Cataract surgery evaluation system
The cataract surgery evaluation system addresses the challenge of evaluating intrascleral fixation proficiency by using video data and an evaluation model to classify surgical steps, enhancing training accuracy and simplifying parameter settings.
Patent Information
- Application Number
- JP2020166363
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-09-10
- Publication Date
- 2024-12-09
- Estimated Expiration
- 2040-09-10
AI Technical Summary
Existing cataract surgery training methods, including the use of simulated eyeballs and CG simulators, face challenges in evaluating the proficiency of intrascleral fixation, a complex surgical procedure, due to difficulties in assessing surgical steps and complex parameter settings.
A cataract surgery evaluation system that inputs video data from intrascleral fixation training using a simulated eyeball, employing an evaluation model to classify surgical steps and evaluate proficiency, thereby simplifying parameter settings and improving evaluation accuracy.
The system enables accurate evaluation of intrascleral fixation proficiency by classifying surgical steps and reducing unnecessary information, allowing for practical-level practice, compared to traditional methods.
Smart Images

Figure 0007595833000001 
Figure 0007595833000002 
Figure 0007595833000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a cataract surgery evaluation system. [Background technology]
[0002] Cataract is a disease in which the crystalline lens, which functions as a lens in the eye, becomes cloudy. The crystalline lens is located behind the iris and is enclosed in an anterior and posterior capsule. In cataract surgery, for example, the anterior capsule is torn, the cloudy crystalline lens is aspirated and removed through an incision in the anterior capsule by ultrasonic emulsification, and an intraocular lens as an artificial crystalline lens is inserted into the capsule through the incision.
[0003] To improve the above-mentioned cataract surgery, a considerable number of cases must be accumulated. However, due to the frequent occurrence of medical lawsuits and the strict requirement of informed consent from patients, it is currently very difficult to provide inexperienced young doctors with surgical training through practical clinical surgery. For this reason, surgical training using freshly extracted pig eyes is often used for cataract surgery training. However, pig eyes hurt quickly, so they must be extracted and used quickly, and furthermore, using pig eyes in an operating room is unsanitary. Therefore, for example, a mock eyeball for surgical training (see Patent Document 1, for example) and a simulator using computer graphics (CG) for surgical training have been proposed. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] JP 2019-028305 A Summary of the Invention [Problem to be solved by the invention]
[0005] During cataract surgery, the anterior and posterior capsules may fall off for some reason. In such a case, the intraocular lens cannot be supported by the anterior and posterior capsules, so intrascleral fixation is used to fix the intraocular lens to the sclera. Intrascleral fixation is a very difficult surgical procedure among cataract surgeries. However, in surgical training using a simulated eyeball, it is difficult to evaluate the proficiency of intrascleral fixation. In addition, in a simulator using CG, the parameter settings are complicated, and it is difficult to practice intrascleral fixation at a practical level in the first place. Therefore, it is desirable to provide a cataract surgery evaluation system that can evaluate the proficiency of intrascleral fixation. [Means for solving the problem]
[0006] A cataract surgery evaluation system according to one aspect of the present invention includes an evaluation model that, when video data obtained by imaging a practical training of intrascleral fixation surgery using a simulated eyeball is input, classifies the surgical steps included in the video data, evaluates the proficiency in the classified surgical steps, and outputs the classification result and the evaluation result. This cataract surgery evaluation system further includes an acquisition unit that acquires the classification result and the evaluation result from the evaluation model by inputting the video data to the evaluation model.
[0007] In the cataract surgery evaluation system according to one aspect of the present invention, the surgical steps included in the video data obtained by imaging the training of intrascleral fixation using a simulated eyeball in an evaluation model are classified, and the proficiency level in the classified surgical steps is evaluated. This makes it possible to reduce information that is unnecessary for the evaluation of proficiency included in the video data compared to the case where an excised pig's eye is used, and therefore makes it possible to perform the evaluation of proficiency with high accuracy. In addition, since it is easier to set parameters for the evaluation model compared to a simulator using CG, the training of intrascleral fixation can be performed at a practical level. Effect of the Invention
[0008] According to one aspect of the present invention, the cataract surgery evaluation system classifies the surgical steps contained in the video data obtained by imaging the practice of intrascleral fixation using a mock eyeball in an evaluation model, and evaluates the proficiency level for the classified surgical steps, thereby making it possible to evaluate the proficiency level of intrascleral fixation. [Brief description of the drawings]
[0009] [Figure 1] 1 is a diagram illustrating an example of a schematic configuration of a cataract surgery evaluation system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram illustrating an example of a perspective configuration of the simulated eyeball in FIG. [Diagram 3] 3 is a diagram illustrating an example of a perspective view of the simulated eyeball of FIG. 2 in development; FIG. [Figure 4] 3 is a diagram illustrating an example of a cross-sectional configuration of the simulated eyeball in FIG. 2. [Diagram 5] 2 is a diagram illustrating an example of the internal configuration of the cataract surgery evaluation system in FIG. 1. [Figure 6] FIG. 6 is a diagram illustrating an example of the video data of FIG. 5. [Figure 7] FIG. 6 is a diagram illustrating an example of the video data of FIG. 5. [Figure 8] FIG. 6 is a diagram illustrating an example of the video data of FIG. 5. [Figure 9] FIG. 6 is a diagram illustrating an example of a schematic configuration of the evaluation model of FIG. 5. [Figure 10] FIG. 6 is a diagram illustrating an example of a functional block of the learning module of FIG. 5. [Figure 11] FIG. 2 is a diagram illustrating a modified example of the schematic configuration of the cataract surgery evaluation system in FIG. [Figure 12] FIG. 12 is a diagram illustrating an example of the internal configuration of the cataract surgery evaluation system in FIG. [Figure 13] FIG. 2 is a diagram illustrating a modified example of the schematic configuration of the cataract surgery evaluation system in FIG. [Figure 14] FIG. 14 is a diagram illustrating an example of the internal configuration of the cataract surgery evaluation system in FIG. [Figure 15] FIG. 2 is a diagram illustrating a modified example of the schematic configuration of the cataract surgery evaluation system in FIG. [Figure 16] FIG. 16 is a diagram illustrating an example of the internal configuration of the cataract surgery evaluation system in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0010] Hereinafter, the embodiment of the invention will be described in detail with reference to the drawings. The following description is one specific example of the present invention, and the present invention is not limited to the following embodiment. The present invention is also not limited to the arrangement, dimensions, and dimensional ratios of each component shown in each drawing. In the following description, directions such as up / down, left / right, front / rear, etc. are directions seen from the medical staff side unless otherwise specified. Regarding the up / down direction, the direction from the objective lens described below toward the surgical field is defined as downward, and the opposite direction is defined as upward. Generally, patients undergo surgery in a supine position, so the up / down direction and the vertical direction are the same. The description will be given in the following order. 1. Embodiment (FIGS. 1 to 10) Example of classification and evaluation of surgical procedures using a server device 2. Variations Modification A: Omit the server device and store the surgical classification and evaluation Example of using an information processing device (Fig. 11, Fig. 12) Modification B: Example of using a camera-equipped information processing device (FIGS. 13 and 14) Modification C: Omit the server device and perform surgery classification and evaluation. Example of using a camera-equipped information processing device (Fig. 15, Fig. 16)
[0011] <1. Preferred embodiment> [composition] 1 shows an example of a schematic configuration of a cataract surgery evaluation system 100 according to an embodiment of the present invention. The cataract surgery evaluation system 100 is a system for evaluating the proficiency of intrascleral fixation surgery, which is an extremely difficult surgical procedure among cataract surgeries. This cataract surgery evaluation system 100 uses a simulated eyeball 400 and an ophthalmic surgery microscope system 200.
[0012] (Simulated eyeballs 400) Fig. 2 shows an example of a perspective configuration of the simulated eyeball 400. Fig. 3 shows an example of a perspective configuration of the simulated eyeball 400 when developed. Fig. 4 shows an example of a cross-sectional configuration of the simulated eyeball 400.
[0013] The simulated eyeball 400 is a training device for intrascleral fixation surgery, and is modeled after a part of an actual eye. The simulated eyeball 400 includes, for example, a lower member 410 and an upper member 420. The lower member 410 is a member that supports the upper member 420, and has a shape that models a part of a portion that corresponds to the posterior segment of the eye. The upper member 420 has a shape that models a part of the anterior segment of the eye.
[0014] The lower member 410 has a substantially hemispherical shell-shaped main body 411 having a flat annular upper end surface 411A at its upper portion, and a recessed portion 412 fixed to the upper end surface 411A of the shell-shaped main body 411. The upper end of the recessed portion 412 is opened with a circular opening, and is fixed to the upper end surface 411A of the shell-shaped main body 411. An intraocular lens 500, which will be described later, is placed in the recessed portion 412. The lower member 410 is formed, for example, by resin molding.
[0015] The upper member 420 has an annular simulated sclera portion 421 placed on the upper end surface 411A of the lower member 410, and an annular simulated cornea portion 422 and an annular simulated iris portion 423 fixed to the simulated sclera portion 421. The simulated sclera portion 421 is an annular member placed on the upper end surface 411A of the lower member 410, and has a shape and softness that imitates a part of the sclera of an actual eye. The simulated sclera portion 421 is formed of, for example, a milky white resin. The simulated cornea portion 422 is an annular member fixed to the simulated sclera portion 421, and has a shape and softness that imitates a real cornea. The simulated cornea portion 422 is provided with a window (corneal window 422A) for ensuring a field of view when performing training. The simulated iris portion 423 is an annular member fixed to the simulated sclera portion 421, and has a shape and softness that imitates the iris of a real eye.
[0016] The upper member 420 does not include a simulated capsule portion for encasing the intraocular lens 500. The simulated capsule portion is a simulation of the upper and lower capsules of an actual eye. When actually performing intrascleral fixation, the lower capsule falls off for some reason and is in a state in which it does not function to hold the crystalline lens. Therefore, the simulated capsule portion is an unnecessary member for the practice of intrascleral fixation. Therefore, the recessed portion 412 of the lower member 410 communicates with the outside when the upper member 420 is placed on the lower member 410.
[0017] In cataract surgery, an intraocular lens 500 is used instead of a crystalline lens. As shown in FIG. 2, the intraocular lens 500 is a member in which two connecting parts 520 for fixing the intraocular lens 500 to an actual sclera part (or simulated sclera part 421) are provided on the edge of the main body lens part 510. The two connecting parts 520 are fixed, for example, on the edge of the main body lens part 510 at positions facing each other with the central part of the main body lens part 510 in between. Each connecting part 520 is, for example, thread-like, and made of a material of a color other than white (for example, blue) to improve visibility. The main body lens part 510 is a substitute for the crystalline lens of an actual eye. The main body lens part 510 is, for example, lens-shaped, and made of transparent resin.
[0018] (Ophthalmic Surgery Microscope System 200) Next, we will explain the ophthalmic surgery microscope system 200 used in conjunction with the cataract surgery evaluation system 100. The ophthalmic surgery microscope system 200 includes, for example, a support 210, a first arm 220, a second arm 230, a drive device 240, a medical practitioner microscope 250, and a foot switch 260.
[0019] The support 210 supports the entire ophthalmic surgery microscope system 200. One end of the first arm 220 is connected to the upper end of the support 210. One end of the second arm 230 is connected to the other end of the first arm 220. The drive unit 240 is connected to the other end of the second arm 230. The medical practitioner microscope 250 is suspended by the drive unit 240. The camera 120 used in the cataract surgery evaluation system 100 is attached to the medical practitioner microscope 250. The camera 120 corresponds to a specific example of the "imaging unit" of the present invention. The foot switch 260 is used to perform various operations with the foot of a medical practitioner or the like. The drive unit 240 moves the medical practitioner microscope 250 and the camera 120 three-dimensionally in the vertical and horizontal directions in response to an operation by the medical practitioner or the like.
[0020] The medical practitioner's microscope 250 has a lens barrel section 251 that houses various optical systems and a drive system, and a pair of left and right eyepieces 252. The medical practitioner looks into the eyepieces 252 and observes the surgical field with both eyes through the lens barrel section 251. The lens barrel section 251 is configured to include, for example, an objective lens, an illumination optical system, an observation optical system, and a photographing optical system. The observation optical system is an optical system connected to the objective lens, and the photographing optical system is an optical system connected to the camera 120.
[0021] The illumination optical system illuminates the surgical field via an objective lens. The observation optical system is used to observe the surgical field illuminated by the illumination optical system with a medical practitioner's microscope 250 via an objective lens. A pair of observation optical systems are provided on the left and right, with the right observation optical system connected to the right eyepiece 252 and the left observation optical system connected to the left eyepiece 252. For example, a beam splitter is provided in the right observation optical system, and a portion of the light guided from the surgical field to the right observation optical system is guided to the imaging optical system by the beam splitter.
[0022] (Cataract Surgery Rating System 100) Next, the internal configuration of the cataract surgery evaluation system 100 will be described. FIG. 5 shows an example of the internal configuration of the cataract surgery evaluation system 100. The cataract surgery evaluation system 100 includes, for example, an information processing device 110, a camera 120, and a server device 130. The information processing device 110 and the server device 130 are connected via a network 140. The information processing device 110 is configured to be able to communicate with the server device 130 via the network 140. The server device 130 is configured to be able to communicate with the information processing device 110 via the network 140. The information processing device 110 and the camera 120 are connected via the network 140. The information processing device 110 is configured to be able to communicate with the camera 120 via the network 140. The camera 120 is configured to be able to communicate with the information processing device 110 via the network 140.
[0023] The network 140 is, for example, a network that performs communication using a communication protocol (TCP / IP) that is commonly used on the Internet. The network 140 may be, for example, a secure network that performs communication using a communication protocol unique to the network. The network 140 is, for example, the Internet, an intranet, or a local area network. The connection between the network 140 and the information processing device 110, the server device 130, or the camera 120 may be, for example, a wired LAN (Local Area Network) such as Ethernet, a wireless LAN such as Wi-Fi, or a mobile phone line.
[0024] The information processing device 110 includes, for example, a control unit 111, a memory 112, a display unit 113, an input unit 114, and a network IF (Interface) 115. The control unit 111 corresponds to a specific example of an "acquisition unit" of the present invention. The control unit 111 includes a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), and executes, for example, an operating system (not shown) stored in the memory 112. The control unit 111 further executes, for example, a program 112A stored in the memory 112. The program 112A causes the control unit 111 to execute a series of procedures for evaluating the proficiency of the surgery when the user practices intrascleral fixation surgery on the simulated eyeball 400 included in the surgical field. The series of procedures that the program 112A causes the control unit 111 to execute will be described in detail later.
[0025] The memory 112 stores a program (e.g., an operating system) executed by the control unit 111. The memory 112 is configured by, for example, a random access memory (RAM), a read only memory (ROM), an auxiliary storage device (such as a hard disk), etc. The memory 112 further stores a program 112A.
[0026] The display unit 113 is composed of a display device such as a liquid crystal panel or an organic EL (Electro Luminescence) panel. The display unit 113 displays an image based on a video signal from the control unit 111 on the display surface 110A. The input unit 114 receives an instruction from the outside (for example, a user) and outputs the received instruction to the control unit 111. The input unit 114 may be, for example, a mechanical input interface including a button or a dial, or may be a voice input interface including a microphone. The input unit 114 may be, for example, a touch panel provided on the display surface 110A of the information processing device 110. The network IF 115 is a communication interface for communicating with the camera 120 and the server device 130 via the network 140.
[0027] The camera 120 includes, for example, a charge coupled device (CCD) image sensor or a complementary metal-oxide-semiconductor (CMOS) image sensor. The camera 120 captures an image of a training session of intrascleral fixation surgery using the simulated eyeball 400. The camera 120 captures an image under the control of the information processing device 110, for example, and outputs a moving image (moving image data Ia) obtained by the imaging to the information processing device 110 via the network 140.
[0028] The server device 130 includes, for example, a control unit 131, an evaluation model 132, a learning module 133, memories 134 and 135, and a network IF 136. The evaluation model 132 corresponds to a specific example of an "evaluation model" of the present invention. The control unit 131 includes a CPU, a GPU, and the like, and executes, for example, an operating system stored in the memory 134. The control unit 131 further inputs, for example, video data Ia to the evaluation model 132, and transmits process evaluation data Da acquired from the evaluation model 132 in response to the input of the video data Ia to the information processing device 110 via the network IF 136.
[0029] The network IF 136 is a communication interface for communicating with the information processing device 110 via the network 140. The memory 134 stores a program (e.g., an operating system) executed by the control unit 131, etc. The memory 134 is composed of, for example, a RAM, a ROM, an auxiliary storage device (such as a hard disk), etc. The memory 134 further stores process data 134A and parameters 134B.
[0030] The process data 134A is data used as input data for the learning module 133, and is data on the classification of the surgical process included in each video data 135A, each video data 135B, and each video data 135C. The process data 134A describes, for example, that the surgical process included in each video data 135A is the first step of intrascleral fixation, that the surgical process included in each video data 135B is the middle step of intrascleral fixation, and that the surgical process included in each video data 135C is the latter step of intrascleral fixation. The parameters 134B are data used in the evaluation model 132, and are, for example, data related to a neural network described later, and include, for example, the number of layers of the neural network, the number of neurons in each layer, the connection relationship between neurons, the weight of the connection between each neuron, and the threshold value of each neuron. The parameters 134B are derived by the learning module 133.
[0031] Memory 135 is configured by, for example, a RAM, a ROM, an auxiliary storage device (such as a hard disk), etc. Memory 135 stores, for example, a plurality of moving image data 135A, a plurality of moving image data 135B, and a plurality of moving image data 135C.
[0032] Each of the video data 135A, 135B, and 135C is data used as input data for the learning module 133. Each of the video data 135A, 135B, and 135C is video data obtained by imaging the state of the practice of intrascleral fixation using the simulated eyeball 400 with the camera 120. Each of the video data 135A is data obtained by imaging, for example, the state of the first stage of the intrascleral fixation using the camera 120. Each of the video data 135A is data obtained by imaging, for example, the state of putting the intraocular lens 500 into the recessed portion 412 in the practice of the intrascleral fixation using the simulated eyeball 400 as shown in FIG. 6. Each of the video data 135B is data obtained by imaging, for example, the state of the middle stage of the intrascleral fixation using the camera 120. Each piece of video data 135B is data obtained by imaging, with the camera 120, a state in which two connecting parts 520 of an intraocular lens 500 are passed through a simulated scleral part 421 in a practical training of intrascleral fixation using a simulated eyeball 400 as shown in Fig. 7. Each piece of video data 135C is data obtained by imaging, with the camera 120, a state in which a later stage of intrascleral fixation is performed. Each piece of video data 135C is data obtained by imaging, with the camera 120, a state in which two connecting parts 520 passed through a simulated scleral part 421 are fixed to the simulated scleral part 421 in a practical training of intrascleral fixation using a simulated eyeball 400 as shown in Fig. 8.
[0033] The evaluation model 132 is an artificial intelligence model for image analysis. The evaluation model 132 uses the parameters 134B generated by the learning module 133 to analyze the input video data Ia, classify the surgical steps (e.g., front, middle, and back) included in the video data Ia into predetermined periods, evaluate the proficiency in the classified surgical steps, and output the classification results and evaluation results as process evaluation data Da. The video data Ia is data obtained by capturing an image of a state in which intrascleral fixation surgery is practiced using the simulated eyeball 400 with the camera 120. The classification results include data in which the surgical steps included in the video data Ia are classified into predetermined periods. The proficiency includes data evaluating the proficiency of the surgery included in the video data Ia. When the video data Ia includes multiple classifications of surgical steps, the proficiency includes data evaluating the proficiency of the surgery included in the video data Ia for each classification of the surgical steps.
[0034] The evaluation model 132 includes, for example, a neural network as shown in FIG. 9. The evaluation model 132 includes, for example, a deep neural network such as a convolutional neural network (CNN). The neural network in the evaluation model 132 includes, for example, an input layer 132A, an intermediate layer 132B (hidden layer), and an output layer 132C. Each layer (132A, 132B, 132C) includes one or more neurons. For example, the number of neurons in the input layer 132A can be set according to the number of pixels of each image data included in the video data Ia. The number of neurons in the intermediate layer 132B can be set appropriately. The output layer 132C can be set according to the number of surgical steps (number of classifications) included in the intrascleral fixation surgery.
[0035] Neurons in adjacent layers are connected to each other, and a weight (connection weight) is set for each connection. The number of connections between neurons may be set appropriately. A threshold is set for each neuron, and the output value of each neuron is determined depending on whether the sum of the products of the input values and the weights to each neuron exceeds the threshold.
[0036] Image data Ia is input to the input layer 132A. Process evaluation data Da is output from the output layer 132C. The evaluation model 132 has been trained to output the process evaluation data Da when image data Ia is input. Trained parameters (parameters 134B) of the evaluation model 132 obtained by training are stored in the memory 134. Note that the parameters 134B may be derived in a device different from the server device 130, and the derived parameters 134B may be stored in the memory 134. In this case, it is possible to omit the learning module 133, the process data 134A, the plurality of video data 135A, the plurality of video data 135B, and the plurality of video data 135C in the server device 130.
[0037] Next, a description will be given of a learning method for the evaluation model 132. Fig. 10 shows functional blocks of a learning module 133 that learns the evaluation model 132. The learning module 133 includes, for example, a processor and a memory.
[0038] The learning module 133 includes a learning data generation module 133A and a parameter generation module 133B. The learning data generation module 133A generates learning data Db from the process data 134A and the video data 135A, 135B, and 135C. The video data 135A, 135B, and 135C are acquired by capturing images of a training session of intrascleral fixation using a simulated eyeball 400 similar to the simulated eyeball 400 used when capturing the video data Ia (or a simulated eyeball 400 having the same configuration) with the camera 120. The parameter generation module 133B optimizes parameters of the evaluation model 132 by training the evaluation model 132 using the learning data Db. The parameter generation module 133B acquires the optimized parameters as learned parameters (parameters 134B).
[0039] The initial values of the parameters 134B may be provided by a template. Alternatively, the initial values of the parameters 134B may be provided manually by human input. The learning module 133 may re-learn the evaluation model 132. When re-learning the evaluation model 132, the learning module 133 may generate new parameters 134B based on the learned parameters (parameters 134B) of the evaluation model 132 to be re-learned. The learning module 133 may update the parameters 134B by periodically executing learning of the evaluation model 132.
[0040] [Operation] Next, the operation of the cataract surgery evaluation system 100 will be described.
[0041] First, the camera 120 captures an image of the practice of intrascleral fixation surgery using the simulated eyeball 400, and outputs the video data Ia obtained thereby to the information processing device 110 via the network 140. When the input of the video data Ia from the camera 120 is completed, the information processing device 110 outputs the video data Ia obtained from the camera 120 to the server device 130 via the network 140. Note that the information processing device 110 may output the video data Ia from the camera 120 to the server device 130 in real time or at predetermined time intervals.
[0042] When the server device 130 acquires the video data Ia, it inputs the acquired video data Ia to the evaluation model 132. The evaluation model 132 performs video analysis on the input video data Ia. For example, the evaluation model 132 performs image analysis based on a neural network on the input video data Ia. The evaluation model 132 outputs process evaluation data Da as a response to the input video data Ia. The server device 130 outputs the process evaluation data Da obtained from the evaluation model 132 to the information processing device 110 via the network 140. The information processing device 110 displays the process evaluation data Da obtained from the server device 130 on the display surface 110A.
[0043] [effect] Next, the effects of the cataract surgery evaluation system 100 will be described.
[0044] In cataract surgery, for example, the anterior capsule is torn, the cloudy crystalline lens is aspirated and removed through an incision in the anterior capsule by phacoemulsification, and an intraocular lens as an artificial crystalline lens is inserted into the incision. It takes a considerable number of cases to improve the above-mentioned cataract surgery. However, due to the frequent occurrence of medical lawsuits and the strict requirement of informed consent from patients, it is very difficult to educate inexperienced young doctors on surgery through clinical surgery. For this reason, surgical training using freshly extracted pig eyes is often used for cataract surgery training. However, pig eyes hurt quickly, so they must be extracted and used early, and it is also unsanitary to use pig eyes in the operating room.
[0045] During cataract surgery, the anterior and posterior capsules may fall off for some reason. In such cases, the intraocular lens cannot be supported by the anterior and posterior capsules, so intrascleral fixation is used to fix the intraocular lens to the sclera using thin threads. In surgical training using a mock eyeball, it was difficult to evaluate the proficiency of intrascleral fixation. In addition, with a CG simulator, parameter settings were complicated, making it difficult to train in intrascleral fixation at a practical level in the first place.
[0046] On the other hand, in the present embodiment, the evaluation model 132 classifies the surgical steps included in the video data Ia obtained by capturing an image of the practice of intrascleral fixation using a simulated eyeball, and evaluates the proficiency in the classified surgical steps. This makes it possible to reduce information that is included in the video data Ia and is unnecessary for the evaluation of the proficiency, compared to the case where an excised pig's eye is used, and therefore the evaluation of the proficiency can be performed with high accuracy. In addition, since the parameter setting of the evaluation model 132 is easier than that of a simulator using CG, the practice of intrascleral fixation can be performed at a practical level. Therefore, the cataract surgery evaluation system 100 can evaluate the proficiency of intrascleral fixation.
[0047] In this embodiment, an artificial intelligence model that generates and outputs process evaluation data Da by analyzing video data Ia using parameters 134B generated using learning video data (135A, 135B, 135C) obtained by imaging the state of training in intrascleral fixation using a simulated eyeball 400 similar to the simulated eyeball 400 is used as the evaluation model 132. As a result, since it is easier to set parameters for the evaluation model 132 than a simulator using CG, training in intrascleral fixation can be performed at a practical level. Therefore, in the cataract surgery evaluation system 100, the proficiency of intrascleral fixation can be evaluated.
[0048] Furthermore, in this embodiment, a camera 120 is provided that captures video data Ia. This makes it possible to obtain process evaluation data Da by analyzing the video data Ia using the evaluation model 132. Therefore, the cataract surgery evaluation system 100 can evaluate the proficiency of intrascleral fixation surgery.
[0049] <2. Modifications> Next, a modification of the cataract surgery evaluation system 100 according to the above embodiment will be described.
[0050] [[Variation A]] In the above embodiment, for example, as shown in Fig. 11, the server device 130 may be omitted and the information processing device 110 may have the functions of the server device 130. In this case, the information processing device 110 may further include an evaluation model 132 and a learning module 133, and may further include a memory 134 that stores process data 134A and parameters 134B, and a memory 135 that stores a plurality of video data 135A, a plurality of video data 135B, and a plurality of video data 135C, as shown in Fig. 12. In this case, the process evaluation data Da can be derived by the information processing device 110.
[0051] [[Variation B]] In the above embodiment, for example, as shown in Fig. 13 and Fig. 14, communication between the information processing device 110 and the camera 120 may be performed via a network 150 different from the network 140. Also, in the above modification A, for example, as shown in Fig. 15 and Fig. 16, communication between the information processing device 110 and the camera 120 may be performed via a network 150 different from the network 140. In this case, the network 150 may be, for example, a short-range wireless communication such as Bluetooth (registered trademark). Even in such a configuration, the same effects as those of the above embodiment can be obtained.
[0052] [[Variation C]] In the above embodiment and its modified examples, a device for detecting an incident angle of a surgical instrument used in intrascleral fixation with respect to the simulated sclera portion 421 may be provided. In this case, the incident angle of a surgical instrument used in intrascleral fixation with respect to the simulated sclera portion 421 can be incorporated into the evaluation of the proficiency of intrascleral fixation, so that the evaluation of the proficiency of intrascleral fixation can be performed more precisely.
[0053] [[Variation D]] In the above embodiment and its modified examples, video data Ia obtained by a camera other than the camera 120 may be used. In this case, the camera other than the camera 120 is configured to capture an image of the surgical field from a direction different from the observation direction of the surgical field when the medical staff looks into the eyepiece 252. The camera other than the camera 120 may capture an image of the surgical field from a direction oblique to the surgical field to obtain video data Ia. In this way, when the surgical field is captured from a direction oblique to the surgical field, it is possible to know the incidence angle, etc. of the surgical instrument used in the intrascleral fixation surgery with respect to the simulated sclera portion 421 from the video data Ia even if a device for detecting the incidence angle, etc. of the surgical instrument used in the intrascleral fixation surgery with respect to the simulated sclera portion 421 is not provided. Therefore, the configuration of the cataract surgery evaluation system 100 can be simplified by the amount that the device for detecting the incidence angle, etc. can be omitted. [Explanation of symbols]
[0054] 100...cataract surgery evaluation system, 110...information processing device, 110A...display surface, 111...control unit, 112...memory, 112A...program, 113...display unit, 114...input unit, 115...network IF, 120...camera, 130...server device, 131...control unit, 132...evaluation model, 132A...input layer, 132B...intermediate layer, 132C...output layer, 133...learning module, 133A...learning data generation module, 133B...parameter generation module, 134...memory, 134A...process data, 134B...parameters, 135...memory, 135A, 135B, 135C...video data data, 136...network IF, 140,150...network, 200...ophthalmic surgery microscope system, 210...support, 220...first arm, 230...second arm, 240...driving device, 250...microscope for medical staff, 260...foot switch, 300...operating table, 400...simulated eyeball, 410...lower member, 411...shell-shaped main body portion, 411A...upper end surface, 412...recess portion, 420...upper member, 421...simulated sclera portion, 422...simulated cornea portion, 423...simulated iris portion, 500...intraocular lens, 510...main body lens portion, 520...connecting portion, Da...process evaluation data, Db...learning data, Ia...video data.
Claims
1. an evaluation model that, when video data obtained by imaging a practical training of intrascleral fixation using a simulated eyeball is input, classifies a plurality of chronological surgical steps constituting the intrascleral fixation included in the video data into periods corresponding to each of the surgical steps, evaluates the proficiency level for each of the classified surgical steps, and outputs the classification results and evaluation results for each of the surgical steps in response to the input of the video data; an acquisition unit that acquires the classification result and the evaluation result for each surgical step from the evaluation model by inputting the video data into the evaluation model; a display unit that displays the classification result and the evaluation result for each surgical procedure acquired by the acquisition unit; Equipped with Cataract surgery rating system.
2. The evaluation model is an artificial intelligence model that generates and outputs the classification result and the evaluation result by analyzing video data using parameters generated using learning video data obtained by imaging a state of a practical training of intrascleral fixation surgery using a mock eyeball similar to the mock eyeball. The cataract surgery evaluation system according to claim 1 .
3. The video data is acquired by capturing an image. The cataract surgery evaluation system according to claim 1 or 2.
Citation Information
Patent Citations
Surgical console operable for simulation of operation procedures
JP2008100060A
Universal Microsurgery Simulator
JP2015506726A
Medical activity training appropriateness evaluation device
JP2016080752A
Medical care procedure training system
JP2018036351A
Device for eyeball surgical operation training
JP2019028305A
