit 280. The output unit 250 is configured using, for example, a speaker.

[0063] Under the control of the control unit 280, the communication unit 260 transmits various pieces of information input from the control unit 280 to an external server or the like, and outputs various pieces of information received from an external server to the control unit 280. The communication unit 260 is configured using, for example, a communication module capable of Wi-Fi (registered trademark) or Bluetooth (registered trademark).

[0064] The recording unit 270 records various information related to the assistance device 200. The recording unit 270 is configured using a hard disk drive (HDD), a solid state drive (SSD), a flash memory, a volatile memory, a non-volatile memory, etc. The recording unit 270 has a program recording unit 271 and a first trained model recording unit 272.

[0065] The program recording unit 271 records various programs executed by the assistance device 200 and data being processed.

[0066] The first trained model recording unit 272 records the first trained model. Specifically, the first trained model is a trained model trained using a plurality of pieces of training data linked to a known trained model and at least any of drive information when an energy device including the treatment tool 110 is driven, device information indicating the state of the energy device, and site information regarding the type, size, and state of the target site as input parameters, and intraoperative information regarding the operation on the target site as an output parameter.

[0067] The drive information includes the voltage, current, supplied power, impedance, drive time, and phase when the energy device is driven. The device information includes the drive status, stop information, treatment status, and the gripping status of the target area by the energy device. The gripping status of the target area includes normal gripping, oblique gripping, and grip position deviation. For example, the gripping status of the target area includes a value based on the impedance of the target area gripped between the jaw 118 and the end 119a or the contact area of ​​the target area at the end 119a. The type of target area includes the type of blood vessel or biological tissue, the type of mucosa, the type of tumor, and the stage of tumor progression. The size of the target area includes the blood vessel size, blood vessel thickness, blood vessel diameter, and blood vessel pressure resistance. The condition of the target area includes the type of treatment being performed by the energy device (e.g., sealing status, bleeding information, hemostasis information) and the condition of the surrounding environment of the biological tissue. The surrounding environment of the biological tissue includes the amount of connective tissue and water content around the target area.

[0068] The intraoperative information includes control parameters for controlling the energy device (treatment tool 110) when it applies energy to the target site, treatment information by the energy device, status information indicating the state of the target site as a result of the energy device with respect to the target site, and accuracy of the estimation results (output parameters). The control parameters include parameters of the voltage, current, high-frequency energy, and ultrasonic energy that the treatment tool control device 120 supplies to the treatment tool 110 to optimize treatment of the target site using the treatment tool 110. The treatment information includes the details of the treatment performed by the energy device (treatment tool 110) on the target site, such as grasping, incision, and sealing. The status information includes whether the energy device (treatment tool 110) incised the target site, the bleeding status, bleeding risk, blood vessel diameter, blood vessel pressure resistance value, and site information related to the target site. The accuracy is the accuracy of the incision state of the target site output by the first trained model 272, such as a percentage accuracy, evaluation value, and AI score.

[0069] The first trained model is a trained model generated by machine learning. Specifically, the first trained model may be a trained model generated using various machine learning techniques, such as deep learning using a neural network, a support vector machine, a decision tree, naive Bayes, or a k-nearest neighbor algorithm. When a neural network is used for the first trained model, a convolutional neural network (CNN) or the like may be used. That is, the first trained model includes an input layer to which input parameters are input, including drive information when an energy device including the treatment tool 110 is driven, device information indicating the state of the energy device, and site information regarding the type, size, and state of the target site; multiple intermediate layers that perform arithmetic processing on the data input through the input layer; and an output layer that outputs, as output parameters, intraoperative information regarding the operation on the biological tissue recognized based on the calculation results output from the multiple intermediate layers.

[0070] The control unit 280 is realized using a processor having hardware such as a GPU, a DSP (Digital Signal Processor), an FPGA, or a CPU, and a memory that is a temporary storage area used by the processor. The control unit 280 controls each unit that constitutes the assistance device 200. The control unit 280 has an acquisition unit 281, an estimation unit 282, a determination unit 283, a parameter control unit 284, and an output control unit 285.

[0071] The acquiring unit 281 acquires electrical information from the treatment instrument control device 120 of the treatment system 100 via the I / O device 220. Specifically, the acquiring unit 281 acquires electrical information provided from the treatment instrument control device 120 in chronological order when the treatment instrument control device 120 applies at least one of a high-frequency energy signal and an ultrasonic energy signal to the treatment instrument 110 to the target site.

[0072] The estimation unit 282 estimates intraoperative information during surgery based on the first trained model 272 and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 to the first trained model 272 as input parameters, and estimates output parameters output from the first trained model 272 as intraoperative information.

[0073] The determination unit 283 determines whether the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value.

[0074] The parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 to the control parameters included in the intraoperative information estimated by the estimation unit 282 and outputs the changed control parameters.

[0075] The output control unit 285 outputs a notification that the control parameters have been changed to the display unit 240 or the output unit 250 .

[0076] [Processing of Support Device] Next, a description will be given of the processing executed by the support device 200. Fig. 3 is a flowchart showing an outline of the processing executed by the support device 200.

[0077] 3 , first, the acquisition unit 281 acquires electrical information from the treatment instrument control device 120 of the treatment system 100 via the I / O device 220 (step S101). Specifically, the acquisition unit 281 acquires electrical information provided in time series when the treatment instrument control device 120 applies at least one of a high-frequency energy signal and an ultrasonic energy signal to the treatment instrument 110 to the target site. More specifically, the electrical information is time-series data of at least one of a US signal and an HF signal supplied by the treatment instrument control device 120 to the treatment instrument 110. Of course, the electrical information may include the voltage, current, driving power, driving time, impedance information, phase information, etc. supplied by the treatment instrument control device 120 to the treatment instrument 110.

[0078] Next, the estimation unit 282 estimates intraoperative information during surgery on the target area based on the first trained model 272 and the electrical information acquired by the acquisition unit 281 (step S102). Specifically, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 to the first trained model 272 as input parameters, and estimates output parameters output from the first trained model 272 as intraoperative information during surgery on the target area.

[0079] Thereafter, the determination unit 283 determines whether the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value (step S103). Here, the certain value is, for example, equal to or greater than 80%. In this case, the determination unit 283 determines whether the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than 80%. Note that the certain value can be set appropriately by the user, but is preferably set to equal to or greater than 80%, more preferably equal to or greater than 90%, to maintain estimation accuracy. If the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value (step S103: Yes), the support device 20 proceeds to step S104, which will be described later. On the other hand, if the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is not equal to or greater than a certain value (step S103: No), the support device 20 proceeds to step S105, which will be described later.

[0080] In step S104, the parameter control unit 284 changes the control parameters used by the treatment instrument control device 120 to control the treatment instrument 110 based on the intraoperative information estimated by the estimating unit 282, and outputs the changed parameters to the treatment instrument control device 120. Specifically, the parameter control unit 284 changes the control parameters used by the treatment instrument control device 120 to the control parameters included in the intraoperative information estimated by the estimating unit 282, and outputs the changed parameters. For example, the parameter control unit 284 outputs to the treatment instrument control device 120 a control parameter that increases or decreases the output current supplied by the treatment instrument control device 120 to the treatment instrument 110. Of course, the parameter control unit 284 may stop the output current supplied by the treatment instrument control device 120 to the treatment instrument 110. This allows the treatment instrument control device 120 to supply to the treatment instrument 110 high-frequency energy signals and ultrasonic energy signals that enable the treatment instrument 110 to perform optimal treatment on the target site being treated. As a result, the operator can perform optimal treatment on the target site using the treatment instrument 110.

[0081] Next, the output control unit 285 outputs a notice of the change in the control parameters to the display unit 240 or the output unit 250 (step S105). In this case, the output control unit 285 may display the bleeding risk (e.g., the probability of bleeding risk), blood vessel diameter, blood vessel pressure resistance value, and the location of the target site on the display unit 240, or may output an alarm sound from the output unit 250, based on the status information included in the intraoperative information estimated by the estimation unit 282. This allows the operator to understand that the control parameters have been changed. After step S105, the assistance device 200 proceeds to step S107, which will be described later. Note that the assistance device 200 may skip step S105 and proceed to step S106, depending on the display settings set by the operator in the assistance device 200 when the control parameters are changed.

[0082] In step S106, the parameter control unit 284 maintains the control parameters used by the treatment tool control device 120 to control the treatment tool 110. Specifically, the parameter control unit 284 maintains the control parameters used by the treatment tool control device 120 to control the treatment tool 110 without adopting the control parameters included in the intraoperative information estimated by the estimating unit 282. This makes it possible to prevent the control parameters from being automatically changed in the state of a low-reliability estimation result when the accuracy of the estimation result estimated by the estimating unit 282 is less than a certain value. After step S106, the assistance device 200 proceeds to step S107, which will be described later.

[0083] In step S107, the determination unit 283 determines whether the operator has finished the treatment on the subject. If the determination unit 283 determines that the operator has finished the treatment on the subject (step S107: Yes), the assistance device 200 ends this process. On the other hand, if the determination unit 283 determines that the operator has not finished the treatment on the subject (step S107: No), the assistance device 200 returns to the above-mentioned step S101.

[0084] According to the first embodiment described above, when the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110.

[0085] Furthermore, according to embodiment 1, when the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value, the parameter control unit 284 maintains the control parameters used by the treatment tool control device 120 to control the treatment tool 110, and therefore, when the accuracy of the estimation result estimated by the estimation unit 282 is less than a certain value, it is possible to prevent the control parameters from being automatically changed when the estimation result has low reliability.

[0086] Furthermore, according to embodiment 1, the output control unit 285 outputs to the display unit 240 or the output unit 250 information indicating that the control parameters have been changed, allowing the operator to intuitively understand that the control parameters used when the treatment tool 110 applies energy to the target area have been changed.

[0087] (Embodiment 2) Next, embodiment 2 will be described. In embodiment 2, a determination is made based on the accuracy of each of a plurality of estimation results. Therefore, in the following, the functional configuration of the support device according to embodiment 2 will be described, followed by a description of the processing executed by the support device according to embodiment 2. Note that the same components as those in the support device 200 according to embodiment 1 described above will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0088] [Functional Configuration of Support Device] Fig. 4 is a block diagram showing the functional configuration of a support device according to embodiment 2. The support device 200A shown in Fig. 4 includes a control unit 280A instead of the control unit 280 of the support device 200 according to embodiment 1 described above. The control unit 280A further includes a calculation unit 286 in addition to the functional configuration of the control unit 280 according to embodiment 1 described above.

[0089] The calculation unit 286 calculates the accuracy of the intraoperative information based on the intraoperative information of each of a plurality of estimation results performed a predetermined number of times by the estimation unit 282. For example, the calculation unit 286 calculates the average value of the accuracy included in the intraoperative information of each of the estimation results performed ten times by the estimation unit 282 as the accuracy to be compared with a fixed value.

[0090] [Support Device Processing] Next, the processing executed by the support device 200A will be described. FIG. 5 is a flowchart showing an outline of the processing executed by the support device 200A. In FIG. 5, steps S201 to S202 and steps S206 to S209 correspond to steps S101 to S102 and steps S104 to S107, respectively, in FIG. 3 described above, and detailed description will be omitted because similar processing is executed. Therefore, steps S203 to S205 executed by the support device 200A will be described below.

[0091] In step S203, the determination unit 283 determines whether the estimation unit 282 has estimated the intraoperative information a predetermined number of times. Specifically, the determination unit 283 determines whether the estimation unit 282 has estimated the intraoperative information, for example, 10 or more times. If the determination unit 283 determines that the estimation unit 282 has estimated the intraoperative information a predetermined number of times (step S203: Yes), the support device 200A proceeds to step S204, which will be described later. On the other hand, if the determination unit 283 determines that the estimation unit 282 has not estimated the intraoperative information a predetermined number of times (step S203: No), the support device 200A returns to step S201.

[0092] In step S204, the calculation unit 286 calculates the accuracy of the intraoperative information based on the intraoperative information for each of the multiple estimation results obtained by the estimation unit 282 a predetermined number of times. Specifically, the calculation unit 286 calculates the accuracy as the average of the accuracy values ​​included in the intraoperative information for each of the ten estimation results obtained by the estimation unit 282. For example, when the estimation unit 282 outputs the ten estimation results as a percentage indicating the completion of incision of the target region included in the intraoperative information, the calculation unit 286 calculates the average value by adding up the probabilities and dividing by the number of times as the accuracy. Of course, the calculation unit 286 may calculate the accuracy not only as the average, but also as the median or mode, or may calculate the standard deviation, maximum value, minimum value, number of outliers, etc. as the accuracy. Note that, in cases other than when the estimation unit 282 outputs the ten estimation results as a percentage indicating the completion of incision of the target region included in the intraoperative information, the calculation unit 286 may calculate the median or mode as the accuracy for the pressure resistance value or size of a blood vessel.

[0093] The determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S205). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S205: Yes), the support device 200A proceeds to step S206. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S205: No), the support device 200A proceeds to step S207.

[0094] According to the second embodiment described above, when the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110 using control parameters with high reliability as estimated results of the first trained model.

[0095] (Embodiment 3) Next, embodiment 3 will be described. In embodiment 3, accuracy is calculated and determined based on multiple pieces of intraoperative information estimated using multiple trained models that have trained learning data under different conditions. Therefore, hereinafter, the functional configuration of the support device according to embodiment 3 will be described, followed by a description of the processing executed by the support device according to embodiment 3. Note that the same components as those in the support devices 200 and 200A according to embodiments 1 and 2 described above will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0096] [Functional Configuration of Support Device] Fig. 6 is a block diagram showing the functional configuration of a support device according to embodiment 3. The support device 200B shown in Fig. 6 includes a recording unit 270B and a control unit 280B instead of the recording unit 270 and the control unit 280A of the support device 200A according to embodiment 2 described above.

[0097] In addition to the functional configuration of the recording unit 270 according to the first embodiment described above, the recording unit 270B further includes a second trained model 273 and a third trained model recording unit 274.

[0098] The second trained model recording unit 273 records the second trained model. Specifically, the second trained model is a trained model trained using a known trained model and a plurality of training data under conditions different from those of the first trained model, in which actuation information, device information, and body part information are used as input parameters and intraoperative information is linked as an output parameter. The second trained model is a trained model generated using the same machine learning method as the first trained model, and is similar to the first trained model except for the training data used and the weight coefficients in the intermediate layer, etc.

[0099] The third trained model recording unit 274 records the third trained model. Specifically, the third trained model is a trained model trained using a known trained model and multiple pieces of training data under conditions different from those of the first trained model and the second trained model, in which actuation information, device information, and body part information are used as input parameters and intraoperative information is linked as an output parameter. The second trained model is a trained model generated using the same machine learning method as the first trained model, and is a similar trained model except for the training data used and the weight coefficients in the intermediate layer, etc.

[0100] The control unit 280B includes a calculation unit 286B instead of the calculation unit 286 of the control unit 280A according to the second embodiment described above.

[0101] The calculation unit 286B calculates the accuracy of the intraoperative information based on multiple pieces of intraoperative information estimated by the estimation unit 282 using the first trained model, the second trained model, and the third trained model, respectively.

[0102] [Support Device Processing] Next, the processing executed by the support device 200B will be described. Fig. 7 is a flowchart showing an outline of the processing executed by the support device 200B. In Fig. 7, step S301 and steps S305 to S308 correspond to step S201 and steps S206 to S209 in Fig. 5 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, hereinafter, steps S302, S303, and S304 executed by the support device 200B will be described, respectively.

[0103] In step S302, the estimation unit 282 estimates multiple pieces of intraoperative information based on the first trained model recorded by the first trained model recording unit 272, the second trained model recorded by the second trained model recording unit 273, the third trained model recorded by the third trained model recording unit 274, and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 as input parameters to the first trained model and estimates output parameters output from the first trained model as intraoperative information (hereinafter simply referred to as "first intraoperative information") (first estimation result). Similarly, the estimation unit 282 estimates output parameters from the second trained model as intraoperative information (hereinafter simply referred to as "second intraoperative information") (second estimation result). Furthermore, the estimation unit 282 estimates the output parameters output from the third trained model as intraoperative information (hereinafter simply referred to as "third intraoperative information") (third estimation result).

[0104] Next, the calculation unit 286B calculates the accuracy of the intraoperative information based on the multiple pieces of intraoperative information estimated by the estimation unit 282 (step S303). Specifically, the calculation unit 286B calculates the accuracy of the intraoperative information based on the accuracy included in each of the first intraoperative information, the second intraoperative information, and the third intraoperative information (step S303). Specifically, the calculation unit 286B calculates the average value of the accuracy of the incision state of the target site included in each of the first intraoperative information, the second intraoperative information, and the third intraoperative information as the accuracy. Of course, the calculation unit 286 may calculate not only the average value of the accuracy, but also the median or mode as the accuracy.

[0105] Next, the determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S304). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S304: Yes), the support device 200B proceeds to step S306. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S304: No), the support device 200B proceeds to step S307.

[0106] According to the third embodiment described above, when the determination unit 283 determines that the accuracy calculated by the calculation unit 286 based on the accuracy included in the intraoperative information of each of the first trained model, the second trained model, and the third trained model is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110 using control parameters with high reliability as estimated results of the first trained model.

[0107] (Fourth Embodiment) Next, a fourth embodiment will be described. In the fourth embodiment, accuracy is determined based on two pieces of intraoperative information estimated using different trained models. Therefore, the functional configuration of the support device according to the fourth embodiment will be described below, followed by a description of the processing executed by the support device according to the fourth embodiment. Note that the same components as those in the support devices 200, 200A, and 200B according to the first to third embodiments will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0108] [Functional Configuration of Support Device] Fig. 8 is a block diagram showing the functional configuration of a support device according to embodiment 4. The support device 200C shown in Fig. 8 includes a recording unit 270C and a control unit 280C instead of the recording unit 270 and the control unit 280 of the support device 200 according to embodiment 1 described above.

[0109] The recording unit 270C further includes the second trained model recording unit 273 according to the second embodiment described above, in addition to the functional configuration of the recording unit 270 according to the first embodiment described above.

[0110] The control unit 280C includes a determination unit 283C instead of the determination unit 283 of the control unit 280 according to the first embodiment described above.

[0111] The judgment unit 283C judges whether the accuracy of each of the first intraoperative information and the second intraoperative information estimated by the estimation unit 282 using the first learned model recorded by the first learned model recording unit 272 and the second learned model recorded by the second learned model recording unit 273 is equal to or greater than a certain value.

[0112] [Support Device Processing] Next, the processing executed by the support device 200C will be described. Fig. 9 is a flowchart showing an outline of the processing executed by the support device 200C. In Fig. 9, step S401 and steps S404 to S406 correspond to step S101 and steps S104 to S107 in Fig. 3 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, below, step S402 and step S404 executed by the support device 200C will be described, respectively.

[0113] In step S402, the estimation unit 282 estimates intraoperative information during surgery based on the first trained model recorded by the first trained model recording unit 272, the second trained model recorded by the second trained model recording unit 273, and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 into the first trained model as input parameters, and estimates output parameters output from the first trained model as first intraoperative information. Similarly, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 into the second trained model as input parameters, and estimates output parameters output from the second trained model as second intraoperative information.

[0114] Next, the determination unit 283C determines whether the accuracy of each of the first intraoperative information and the second intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value (step S403). If the determination unit 283C determines that the accuracy of each of the first intraoperative information and the second intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value (step S403: Yes), the support device 200C proceeds to step S404. On the other hand, if the determination unit 283C determines that the accuracy of each of the first intraoperative information and the second intraoperative information estimated by the estimation unit 282 is not equal to or greater than a certain value (step S403: No), the support device 200C proceeds to step S405.

[0115] Fig. 10 is a diagram schematically showing the determination content determined by the determination unit 283C. Fig. 11 is a diagram showing an example of the determination result determined by the determination unit 283C. In Fig. 10, the vertical axis indicates the accuracy (%) of the first estimation result, which is the first intraoperative information, and the horizontal axis indicates the accuracy (%) of the second estimation result, which is the second intraoperative information.

[0116] As shown in Fig. 10 , the determination unit 283C determines whether the accuracy of each of the first intraoperative information and the second intraoperative information is included in region D1 when the estimation unit 282C determines whether the accuracy of each of the first intraoperative information and the second intraoperative information is included in region D1. For example, as shown in the first row of the determination result table T1 in Fig. 11 , when the accuracy of the first intraoperative information is "80%" and the accuracy of the second intraoperative information is "10%" (region D1 in Fig. 10 ), the determination unit 283C determines that the accuracy is high and that the accuracy of each of the first intraoperative information and the second intraoperative information is equal to or greater than a certain value. In other words, the determination unit 283C determines that the vascular pressure resistance value is high based on the accuracy of each of the first intraoperative information and the second intraoperative information.

[0117] Similarly, as shown in the second row of the determination result table T1 in Fig. 11, when the accuracy of the first intraoperative information is "20%" and the accuracy of the second intraoperative information is "90%" (area D1 in Fig. 10), the determination unit 283C determines that the accuracy is high and that the accuracy of each of the first intraoperative information and the second intraoperative information is equal to or greater than a certain value. That is, the determination unit 283C determines that the blood vessel pressure resistance value is low based on the accuracy of each of the first intraoperative information and the second intraoperative information.

[0118] In contrast, as shown in the third row of the determination result table T1 in Fig. 11, if the accuracy of the first intraoperative information is "70%" and the accuracy of the second intraoperative information is "80%" (outside of region D1 in Fig. 10), the determination unit 283C determines that the accuracy is low and that the accuracy of each of the first intraoperative information and the second intraoperative information is not equal to or greater than a certain value. Also, as shown in the fourth row of the determination result table T1 in Fig. 11, if the accuracy of the first intraoperative information is "30%" and the accuracy of the second intraoperative information is "30%" (outside of region D1 in Fig. 10), the determination unit 283C determines that the accuracy is low and that the accuracy of each of the first intraoperative information and the second intraoperative information is not equal to or greater than a certain value.

[0119] According to the fourth embodiment described above, when the determination unit 283C determines that the accuracy calculated based on the accuracy included in the intraoperative information of each of the first trained model and the second trained model is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110 using control parameters with high reliability as estimated results of the first trained model.

[0120] (Fifth Embodiment) Next, a fifth embodiment will be described. In the fifth embodiment, it is determined whether the accuracy (deviation) of the electrical information acquired by the acquisition unit 281 with respect to the learning data learned by the first trained model is equal to or greater than a certain value, and if the determination result is equal to or greater than the certain value, the electrical information is input as an input parameter of the first trained model. Therefore, hereinafter, the functional configuration of the assistance device according to the fifth embodiment will be described first, followed by a description of the processing executed by the assistance device according to the fifth embodiment. Note that the same components as those of the assistance devices 200, 200A to 200C according to the first to fourth embodiments described above will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0121] [Functional Configuration of Support Device] Fig. 12 is a block diagram showing the functional configuration of a support device according to embodiment 5. A support device 200D shown in Fig. 12 includes a recording unit 270D and a control unit 280D instead of the recording unit 270 and the control unit 280 of the support device 200 according to embodiment 1 described above.

[0122] The recording unit 270D has an authenticity-trained model recording unit 275 in addition to the functional configuration of the recording unit 270 according to the first embodiment described above.

[0123] The authenticity-trained model recording unit 275 records the authenticity-trained model. Specifically, the authenticity-trained model is a trained model trained using a known trained model and a plurality of pieces of training data that are training data under conditions different from those of the first trained model, and that use drive information, device information, and part information as input parameters and link the accuracy of the degree of deviation from the training data trained by the first trained model as an output parameter. The authenticity-trained model uses a trained model generated using the same machine learning method as the first trained model, and is a similar training model except that the training data used is different.

[0124] The control unit 280D further includes the authenticity determination unit 287 of the control unit 280 according to the first embodiment described above.

[0125] The authenticity determination unit 287 inputs the electrical information acquired by the acquisition unit 281 into the authenticity learned model recorded by the authenticity learned model recording unit 275, estimates the accuracy of the electrical information as an estimated result, and determines whether the accuracy of this electrical information is above a certain value.

[0126] [Support Device Processing] Next, the processing executed by the support device 200D will be described. Fig. 13 is a flowchart showing an outline of the processing executed by the support device 200D. In Fig. 13, step S501 and steps S504 to S509 correspond to step S101 and steps S102 to S107 in Fig. 3 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, steps S502 and S503 executed by the support device 200D will be described below.

[0127] In step S502, the authenticity determination unit 287 inputs input parameters into the authenticity trained model recorded by the authenticity trained model recording unit 275 based on the electrical information acquired by the acquisition unit 281, and estimates the output parameters output from the authenticity trained model as the accuracy of the electrical information indicating the degree of deviation from the learning data learned by the first trained model.

[0128] Next, the authenticity determination unit 287 determines whether the accuracy of the electrical information estimated in step S502 is equal to or greater than a certain value (step S503). If the authenticity determination unit 287 determines that the accuracy of the electrical information is equal to or greater than a certain value (step S403: Yes), the support device 200D proceeds to step S504. On the other hand, if the authenticity determination unit 287 determines that the accuracy of the electrical information is not equal to or greater than a certain value (step S403: No), the support device 200D proceeds to step S508.

[0129] According to the fifth embodiment described above, when the authenticity determination unit 287 determines that the accuracy of the electrical information is equal to or greater than a certain value, and the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110.

[0130] Sixth Embodiment Next, a sixth embodiment will be described. Conventionally, there are individual differences in target biological tissues, and the tissue state is also affected by the target tissue and the treatment environment (grasping state (oblique grasping, grasping at the tip, attachment of extra tissue, wet environment)), making it difficult to select optimal control parameters. Furthermore, in recent years, machine learning algorithms have been used to estimate tissue discrimination and sealing states, but deviations from pre-generated trained models may occur in actual use situations. For this reason, in the sixth embodiment, it is estimated whether or not a scene (situation) is compatible with the estimation of a first trained model, and if the scene is compatible with the first trained model, electrical information is input to the first trained model to estimate intraoperative information. Below, the functional configuration of the support device according to the sixth embodiment will be described, and then the processing executed by the support device according to the sixth embodiment will be described. Note that the same components as those of the support devices 200 and 200A to 200D according to the first to fifth embodiments will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0131] [Functional Configuration of Support Device] Fig. 14 is a block diagram showing the functional configuration of a support device according to embodiment 6. The support device 200E shown in Fig. 14 includes a recording unit 270E and a control unit 280E instead of the recording unit 270 and the control unit 280A of the support device 200A according to embodiment 2 described above.

[0132] The recording unit 270E further includes a judgment-trained model recording unit 276 in addition to the functional configuration of the recording unit 270 according to the second embodiment described above.

[0133] The judgment learned model recording unit 276 records the judgment learned model. Specifically, the judgment learned model is a trained model trained using a known trained model and a plurality of pieces of training data that are training data under conditions different from those of the first trained model, and in which drive information, device information, and part information are used as input parameters and correspondence information indicating whether the first trained model is compatible is linked as an output parameter. The judgment learned model is a trained model generated using the same machine learning method as the first trained model, and a similar training model is used except for the training data used.

[0134] The control unit 280E further includes a determination estimation unit 288 in addition to the functional configuration of the control unit 280A according to the second embodiment described above.

[0135] The determination estimation unit 288 estimates correspondence information indicating whether the first trained model is compatible, based on the determination trained model recorded by the determination trained model recording unit 276 and the electrical information acquired by the acquisition unit 281. Specifically, the determination estimation unit 288 inputs the electrical information as input parameters to the determination trained model, and estimates the output parameters output from the determination trained model as correspondence information.

[0136] [Support Device Processing] Next, the processing executed by the support device 200E will be described. Fig. 15 is a flowchart showing an outline of the processing executed by the support device 200E. In Fig. 14, steps S606 to S613 correspond to steps S201 to S209 in Fig. 5 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, steps S601 to S605 executed by the support device 200E will be described below.

[0137] The acquisition unit 281 acquires electrical information from the treatment instrument control device 120 of the treatment system 100 via the I / O device 220 (step S601).

[0138] In step S602, the judgment estimation unit 288 estimates correspondence information indicating whether the first trained model is compatible, based on the judgment trained model recorded by the judgment trained model recording unit 276 and the electrical information acquired by the acquisition unit 281 (step S602). Specifically, the judgment estimation unit 288 inputs the electrical information as input parameters to the judgment trained model, and estimates the output parameters output from the judgment trained model as correspondence information.

[0139] The determination unit 283 determines whether the determination and estimation unit 288 has estimated the corresponding information a predetermined number of times (step S603). Specifically, the determination unit 283 determines whether the determination and estimation unit 288 has estimated the corresponding information a predetermined number of times (for example, 10 or more times). If the determination unit 283 determines that the determination and estimation unit 288 has estimated the corresponding information a predetermined number of times (step S603: Yes), the support device 200E proceeds to step S604, which will be described later. On the other hand, if the determination unit 283 determines that the determination and estimation unit 288 has not estimated the corresponding information a predetermined number of times (step S603: No), the support device 200E returns to step S601.

[0140] In step S604, the calculation unit 286 calculates the accuracy of the correspondence information based on the plurality of pieces of correspondence information obtained a predetermined number of times by the determination and estimation unit 288. Specifically, the calculation unit 286 calculates the average value of the accuracy of each of the plurality of pieces of correspondence information obtained a predetermined number of times by the determination and estimation unit 288 as the accuracy of the correspondence information.

[0141] Next, the determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S605). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S605: Yes), the support device 200E proceeds to step S606. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S605: No), the support device 200E proceeds to step S613.

[0142] According to the sixth embodiment described above, when the determination unit 283 determines that the determination and estimation unit 288 has estimated the correspondence information a predetermined number of times and the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110.

[0143] (Seventh Embodiment) Next, a seventh embodiment will be described. In the seventh embodiment, situation information during surgery is estimated, and an optimal trained model is selected from a plurality of trained models based on the estimation result. Therefore, in the following, the functional configuration of the support device according to the seventh embodiment will be described, followed by the processing executed by the support device according to the seventh embodiment. Note that the same components as those in the support devices 200, 200A to 200E according to the first to sixth embodiments will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0144] [Functional Configuration of Support Device] Fig. 16 is a block diagram showing the functional configuration of a support device according to embodiment 7. The support device 200F shown in Fig. 16 includes a recording unit 270F and a control unit 280F instead of the recording unit 270 and the control unit 280A of the support device 200A according to embodiment 2 described above.

[0145] In addition to the first trained model recording unit 272 of the recording unit 270 according to the first embodiment described above, the recording unit 270F has a plurality of Nth trained model recording units 272_N (N is an integer greater than or equal to 2) that record each of a plurality of trained models, and a situation discrimination trained model recording unit 277.

[0146] Each of the multiple Nth trained model recording units 272_N records a trained model trained using each of multiple pieces of training data acquired under different circumstances. Specifically, the second trained model recording unit 272_2 records a trained model that takes electrical information as an input parameter and is capable of outputting intraoperative information including control parameters for carotid artery treatment as an output parameter. The third trained model recording unit 272_3 records a trained model that takes electrical information as an input parameter and is capable of outputting intraoperative information including control parameters for woozing treatment as an output parameter. The fourth trained model recording unit 272_4 records a trained model that takes electrical information as an input parameter and is capable of outputting intraoperative information including control parameters for diagonal grasping as an output parameter. Each of the multiple Nth trained model recording units 272_N uses a trained model generated using the same machine learning technique as the first trained model, and similar trained models are used except for different training data. In the seventh embodiment, a case where first to fourth trained models are provided will be described as an example, but this is not limited to this, and the number of trained models can be changed as appropriate.

[0147] The situation discrimination trained model recording unit 277 records the situation discrimination trained model. Specifically, the situation discrimination trained model is a trained model trained using a known trained model and multiple pieces of training data under a situation different from that of the first trained model, in which actuation information, device information, and body part information are used as input parameters and situation discrimination is linked as an output parameter to select the trained model that is optimal for the intraoperative situation from multiple Nth trained models. The situation discrimination trained model uses a trained model generated using the same machine learning method as the first trained model, and is similar to the first trained model except for the training data used.

[0148] The control unit 280F further includes a situation estimation unit 289 and a selection unit 290 in addition to the functional configuration of the control unit 280F according to the second embodiment described above.

[0149] The situation estimation unit 289 estimates situation information related to the situation of the target body part based on the situation discrimination trained model recorded by the situation discrimination trained model recording unit 277 and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282F inputs the electrical information as input parameters to the situation discrimination trained model, and estimates output parameters output from the situation discrimination trained model as situation information.

[0150] The selection unit 290 selects the trained model that is best suited to the current situation of the target area from the first trained model recording unit 272 to the Nth trained model recording unit 272_N based on the situation information estimated by the situation estimation unit 289.

[0151] [Support Device Processing] Next, the processing executed by the support device 200F will be described. Fig. 17 is a flowchart showing an outline of the processing executed by the support device 200F. In Fig. 17, step S701 and steps S709 to S712 correspond to step S101 and steps S104 to S106 in Fig. 3 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, hereinafter, steps S702 to S708 executed by the support device 200F will be described, respectively.

[0152] In step S702, the estimation unit 282F estimates situation information of the site during surgery based on the situation discrimination trained model recorded by the situation discrimination trained model recording unit XX and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282F inputs the electrical information as an input parameter to the situation discrimination trained model and causes the model to output the situation information as an output parameter, thereby making the estimation.

[0153] Next, the determination unit 283 determines whether the estimation unit 282F has estimated the situation information a predetermined number of times (step S703). Specifically, the determination unit 283 determines whether the estimation unit 282F has estimated the situation information five or more times. If the determination unit 283 determines that the estimation unit 282F has estimated the situation information a predetermined number of times (step S703: Yes), the support device 200F proceeds to step S304, which will be described later. On the other hand, if the determination unit 283 determines that the estimation unit 282F has not estimated the situation information a predetermined number of times (step S203: No), the support device 200F returns to step S701.

[0154] In step S704, the calculation unit 286 calculates the accuracy of the situation information based on the multiple pieces of situation information obtained by the estimation unit 282F a predetermined number of times. Specifically, the calculation unit 286 calculates, as the accuracy, the number of times for each scene included in the situation information for each of the ten estimation results obtained by the estimation unit 282F. Of course, the calculation unit 286 may calculate, as the accuracy, the number of times for each tissue type, the number of times for each blood vessel type, the number of times for each blood vessel size, the number of times for each treatment type, the number of times for ...

[0155] Next, the determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S705). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S705: Yes), the support device 200F proceeds to step S706. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S705: No), the support device 200F proceeds to step S711.

[0156] In step S706, the selection unit 290 selects the trained model that is best suited to the current situation of the target area from among the multiple trained models recorded by each of the multiple Nth trained model recording units 272_N, based on the situation information estimated by the situation estimation unit 289.

[0157] Next, the acquisition unit 281 acquires electrical information from the treatment tool control device 120 (step S707).

[0158] Next, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 into the optimal trained model selected by the selection unit 290, and estimates the output parameters output from the optimal trained model as intraoperative information including control parameters (step S708).

[0159] According to the seventh embodiment described above, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 into the optimal trained model selected by the selection unit 290, and causes the optimal trained model to output the information to estimate intraoperative information including control parameters. The parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110.

[0160] Furthermore, according to embodiment 7, the selection unit 290 selects the optimal trained model, and then the stay unit 282 estimates and changes the control parameters, which enables more precise judgment and adjustment of the control parameters than simply changing the control parameters using a trained model, and is also advantageous in terms of processing speed, etc.

[0161] (Embodiment 8) Next, embodiment 8 will be described. Conventionally, if the treatment tool 110 continues to output energy while still in a gripping state even after tissue incision at the target site is complete, this could lead to an excessive temperature rise at the tip of the treatment tool 110 or damage to the vibration transmission member 119. To prevent this, a technique is known in which the completion of tissue incision is detected based on load fluctuations on the treatment tool 110 and the output is automatically stopped. However, depending on the tissue gripping position of the treatment tool 110 (for example, when the target site is gripped with the tip of the vibration transmission member 119 of the treatment tool 110 and output is performed), the detection rate of the completion of tissue incision at the target site may decrease. For this reason, in embodiment 8, the gripping state of the target site by the treatment tool 110 is estimated, and control parameters are estimated based on this gripping state. Below, the functional configuration of the support device according to embodiment 8 will be described, followed by the processing executed by the support device according to embodiment 8. Note that the same components as those in the support devices 200, 200A to 200F according to embodiments 1 to 7 described above are designated by the same reference numerals, and detailed description thereof will be omitted.

[0162] [Functional Configuration of Support Device] Fig. 18 is a block diagram showing the functional configuration of a support device according to embodiment 8. The support device 200G shown in Fig. 18 includes a recording unit 270G and a control unit 280G instead of the recording unit 270 and the control unit 280A of the support device 200A according to embodiment 2 described above.

[0163] The recording unit 270G further includes a grip determination learned model recording unit 278 in addition to the functional configuration of the recording unit 270 according to the second embodiment described above.

[0164] The grasp determination learned model recording unit 278 records the grasp determination learned model. Specifically, the grasp determination learned model is a learned model trained using a known learned model and a plurality of pieces of learned data under conditions different from those of the first learned model, in which drive information, device information, and body part information are used as input parameters and grasp information indicating the state of grasping by the treatment tool 110 is linked to the target body part as an output parameter. The grasp determination learned model uses a learned model generated using the same machine learning method as the first learned model, and is similar to the first learned model except for the learning data used to learn it.

[0165] The control unit 280G further includes a grip estimation unit 291 in addition to the functional configuration of the control unit 280A according to the second embodiment described above.

[0166] The grip estimation unit 291 inputs the electrical information acquired by the acquisition unit 281 as input parameters into the grip judgment learned model recorded by the grip judgment learned model recording unit 278, and estimates the output parameters output from the grip judgment learned model as grip information.

[0167] [Support Device Processing] Next, the processing executed by the support device 200G will be described. Fig. 19 is a flowchart showing an outline of the processing executed by the support device 200G. Step S801 and steps S806 to S810 correspond to steps S101, S102, and S104 to S107 in Fig. 3 described above, respectively, and execute similar processing, so detailed description will be omitted. Therefore, steps S802 to S805 executed by the support device 200G will be described below.

[0168] In step S802, the grasping estimation unit 291 estimates grasping state information of the target site by the treatment tool 110 based on the grasping situation learned model recorded by the grasping judgment learned model recording unit 278 and the electrical information acquired by the acquisition unit 281. Specifically, the estimation unit 282G inputs the electrical information as an input parameter to the grasping situation learned model and outputs the grasping information as an output parameter, thereby making the estimation.

[0169] Next, the determination unit 283 determines whether the grip estimation unit 291 has estimated the grip information a predetermined number of times (step S803). If the determination unit 283 determines that the grip estimation unit 291 has estimated the grip information a predetermined number of times (step S803: Yes), the support device 200G proceeds to step S804, which will be described later. On the other hand, if the determination unit 283 determines that the grip estimation unit 291 has not estimated the grip information a predetermined number of times (step S803: No), the support device 200G returns to step S801.

[0170] In step S704, the calculation unit 286 calculates the accuracy of the grip information based on the plurality of pieces of grip information obtained a predetermined number of times by the grip estimation unit 291. Specifically, the calculation unit 286 calculates, as the accuracy, the average value of the accuracy of the grip state included in each piece of grip information obtained as the ten estimation results by the grip estimation unit 291. Of course, the calculation unit 286 may calculate, as the accuracy, the region or contact area of ​​the target part grasped included in each piece of grip information obtained as the ten estimation results by the grip estimation unit 291, or may calculate the median or mode as the accuracy.

[0171] Next, the determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S805). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S805: Yes), the support device 200G proceeds to step S806. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S805: No), the support device 200G proceeds to step S809.

[0172] According to the eighth embodiment described above, when the determination unit 283 determines that the accuracy of the grip information of the estimation result of the grip estimation unit 291 calculated by the calculation unit 286 is equal to or greater than a certain value, the parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 based on the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters to the treatment tool control device 120. This allows the operator to appropriately control the treatment tool 110 using control parameters with high reliability as the estimation result of the first trained model.

[0173] Ninth Embodiment Next, a ninth embodiment will be described. Conventionally, energy devices that use high-frequency energy for surgical procedures control output power, voltage, and the like based on detected electrical signals. When used for blood vessel sealing, these energy devices control output power, output voltage, and other parameters by detecting tissue conditions to achieve a satisfactory sealing state. Furthermore, in conventional technologies, tissue discrimination is performed using two sizes (small and large), and the output of the energy device is stopped when the respective set termination resistances and minimum output times are met. As a result, in conventional technologies, due to the limited number of tissue discrimination categories, control for blood vessels of each target region is not optimized, and control parameters are fixed to one of the two sizes. This has led to a demand for a technology that can treat blood vessels with optimal control parameters for each blood vessel size, due to the risk of tissue misidentification and the limited number of tissue discrimination categories. Therefore, in the ninth embodiment, after the treatment system 100 determines the control parameters to be used, electrical information is input into a trained model to estimate intraoperative information, and detailed control parameters are changed based on the intraoperative information, thereby performing treatment using the energy device with control parameters optimal for the size of the target region. In the following, the functional configuration of the support device according to embodiment 9 will be described, followed by a description of the processing executed by the support device according to embodiment 9. Note that the same components as those of the support devices 200, 200A to 200G according to the above-described embodiments 1 to 8 will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0174] [Functional Configuration of Support Device] Fig. 20 is a block diagram showing the functional configuration of a support device according to embodiment 9. The support device 200H shown in Fig. 20 includes a recording unit 270H and a control unit 280H instead of the recording unit 270 and the control unit 280A of the support device 200A according to embodiment 2 described above.

[0175] The recording unit 270H has a secondary discrimination trained model recording unit 279 in addition to the functional configuration of the recording unit 270 according to the second embodiment described above.

[0176] The secondary discrimination trained model recording unit 279 records the secondary discrimination trained model. Specifically, the secondary discrimination trained model is a trained model trained using a known trained model and a plurality of training data that are training data under conditions different from those of the first trained model, and in which drive information, device information, and site information are used as input parameters and secondary discrimination information including the size of the target site, the vascular sealing state, and the vascular pressure resistance value, etc., is linked as output parameters. The secondary discrimination trained model uses a trained model generated using the same machine learning method as the first trained model, and is similar to the first trained model except for the training data used.

[0177] The control unit 280H further includes a secondary discrimination estimation unit 292 in addition to the functional configuration of the control unit 280A according to the second embodiment described above.

[0178] The secondary discrimination estimation unit 292 estimates secondary discrimination information of the target region before estimating intraoperative information for treating the target region, after determining control parameters for treatment by the treatment system 100, based on the electrical information acquired by the acquisition unit 281 and the secondary discrimination trained model recorded by the secondary discrimination trained model recording unit 279. Specifically, the secondary discrimination estimation unit 292 inputs the electrical information acquired by the acquisition unit 281 to the secondary discrimination trained model as input parameters, and estimates output parameters output from the secondary discrimination trained model 272 as secondary discrimination information.

[0179] [Processing of Support Device] Next, a process executed by the support device 200H will be described. Fig. 21 is a flowchart showing an outline of the process executed by the support device 200H.

[0180] As shown in FIG. 21, first, the acquisition unit 281 acquires electrical information from the treatment instrument control device 120 of the treatment system 100 (step S901).

[0181] Next, the parameter control unit 284 determines control parameters for the treatment tool 110 to treat the target region based on the electrical information acquired by the acquisition unit 281 (step S902). In this case, the parameter control unit 284 determines control parameters corresponding to the size of the target region based on at least one of the voltage, current, power corresponding to the voltage and current, impedance, and phase included in the electrical information acquired by the acquisition unit 281, and outputs the control parameters to the treatment tool control device 120. Specifically, the parameter control unit 284 determines whether the size of the target region is larger than a predetermined value based on at least one of the voltage, current, power corresponding to the voltage and current, impedance, and phase. If it is determined that the size of the target region is larger than the predetermined value, the parameter control unit 284 determines the control parameters for when the target region is larger than the predetermined value (L size). On the other hand, if it is determined that the size of the target region is not larger than the predetermined value (S size), the parameter control unit 284 determines the control parameters for when the target region is not larger than the predetermined value (S size). Here, the control parameters are the output voltage, output current, output time, and termination resistance value output by the treatment tool 110 to the target region.

[0182] Next, the acquiring unit 281 acquires electrical information from the treatment tool control device 120 (step S903). Thereafter, the secondary discrimination estimation unit 292 estimates secondary discrimination information based on the secondary discrimination trained model recorded by the secondary discrimination trained model recording unit 279 and the electrical information acquired by the acquiring unit 281 (step S904). Specifically, the secondary discrimination estimation unit 292 inputs the electrical information acquired by the acquiring unit 281 to the secondary discrimination trained model as input parameters, and estimates output parameters output from the secondary discrimination trained model 272 as secondary discrimination information.

[0183] Thereafter, the determination unit 283 determines whether the secondary discrimination estimation unit 292 has estimated the secondary discrimination information a predetermined number of times (step S905). If the determination unit 283 determines that the secondary discrimination estimation unit 292 has estimated the secondary discrimination information a predetermined number of times (step S905: Yes), the support device 200H proceeds to step S906, which will be described later. On the other hand, if the determination unit 283 determines that the secondary discrimination estimation unit 292 has not estimated the secondary discrimination information a predetermined number of times (step S905: No), the support device 200H returns to step S903.

[0184] In step S906, the calculation unit 286 calculates the accuracy of the secondary discrimination information based on the multiple pieces of secondary discrimination information obtained by the secondary discrimination estimation unit 292 performing estimation a predetermined number of times. Specifically, the calculation unit 286 calculates, as the accuracy, the number of times the size of the target region (e.g., 7 mm, 6 mm, 5 mm, 4 mm) is included in the secondary discrimination information for each of the ten estimation results performed by the secondary discrimination estimation unit 292. Of course, the calculation unit 286 may calculate, as the accuracy, the number of times the tissue type, the number of times the blood vessel type, the number of times the blood vessel size, the number of times the treatment type, the number of times a sealed state occurs due to treatment, the number of times the presence or absence of bleeding occurs, the number of times a state is grasped by the treatment tool 110 (energy device), and the number of times the surrounding environment of the target region (e.g., the amount of connective tissue and the amount of water) occurs, or may calculate the median or mode as the accuracy.

[0185] Next, the determination unit 283 determines whether the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S907). If the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is equal to or greater than a certain value (step S907: Yes), the support device 200H proceeds to step S908. On the other hand, if the determination unit 283 determines that the accuracy calculated by the calculation unit 286 is not equal to or greater than a certain value (step S907: No), the support device 200H proceeds to step S912.

[0186] In step S908 , the acquisition unit 281 acquires electrical information from the treatment instrument control device 120 of the treatment system 100 .

[0187] Next, the estimation unit 282 estimates intraoperative information during surgery based on the first trained model 272, the electrical information acquired by the acquisition unit 281, and the secondary discrimination information estimated by the secondary discrimination estimation unit 292 (step S909). Specifically, the estimation unit 282 inputs the electrical information acquired by the acquisition unit 281 and the secondary discrimination information estimated by the secondary discrimination estimation unit 292 to the first trained model 272 as input parameters, and causes the first trained model 272 to output the intraoperative information as an output parameter, thereby performing estimation.

[0188] The parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 to the control parameters included in the intraoperative information estimated by the estimation unit 282, and outputs the changed control parameters (step S910). For example, the parameter control unit 284 outputs to the treatment tool control device 120 a control parameter that increases or decreases the output voltage supplied by the treatment tool control device 120 to the treatment tool 110. Of course, the parameter control unit 284 may stop the output voltage supplied by the treatment tool control device 120 to the treatment tool 110. This allows the treatment tool control device 120 to supply to the treatment tool 110 high-frequency energy signals and ultrasonic energy signals that enable the treatment tool 110 to perform optimal treatment on the target site being treated. As a result, the operator can perform optimal treatment on the target site using the treatment tool 110.

[0189] Subsequently, the output control unit 285 outputs a notice to the effect that the control parameters have been changed to the display unit 240 or the output unit 250 (step S911). After step S911, the support device 200H proceeds to step S913.

[0190] In step S912, the parameter control unit 284 maintains the control parameters determined in step S902. This prevents the control parameters from being automatically changed when the accuracy of the estimation result obtained by the estimation unit 282 is less than a certain value and the estimation result has low reliability. After step S912, the assistance device 200H proceeds to step S913, which will be described later.

[0191] Next, the determination unit 283 determines whether the operator has finished the treatment on the subject (step S913). If the determination unit 283 determines that the operator has finished the treatment on the subject (step S913: Yes), the support device 200H ends this process. On the other hand, if the determination unit 283 determines that the operator has not finished the treatment on the subject (step S913: No), the support device 200H returns to the above-mentioned step S903.

[0192] According to the ninth embodiment described above, the treatment tool control device 120 can supply to the treatment tool 110 a high-frequency energy signal and an ultrasonic energy signal for performing optimal treatment on the target site being treated by the treatment tool 110. As a result, the operator can perform optimal treatment on the target site using the treatment tool 110.

[0193] Furthermore, according to embodiment 9, the estimation unit 282 estimates intraoperative information during surgery based on the first trained model 272, the electrical information acquired by the acquisition unit 281, and the secondary discrimination information estimated by the secondary discrimination estimation unit 292, so that it is possible to perform detailed optimal control according to each of multiple sizes, such as blood vessel size, not just two sizes, using the estimation results obtained by machine learning.

[0194] (Variation 1 of Embodiment 9) Next, Variation 1 of Embodiment 9 will be described. Variation 1 of Embodiment 9 has the same configuration as the support device 200H according to the above-described embodiment 9, but executes different processes. Therefore, the process executed by the support device according to Variation 1 of Embodiment 9 will be described below. Note that the same components as those of the support device 200H according to the above-described embodiment 9 are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0195] [Processing of Support Device] Figure 22 is a flowchart showing an outline of the processing executed by the support device 200H in Modification 1 of Embodiment 9. In Figure 22, steps S1001 to S1010 correspond to steps S901 to S910 in Figure 21, respectively. Therefore, step S1011 and subsequent steps will be described below.

[0196] In step S1011 , the acquisition unit 281 acquires electrical information from the treatment tool control device 120 of the treatment system 100 .

[0197] Next, the estimation unit 282 estimates intraoperative information during surgery based on the first learned model 272, the electrical information acquired by the acquisition unit 281, and the secondary discrimination information estimated by the secondary discrimination estimation unit 292 (step S1012).

[0198] Thereafter, the determination unit 283 determines whether the vascular withstand pressure value included in the intraoperative information estimated by the estimation unit 282 is equal to or less than a predetermined value (step S1013). For example, the determination unit 283 determines whether the vascular withstand pressure value included in the intraoperative information estimated by the estimation unit 282 is equal to or less than 1500 mmHg. If the determination unit 283 determines that the vascular withstand pressure value included in the intraoperative information estimated by the estimation unit 282 is equal to or less than the predetermined value (step S1013: Yes), the support device 200H proceeds to step S1014, which will be described later. On the other hand, if the determination unit 283 determines that the vascular withstand pressure value included in the intraoperative information estimated by the estimation unit 282 is not equal to or less than the predetermined value (step S1013: No), the support device 200H proceeds to step S1017, which will be described later.

[0199] In step S1014 , the acquisition unit 281 acquires electrical information from the treatment tool control device 120 of the treatment system 100 .

[0200] Next, the estimation unit 282 estimates intraoperative information during surgery based on the first trained model 272, the electrical information acquired by the acquisition unit 281, the secondary discrimination information estimated by the secondary discrimination estimation unit 292, and the judgment result of the above-mentioned step S1013 (step S1015).

[0201] The parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 to the control parameters included in the intraoperative information estimated by the estimation unit 282 in step S1015 described above, and outputs the changed control parameters (step S1016). For example, if the determination unit 283 determines in step S1013 that the blood vessel pressure resistance value is 500 mmHg or less and the estimation unit 282 estimates in step S1015 that the blood vessel size included in the intraoperative information is 4 mm, the parameter control unit 284 changes the control parameters to those corresponding to a blood vessel pressure resistance value of 500 mmHg or less and a blood vessel size of 4 mm, and outputs the changed control parameters to the treatment tool control device 120. After step S1016, the assistance device 200H proceeds to step S1020 described below.

[0202] In step S1017 , the acquisition unit 281 acquires electrical information from the treatment tool control device 120 of the treatment system 100 .

[0203] Next, the estimation unit 282 estimates intraoperative information during surgery based on the first trained model 272, the electrical information acquired by the acquisition unit 281, the secondary discrimination information estimated by the secondary discrimination estimation unit 292, and the judgment result of the above-mentioned step S1013 (step S1018).

[0204] The parameter control unit 284 changes the control parameters used by the treatment tool control device 120 to control the treatment tool 110 to the control parameters included in the intraoperative information estimated by the estimation unit 282 in step S1015 described above, and outputs the changed control parameters (step S1019). For example, if the determination unit 283 determines in step S1013 that the blood vessel pressure resistance value is not equal to or less than 500 mmHg, and the estimation unit 282 estimates in step S1015 that the blood vessel size included in the intraoperative information is 4 mm, and the blood vessel pressure resistance value is 1000 mmHg, the parameter control unit 284 changes the control parameters to those corresponding to the blood vessel pressure resistance value of 1000 mmHg and the blood vessel size of 4 mm, and outputs the changed control parameters to the treatment tool control device 120. After step S1016, the assistance device 200H proceeds to step S1020 described below.

[0205] In step S1020, the determination unit 283 determines whether the operator has finished the treatment on the subject. If the determination unit 283 determines that the operator has finished the treatment on the subject (step S1020: Yes), the support device 200H ends this process. On the other hand, if the determination unit 283 determines that the operator has not finished the treatment on the subject (step S1020: No), the support device 200H returns to the above-mentioned step S1011.

[0206] In step S1121, the parameter control unit 284 maintains the control parameters used by the treatment tool control device 120 to control the treatment tool 110. Specifically, the parameter control unit 284 maintains the control parameters determined in the above-mentioned step S1002 without adopting the control parameters included in the intraoperative information estimated by the estimation unit 282. This makes it possible to prevent the control parameters from being automatically changed in the state of a low-reliability estimation result when the accuracy of the estimation result estimated by the estimation unit 282 is less than a certain value.

[0207] Next, the determination unit 283 determines whether the operator has finished the treatment on the subject. If the determination unit 283 determines that the operator has finished the treatment on the subject (step S1022: Yes), the support device 200H ends this process. On the other hand, if the determination unit 283 determines that the operator has not finished the treatment on the subject (step S1022: No), the support device 200H returns to the above-mentioned step S1001.

[0208] According to the above-described first modification of the ninth embodiment, similarly to the ninth embodiment, the treatment tool control device 120 can supply to the treatment tool 110 a high-frequency energy signal and an ultrasonic energy signal for performing an optimal treatment on the target site being treated by the treatment tool 110. As a result, the operator can perform an optimal treatment on the target site using the treatment tool 110.

[0209] (Variation 2 of Embodiment 9) Next, Variation 2 of Embodiment 9 will be described. Variation 2 of Embodiment 9 has the same configuration as the support device 200H according to the above-described embodiment 9, but executes different processes. Therefore, the process executed by the support device according to Variation 2 of Embodiment 9 will be described below. Note that the same components as those of the support device 200H according to the above-described embodiment 9 will be assigned the same reference numerals, and detailed description thereof will be omitted.

[0210] [Support Device Processing] Figure 23 is a flowchart showing an outline of the processing executed by the support device 200H in Modification 2 of Embodiment 9. In Figure 23, steps S1101 to S1112 and steps S1114 to S1122 correspond to steps S1001 to S1012 and steps S1014 to S1022 in Figure 22 described above, respectively, and detailed description will be omitted because similar processing is executed. Therefore, the following will describe step S1113 executed by the support device 200H.

[0211] In step S1113, the determination unit 283 determines whether the state of the target region included in the intraoperative information estimated by the estimation unit 282 is a specific state. For example, the determination unit 283 determines whether the state of the target region included in the intraoperative information estimated by the estimation unit 282 is a bleeding state. In this case, the determination unit 283 determines that the state of the target region included in the intraoperative information estimated by the estimation unit 282 is a specific state. If the determination unit 283 determines that the state of the target region included in the intraoperative information estimated by the estimation unit 282 is a specific state (step S1113: Yes), the support device 200H proceeds to step S1114. On the other hand, if the determination unit 283 determines that the state of the target region included in the intraoperative information estimated by the estimation unit 282 is not a specific state (step S1113: No), the support device 200H proceeds to step S1117.

[0212] According to the above-described second modification of the ninth embodiment, similarly to the ninth embodiment, the treatment tool control device 120 can supply to the treatment tool 110 a high-frequency energy signal and an ultrasonic energy signal for performing an optimal treatment on the target site being treated by the treatment tool 110. As a result, the operator can perform an optimal treatment on the target site using the treatment tool 110.

[0213] (Embodiment 10) Next, embodiment 10 will be described. In embodiment 10, the functions of the support device are provided in the treatment tool control device. Therefore, the configuration of the treatment tool control device according to embodiment 10 will be described below. Note that the same components as those in the support devices 200, 200A to 200H according to embodiments 1 to 9 described above are assigned the same reference numerals, and detailed description thereof will be omitted.

[0214] [Configuration of the treatment tool control device] Fig. 24 is a block diagram showing the functional configuration of the treatment tool and the treatment tool control device 120. The treatment tool control device 120 shown in Fig. 24 includes a first power supply 121, a first detection circuit 122, a first ADC (Analog-to-Digital Converter) 123, a second power supply 124, a second detection circuit 125, a second ADC 126, a notification unit 127, a recording unit 128, and a processor 129.

[0215] Here, a pair of transducer lead wires C1, C1' constituting an electric cable 130 are joined to the ultrasonic transducer 112b. For ease of explanation, only one of the pair of transducer lead wires C1, C1' is shown in FIG. 24 . The first power supply 121 outputs a drive signal, which is power for generating ultrasonic vibrations, to the ultrasonic transducer 112b via the pair of transducer lead wires C1, C1' under the control of the processor 129. This causes the ultrasonic transducer 112b to generate ultrasonic vibrations. The vibration transmission member 119 thus functions as an electrode 1 by transmitting the ultrasonic vibrations generated by the ultrasonic transducer 112b from the end on the proximal side Ar2 to the end 119a on the distal side Ar1.

[0216] In the following, for the sake of convenience, the drive signal output from the first power source 121 to the ultrasonic vibrator 112b will be referred to as the input drive signal, and the signal obtained by changing the input drive signal based on the frequency response of the ultrasonic vibrator 112b will be referred to as the output drive signal.

[0217] The first detection circuit 122 includes a first voltage detection circuit 122a, which is a voltage sensor for detecting voltage values, and a first current detection circuit 122b, which is a current sensor for detecting current values. The first detection circuit 122 detects an US signal (analog signal) corresponding to the output drive signal over time. This US signal corresponds to an electrical characteristic value of the ultrasonic transducer. Specifically, examples of the US signal include a phase signal of the voltage in the output drive signal (hereinafter referred to as a "US voltage phase signal"), a phase signal of the current in the output drive signal (hereinafter referred to as a "US current phase signal"), a phase difference between the voltage and current in the output drive signal (hereinafter referred to as a "US phase difference"), a current value in the output drive signal (hereinafter referred to as a "US current"), a voltage value in the output drive signal (hereinafter referred to as a "US voltage"), a power value in the output drive signal (hereinafter referred to as a "US power"), and an impedance value calculated from the US current and the US voltage (hereinafter referred to as an "ultrasonic impedance value").

[0218] The first ADC 123 converts the US signal (analog signal) output from the first detection circuit 122 into a digital signal. Then, the first ADC 123 outputs the converted US signal (digital signal) to the processor 129.

[0219] Then, under the control of the processor 129, the second power supply 124 outputs a high-frequency current and a high-frequency voltage between the jaw 118 and the vibration transmission member 119 via a pair of high-frequency lead wires C2, C2'. That is, the jaw 118 functions as the electrode 2. As a result, a high-frequency current flows through the target area grasped between the jaw 118 and the end 119a (treatment portion) of the vibration transmission member 119. That is, high-frequency energy is applied to this target area. Then, Joule heat is generated in this target area by the flow of the high-frequency current, and the target area is treated.

[0220] The second detection circuit 125 has a second voltage detection circuit 125a which is a voltage sensor that detects a voltage value and a second current detection circuit 125b which is a current sensor that detects a current value, and detects over time an HF signal corresponding to the high-frequency current and high-frequency voltage output from the second power supply 124 to the jaw 118 and the end 119a. Specifically, examples of the HF signal include the high-frequency current (hereinafter referred to as "HF current") and high-frequency voltage (hereinafter referred to as "HF voltage") output from the second power supply 124 to the jaw 118 and the end 119a, the high-frequency power (hereinafter referred to as "HF power") calculated from this HF current and this HF voltage, the impedance value (hereinafter referred to as "HF impedance value") calculated from this HF current and this HF voltage, and the phase difference (hereinafter referred to as "HF phase difference") between this HF current and this HF voltage.

[0221] The second ADC 126 converts the HF signal (analog signal) output from the second detection circuit 125 into a digital signal. Then, the second ADC 126 outputs the converted HF signal (digital signal) to the processor 129.

[0222] The notification unit 127 notifies predetermined information under the control of the processor 129. Examples of this notification unit 127 include an LED (Light Emitting Diode) that notifies predetermined information by lighting up, blinking, or by changing the color when lit, a display device that displays predetermined information, and a speaker that outputs predetermined information by sound.

[0223] The recording unit 128 records various information related to the treatment tool control device 120. The recording unit 128 is configured using an HDD, an SSD, a flash memory, a volatile memory, a non-volatile memory, etc. The recording unit 128 has a program recording unit 271 similar to the recording unit 270 according to the first embodiment described above, and a first trained model recording unit 272.

[0224] The processor 129 is realized using a processor having hardware such as a GPU, a DSP (Digital Signal Processor), an FPGA, or a CPU, and a memory that is a temporary storage area used by the processor. The processor 129 controls each unit that constitutes the treatment tool control device 120. The processor 129 also has the same functions as the control unit 280 according to the first embodiment described above. Specifically, the processor 129 has an estimation unit 282, a determination unit 283, a parameter control unit 284, and an output control unit 285. The processor 129 also executes the same processes as those in the first embodiment described above.

[0225] According to the tenth embodiment described above, when the determination unit 283 determines that the accuracy included in the intraoperative information estimated by the estimation unit 282 is equal to or greater than a certain value, the parameter control unit 284 changes and outputs the control parameters used by the processor 129 to control the treatment tool 110, based on the intraoperative information estimated by the estimation unit 282. This allows the operator to appropriately control the treatment tool 110.

[0226] In addition, in the tenth embodiment, the processor 129 may be configured by appropriately combining the functions of the control units 280A to 280H according to the second to ninth embodiments described above, and the processing according to the second to ninth embodiments may be executed.

[0227] (Other Embodiments) Various inventions can be formed by appropriately combining multiple components disclosed in the medical systems according to the above-described first to ninth embodiments of the present disclosure. For example, some components may be deleted from all of the components described in the medical systems according to the above-described first to ninth embodiments of the present disclosure. Furthermore, the components described in the endoscope systems according to the above-described embodiments of the present disclosure may be appropriately combined.

[0228] Furthermore, in the medical systems according to embodiments 1 to 9 of the present disclosure, the support device is connected to the endoscope system and the treatment system via a wired connection, but this is not limited to this. For example, the support device may be configured as a server or the like, and may be capable of two-way communication via a network.

[0229] Furthermore, in the assistance devices according to embodiments 1 to 9 of the present disclosure, the trained model is recorded in the recording unit, but this is not limited to this and may be provided in the recording unit of, for example, an external server.

[0230] Furthermore, in the medical systems according to the first to ninth embodiments of the present disclosure, the above-described "unit" can be read as "means" or "circuit," etc. For example, a control unit can be read as control means or a control circuit.

[0231] In addition, the programs to be executed by the medical systems according to embodiments 1 to 9 of the present disclosure are provided as file data in an installable or executable format recorded on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, a DVD (Digital Versatile Disk), a USB medium, or a flash memory.

[0232] In addition, the programs to be executed by the medical systems according to embodiments 1 to 9 of the present disclosure may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network.

[0233] In the explanation of the flowcharts in this specification, the order of processing between steps is clearly indicated using expressions such as "first," "then," and "continue," but the order of processing required to implement the present invention is not uniquely determined by these expressions. In other words, the order of processing in the flowcharts described in this specification can be changed within a consistent range.

[0234] Although some of the embodiments of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be implemented in other forms that have undergone various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the disclosure of the present invention.

[0235] REFERENCE SIGNS LIST 1 Medical system 2 Insertion section 3 Light source device 4 Light guide 5 Endoscopic camera head 6 Display device 7 Endoscopic control device 10 Endoscopic system 20 Support device 100 Treatment system 110 Treatment tool 120 Treatment tool control device 130 Electric cable 200, 200A to 200h Support device 210, 220 I / O device 230 Input unit 240 Display unit 250 Output unit 260 Communication unit 270, 270B to 270H Recording unit 271 Program recording unit 272 First trained model recording unit 272_N Nth trained model recording unit 273 Second trained model recording unit 274 Third trained model recording unit 275 Authenticity trained model recording unit 276 Judgment trained model recording unit 277 Situation discrimination trained model recording unit 278 Grasp judgment learned model recording unit 279 Secondary discrimination learned model recording unit 280, 280A to 280H Control unit 281 Acquisition unit 282, 282C, 280F, 282G Estimation unit 283, 283C Judgment unit 284 Parameter control unit 285 Output control unit 286, 286B Calculation unit 287 Authenticity judgment unit 288 Judgment estimation unit 289 Situation estimation unit 290 Selection unit 291 Grasp estimation unit 292 Secondary discrimination estimation unit

Claims

1. An assistance device having a processor, wherein the processor: acquires electrical information regarding an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue that is to be treated; inputs the electrical information into a trained model trained from a plurality of learning data in which at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information regarding the size, type, and state of the portion is used as input parameters, and intraoperative information regarding the operation on the portion is linked as an output parameter, estimates the intraoperative information, determines whether the accuracy of the intraoperative information is equal to or greater than a certain value, and performs control according to the intraoperative information if it is determined that the accuracy is equal to or greater than the certain value.

2. An assistance device as described in claim 1, wherein the processor calculates the accuracy of the intraoperative information, determines whether the accuracy is equal to or greater than a certain value, and, if it determines that the accuracy is equal to or greater than the certain value, performs control according to the intraoperative information.

3. An assistance device according to claim 2, wherein the processor calculates the degree of accuracy based on a plurality of pieces of intraoperative information.

4. An assistance device according to claim 1, wherein the processor outputs the intraoperative information to the outside when it determines that the accuracy is equal to or greater than a certain value.

5. An assistance device as described in claim 1, wherein the processor changes a control parameter for driving the energy device based on the intraoperative information when it determines that the accuracy is equal to or greater than a certain value, while maintaining the control parameter for driving the energy device when it determines that the accuracy is not equal to or greater than the certain value.

6. The support device according to claim 1, wherein the electrical information is time-series data of at least one of an US signal and an HF signal for driving the energy device.

7. An assistance device according to claim 1, wherein the intraoperative information is at least one of control parameters for controlling the energy device when the energy device applies energy to the part and information about the part.

8. An assistance device as described in claim 1, wherein the processor inputs the electrical information to each of a plurality of trained models that have learned each of the plurality of learning data under different conditions, causes each of the plurality of trained models to estimate a plurality of pieces of intraoperative information, calculates the accuracy of the intraoperative information based on the plurality of pieces of intraoperative information, determines whether the accuracy is equal to or greater than a certain value, and performs control according to the intraoperative information if it is determined that the accuracy is equal to or greater than the certain value.

9. An assistance device as described in claim 1, wherein the processor: uses at least any one of driving information when the energy device is driven, device information indicating the state of the energy device, and part information related to the size, type and state of the part as input parameters, inputs the electrical information into an authenticity-trained model that has learned multiple authenticity learning data linked to the accuracy of the deviation between the electrical information and the multiple learning data as an output parameter, and estimates the accuracy of the electrical information as an estimation result; determines whether the accuracy of the electrical information is equal to or greater than a certain value; and, if the accuracy of the electrical information is equal to or greater than the certain value, inputs the electrical information into the trained model and estimates the intraoperative information.

10. An assistance device as described in claim 1, wherein the processor: uses at least any of the following as input parameters: driving information when the energy device is driven, device information indicating the state of the energy device, part information related to the size, type and state of the part, and correspondence information indicating whether the learned model is compatible; inputs the electrical information into a judgment-trained model that has learned a plurality of judgment-trained data linked to the correspondence information as an output parameter, and estimates the correspondence information; determines whether the accuracy of the correspondence information is greater than or equal to a certain value; and, if it determines that the accuracy of the correspondence information is greater than or equal to the certain value, inputs the electrical information into the trained model and estimates the intraoperative information.

11. An assistance device as described in claim 1, wherein the processor: inputs the electrical information into a situation discrimination trained model that has learned a plurality of situation learning data linked to situation information regarding the situation of the part as output parameters, with at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and part information regarding the size, type and state of the part as input parameters, and estimates the situation information; selects an optimal trained model from a plurality of trained models that have learned each of the plurality of learning data obtained under different situations based on the situation information; inputs the electrical information into the optimal trained model, and estimates the intraoperative information.

12. An assistance device as described in claim 11, which determines whether the accuracy of the situation information is equal to or greater than a certain value, and if it is determined that the accuracy of the situation information is equal to or greater than the certain value, inputs the electrical information into an optimal trained model to estimate the intraoperative information.

13. An assistance device as described in claim 11, which, when it is determined that the accuracy of the situation information is not a certain value or more, inputs the electrical information into an optimal trained model to estimate the intraoperative information, while, when it is determined that the accuracy of the situation information is not a certain value or more, maintains control parameters for driving the energy device.

14. An assistance device as described in claim 1, wherein the processor uses as input parameters at least driving information when the energy device is driven, device information indicating the state of the energy device, part information relating to the size, type and state of the part, and gripping information relating to the gripping state of the part by the energy device, and inputs the electrical information into a gripping discrimination trained model that has learned a plurality of gripping learning data linked to the gripping information as an output parameter, to estimate the gripping information, and further inputs the gripping information into the trained model to estimate the intraoperative information.

15. An assistance device as described in claim 14, which determines whether the accuracy of the grasping information is equal to or greater than a certain value, and if it is determined that the accuracy of the grasping information is equal to or greater than the certain value, further inputs the grasping information into the trained model to estimate the intraoperative information.

16. An assistance device as described in claim 1, wherein the processor determines and outputs control parameters for controlling the energy device based on the electrical information, acquires the electrical information after outputting the control parameters to the energy device, and inputs the electrical information into the trained model to estimate the intraoperative information.

17. An assistance device as described in claim 16, wherein the processor inputs the electrical information into a secondary discrimination trained model that has learned a plurality of secondary discrimination learning data linked to the secondary discrimination information as an output parameter, and uses at least one of driving information when the energy device is driven, device information indicating the state of the energy device, part information related to the size, type and state of the part, and secondary discrimination information indicating whether secondary discrimination is possible using the trained model as input parameters, and estimates the secondary discrimination; and inputs the electrical information and the secondary discrimination information into the trained model and estimates the intraoperative information.

18. An assisting device according to claim 17, wherein the secondary discrimination information is a blood vessel pressure resistance value in the region or a state of treatment by the energy device on the region.

19. An assistance method executed by an assistance device having a processor, the assistance method including the processor: acquiring electrical information regarding an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue that is to be treated; inputting the electrical information into a trained model trained from a plurality of learning data in which at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information regarding the size, type, and state of the portion is used as input parameters, and intraoperative information regarding the operation on the portion is linked as an output parameter, to estimate the intraoperative information; determining whether the accuracy of the intraoperative information is a certain value or greater; and if it is determined that the accuracy is the certain value or greater, performing control according to the intraoperative information.

20. A non-transitory computer-readable recording medium having an executable program recorded thereon, the program obtaining electrical information regarding an energy signal for driving an energy device that performs treatment by applying energy to a portion of biological tissue to be treated; inputting the electrical information into a trained model trained using a plurality of learning data in which at least one of driving information when the energy device is driven, device information indicating the state of the energy device, and portion information regarding the size, type, and state of the portion is used as input parameters, and intraoperative information regarding the operation on the portion is linked as an output parameter, to estimate the intraoperative information; determining whether the accuracy of the intraoperative information is equal to or greater than a certain value; and performing control according to the intraoperative information if it is determined that the accuracy is equal to or greater than the certain value.

Images