Endoscopic examination operation monitoring method and device

Abstract

The invention relates to an endoscopic examination operation monitoring method and device. The method comprises the steps that in the process that a to-be-evaluated person stretches an endoscope into a preset complete model of a target organ and conducts inspection operation on a preset target, an inspection operation video and the inspection operation duration of the to-be-evaluated person on the preset target are obtained; the preset target comprises at least one of a key part, a lesion marker and an inner wall; a preset lesion marker is arranged on the inner surface of the complete model of the target organ; and generating an endoscopic examination operation monitoring result of the to-be-evaluated person according to the examination operation video and the examination operation duration of the to-be-evaluated person on the preset target. By adopting the method, the problems that an existing endoscope inspection operation monitoring method needs to utilize virtual reality equipment, the cost is high, and popularization is not facilitated can be solved.

Description

Technical Field

[0001] This application relates to the field of endoscopic education and training technology, and in particular to a method and device for monitoring endoscopic examination operations. Background Technology

[0002] With the development of endoscopic technology, endoscopes have been widely used in the medical field. Performing endoscopic examinations of target areas (such as cystoscopy) is an essential clinical skill for endoscopic practitioners. To master this clinical skill, training in endoscopic examination procedures is necessary for endoscopic practitioners.

[0003] Existing training methods for endoscopic procedures mainly involve monitoring the endoscopic procedures performed by endoscopic practitioners, and then providing feedback on the quality of these procedures based on the monitoring results, in order to achieve the training objectives for endoscopic procedures.

[0004] Existing methods for monitoring endoscopic procedures mainly involve performing the procedures on animals or simulated patients. However, these methods often lack standardization, making it difficult to provide objective monitoring results and to provide feedback on problems in the procedures based on standardized monitoring results, thus hindering teaching and training.

[0005] Specifically, existing methods construct a 3D hand skeleton model and an improved bounding box by acquiring standard lesion information and examinee operation data in a simulated endoscopic scenario, and then calculate the similarity of the endoscope-holding posture, the endoscopic operation score, and the consistency of lesion judgment. The monitoring results of the endoscopic examination operation are determined by these factors. However, this method of monitoring endoscopic examination operations requires virtual reality equipment, which is costly and not conducive to widespread adoption.

[0006] There is currently no effective solution to the problem that existing methods for monitoring endoscopic examinations require the use of virtual reality equipment, which is costly and hinders widespread adoption. Summary of the Invention

[0007] Therefore, it is necessary to provide a method and apparatus for monitoring microscopic examination operations to address the aforementioned technical problems.

[0008] Firstly, this application provides a method for monitoring endoscopy procedures. The method includes:

[0009] During the process of the person to be evaluated inserting the endoscope into a complete model of a preset target organ and performing an examination on the preset target, the video of the person to be evaluated's examination of the preset target and the duration of the examination are acquired; the preset target includes at least one of key parts, lesion markers, and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers;

[0010] Based on the video of the person to be evaluated's examination of the preset target and the duration of the examination, the monitoring results of the person's endoscopic examination are generated.

[0011] In one embodiment, the complete model of the target organ is obtained by dividing the inner surface of the original model of the target organ into grids, labeling each grid region obtained after grid division to obtain grid markers for each grid region, and adsorbing the lesion markers onto the surface of the target grid region to obtain the model.

[0012] In one embodiment, generating the endoscopic examination operation monitoring results of the person to be evaluated based on the examination operation video and examination operation duration of the person to be evaluated for the preset target includes:

[0013] Based on the inspection operation video, a two-dimensional reconstruction unfolded diagram of the interior of the target organ detected by the person to be evaluated is generated.

[0014] Based on the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard diagram of the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0015] In one embodiment, before obtaining the monitoring results of the endoscopic examination operation of the person to be evaluated based on the two-dimensional reconstruction unfolded map inside the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard map inside the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, the process includes:

[0016] A standard two-dimensional reconstruction of the interior of the complete model of the target organ is obtained. The standard image is obtained by professional personnel inspecting and photographing the target of the complete model of the target organ according to a preset operating procedure, obtaining a professional inspection video of the target, and performing two-dimensional reconstruction of the interior of the complete model of the target organ based on the professional inspection video of the target.

[0017] In one embodiment, the monitoring results of the endoscopic examination operation of the person being evaluated are obtained based on the two-dimensional reconstructed unfolded map of the complete model of the target organ detected by the person being evaluated, the standard two-dimensional reconstructed map of the complete model of the target organ, and the duration of the examination operation performed by the person being evaluated. This includes:

[0018] Based on the two-dimensional reconstruction unfolded diagram of the interior of the target organ detected by the person to be evaluated, the total number of grid regions divided, the number of lesion markers set, the grid region numbering monitoring results and lesion marker monitoring results of the endoscopic examination operation of the person to be evaluated are determined.

[0019] Based on the two-dimensional reconstruction unfolded image of the interior of the complete model of the target organ detected by the person to be evaluated, and the two-dimensional reconstruction standard image of the interior of the complete model of the target organ, the image quality monitoring result of the endoscopic examination operation of the person to be evaluated is determined.

[0020] Based on the duration of the examination operation performed by the person to be evaluated, the preset standard operation duration, and the preset operation duration threshold, the operation time monitoring result of the endoscopic examination operation of the person to be evaluated is determined.

[0021] Based on the monitoring results of grid area numbering, lesion marker monitoring, image quality monitoring, and operation time monitoring of the endoscopic examination operation of the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0022] In one embodiment, the determination of the monitoring results of the grid region numbering and lesion marker monitoring results for the endoscopic examination operation of the person under evaluation, based on the two-dimensional reconstruction unfolded map of the interior of the target organ detected by the person under evaluation, the total number of grid regions divided, the number of lesion markers set, includes:

[0023] The grid region numbering monitoring result of the endoscopic examination operation of the person under evaluation is determined based on the ratio between the number of correct grid markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person under evaluation and the total number of grid regions.

[0024] The lesion marker monitoring result of the endoscopic examination operation of the person to be evaluated is determined by the ratio between the number of correct lesion markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated and the number of lesion markers set.

[0025] In one embodiment, determining the image quality monitoring results of the endoscopic examination procedure of the person being evaluated, based on the two-dimensional reconstructed unfolded map of the interior of the complete model of the target organ detected by the person being evaluated, and the two-dimensional reconstructed standard map of the interior of the complete model of the target organ, includes:

[0026] Based on the two-dimensional reconstruction unfolded image of the interior of the complete model of the target organ detected by the person to be evaluated, the brightness and sharpness of each grid region in the two-dimensional reconstruction unfolded image are determined;

[0027] The brightness and sharpness of each grid region in the standard two-dimensional reconstruction image inside the complete model of the target organ are used as the standard brightness and standard sharpness of each grid region.

[0028] Based on the brightness and sharpness of each grid region in the two-dimensional reconstruction unfolded image, as well as the standard brightness and standard sharpness of each grid region, the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstruction unfolded image are determined.

[0029] Based on the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstruction unfolded image, as well as the preset weights of the brightness monitoring results and sharpness monitoring results, the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image are determined.

[0030] Based on the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image, the image quality monitoring results of the endoscopic examination operation of the person to be evaluated are determined.

[0031] In one embodiment, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained based on the grid area numbering monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated, including:

[0032] Based on preset weights for grid area number monitoring results, preset weights for lesion marker monitoring results, preset weights for image quality monitoring results, and preset weights for operation time monitoring results, the weighted sums of the grid area number monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0033] In one embodiment, after generating the endoscopic examination operation monitoring results of the person to be evaluated based on the examination operation video and examination operation duration of the person to be evaluated on the preset target, the process includes:

[0034] The monitoring results of the endoscopic examination operation of the person to be evaluated are input into a preset reinforcement learning network, and suggestions for improvement of the endoscopic examination operation of the person to be evaluated are output.

[0035] Based on the monitoring results of the endoscopic examination operation of the person to be evaluated, and the suggestions for improvement of the endoscopic examination operation of the person to be evaluated, a monitoring result report of the endoscopic examination operation of the person to be evaluated is generated; the monitoring result report includes the scoring points, deduction points, and suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

[0036] Secondly, this application also provides an endoscopic examination operation monitoring device. The device includes:

[0037] The monitoring information acquisition module is used to acquire video footage and duration of the examination operation performed by the person to be evaluated on the preset target organ during the process of inserting the endoscope into a complete model of the preset target organ to perform an examination operation on the preset target; the preset target includes at least one of key parts, lesion markers, and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers;

[0038] And a monitoring result determination module, used to generate the monitoring results of the endoscopic examination operation of the person to be evaluated based on the video of the examination operation of the person to be evaluated on the preset target and the duration of the examination operation.

[0039] The aforementioned endoscopic examination operation monitoring method and device acquires video footage and the duration of the examination operation performed by the examinee inserting an endoscope into a complete model of a pre-set target organ. Based on this video footage and duration, it generates endoscopic examination operation monitoring results for the examinee. By performing the examination on a complete model of a pre-set target organ with pre-set lesion markers on its inner surface, it generates monitoring results for the endoscopic examination operation simply by acquiring the video footage and duration of the pre-set target examination operation. This eliminates the need for virtual reality equipment, making it simple, cost-effective, and easy to promote. It solves the problem that existing endoscopic examination operation monitoring methods require virtual reality equipment, resulting in higher costs and hindering widespread adoption.

[0040] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0042] Figure 1 A hardware structure block diagram of a terminal for an endoscopic examination operation monitoring method provided in an embodiment of this application;

[0043] Figure 2 A flowchart illustrating an embodiment of the endoscopic examination operation monitoring method provided in this application;

[0044] Figure 3 A flowchart of an endoscopy operation monitoring method provided in a preferred embodiment of this application;

[0045] Figure 4 This is a structural block diagram of an endoscopic examination operation monitoring device provided in an embodiment of this application. Detailed Implementation

[0046] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0047] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these” used in this application do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to these processes, methods, products, or devices. Words such as “connected,” “linked,” and “coupled” used in this application are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. Normally, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," "third," etc., used in this application are merely to distinguish similar objects and do not represent a specific order of objects.

[0048] The method embodiments provided in this example can be executed on a terminal, computer, or similar computing device. For example, it can run on a terminal of an endoscopy operation training device. Figure 1 This is a hardware structure block diagram of the terminal for the endoscopic examination operation monitoring method of this embodiment. (See diagram below.) Figure 1 As shown, a terminal may include one or more ( Figure 1Only one is shown in the diagram. A processor 102 and a memory 104 for storing data are also included. The processor 102 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown are illustrated.

[0049] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the endoscopic examination operation monitoring method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0050] The transmission device 106 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0051] This embodiment provides a method for monitoring endoscopic examination procedures. Figure 2 This is a flowchart of the endoscopic examination operation monitoring method in this embodiment, as follows: Figure 2 As shown, the process includes the following steps:

[0052] In step S210, during the process of the person to be evaluated inserting the endoscope into the complete model of the preset target organ and performing the examination operation on the preset target, the video of the examination operation and the duration of the examination operation are acquired; the preset target includes at least one of key parts, lesion markers and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers.

[0053] The aforementioned personnel to be evaluated may be students or other endoscopic operators requiring training in endoscopic procedures, i.e., operators who need to be observed, recorded, or evaluated by the system in real time. The aforementioned endoscope may refer to optical or electronic imaging instruments that can be inserted into natural cavities or surgical channels, such as gastroscopes, colonoscopes, and bronchoscopes. The aforementioned examination of the preset target may be performed according to a preset operating procedure. The aforementioned preset operating procedure may be specifically set according to different target organs and based on specific examination requirements; this embodiment does not impose specific limitations here. The aforementioned preset target organ may refer to a specific organ that needs to be reached and examined in advance, as set in the software or experimental procedure, such as "stomach," "colon," or "bronchus." The aforementioned complete model of the preset target organ may be a scaled-down simulation model of the target organ to be examined endoscopically. For example, if a cystoscopy is required, the complete model of the target organ is a bladder model, specifically a physical bladder model that is scaled down to the actual bladder and whose internal structure simulates the internal structure of a real bladder. The aforementioned pre-defined complete model of the target organ is a complete model obtained by pre-defined processing of the original model of the target organ. The original model of the target organ can be a model whose shape, size, and cavity orientation are consistent with the real organ, with no additional markings on its inner surface. The pre-defined processing can involve setting lesion markers or performing other operations at the target location of the original model of the target organ to make the simulated organ more closely resemble reality. The aforementioned examination procedures can be the systematic observation, photography, biopsy, or treatment of the pre-defined target by the person being evaluated using an endoscope. The aforementioned examination procedure videos can be multiple video clips taken by the person being evaluated of the pre-defined target. The aforementioned examination procedure duration can be the time elapsed from the moment the endoscope reaches the entrance of the target organ until it exits the target organ; that is, the total time spent by the person being evaluated from the start to the end of the examination.

[0054] Specifically, the aforementioned key locations can be anatomical landmarks such as the cardia, pylorus, and ileocecal valve. The aforementioned lesion markers can be artificially placed on the inner surface of the original model of the target organ to simulate lesions (such as black spots, erythema, tumors, polyps, etc.). The aforementioned inner wall can be any area on the inner surface of the original model of the target organ. Pre-set lesion markers are placed on the inner surface of the complete model of the target organ to facilitate assessment of whether the personnel being evaluated can correctly identify and treat these lesions, thereby determining their ability to identify and treat lesions.

[0055] Step S220: Based on the video of the person to be evaluated's inspection of the preset target and the duration of the inspection operation, generate the monitoring results of the person to be evaluated's endoscopic inspection operation.

[0056] In this step, the monitoring results of the above-mentioned endoscopic examination operation can be monitoring results of multi-dimensional data, such as the proportion of preset lesion markers being examined, the proportion of each region being captured, the image quality of the two-dimensional reconstruction unfolded map generated from the examination operation video based on the preset target, and whether the operation time of the examination operation is within the qualified range.

[0057] Steps S210 to S220 involve acquiring video footage and duration of the examination of a pre-defined target organ by inserting an endoscope into a complete model of the organ. Based on this video and duration, a monitoring result for the endoscopic examination is generated. This method, which examines a pre-defined target organ with pre-defined lesion markers on its inner surface, generates monitoring results simply by acquiring the video and duration of the examination. It eliminates the need for virtual reality equipment, making it simple, cost-effective, and easily implemented. This solves the problem that existing endoscopic examination monitoring methods require virtual reality equipment, resulting in higher costs and hindering widespread adoption.

[0058] In one embodiment, the complete model of the target organ is obtained by dividing the inner surface of the original model of the target organ into grids, labeling each grid region obtained after grid division to obtain grid markers for each grid region, and adsorbing the lesion markers onto the surface of the target grid region to obtain the model.

[0059] The above-mentioned method involves dividing the inner surface of the original model of the target organ into grids and labeling each grid region to obtain a grid label for each region. This can be achieved by dividing the inner surface of the original model of the target organ into grids according to different regions, and then labeling each grid region (e.g., using different numbers) to obtain a grid label for each region. Different labels (grid labels) represent different anatomical regions. These different anatomical regions can be different areas of the target organ. For example, when the target organ is the bladder, different anatomical regions may include the bladder wall, the trigone of the bladder, the ureteral orifice, the urethral orifice, etc. The above-mentioned grid division of the inner surface of the original model of the target organ can also be based on a preset size or a preset shape. The specific division method is not specifically limited in this embodiment, as long as the grid division of the inner surface of the original model allows for the determination of the proportion of each region that will be photographed during the examination by the person being evaluated.

[0060] The process of attaching the lesion marker to the surface of the target grid area can be achieved using micro-magnets, adhesive dots, or slots.

[0061] Specifically, in one embodiment, step S220 generates endoscopic examination operation monitoring results for the person to be evaluated based on the video of the examination operation performed by the person to be evaluated on the preset target and the duration of the examination operation, including:

[0062] Step S222: Based on the inspection operation video, generate a two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person to be evaluated.

[0063] In this step, the aforementioned generation of a complete model of the target organ detected by the person being evaluated, based on the examination operation video, can be achieved by processing the examination operation video using a 2D reconstruction algorithm to obtain the complete model of the target organ detected by the person being evaluated. The aforementioned 2D reconstruction algorithm can be one or more methods, such as using a trained 2D reconstruction network model or direct cylindrical unfolding.

[0064] Step S224: Based on the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard diagram of the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0065] In this step, the monitoring results of the endoscopic examination operation of the person under evaluation are obtained based on the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person under evaluation, the standard two-dimensional reconstruction map of the complete model of the target organ, and the duration of the examination operation performed by the person under evaluation. This can be based on the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person under evaluation, the total number of divided grid regions, the number of lesion markers set, to determine the grid region numbering and lesion marker monitoring results of the endoscopic examination operation of the person under evaluation. Furthermore, based on the examination results of the person under evaluation... The two-dimensional reconstruction unfolded image of the interior of the complete model of the target organ, the standard two-dimensional reconstruction image of the interior of the complete model of the target organ, the image quality monitoring results of the endoscopic examination operation of the person to be evaluated are determined, the operation time monitoring results of the endoscopic examination operation of the person to be evaluated are determined based on the duration of the examination operation of the person to be evaluated, the preset standard operation time, and the preset operation time threshold, and the grid area number monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0066] Steps S222 to S224 above involve constructing a two-dimensional reconstructed unfolded map of the internal structure of the target organ detected by the examinee based on the examination operation video. Then, based on this two-dimensional reconstructed unfolded map and the standard two-dimensional reconstructed map of the target organ, the coverage rate of the examination area during the examinee's operation is determined. Finally, considering the duration of the examination operation, the monitoring results of the examinee's endoscopic examination operation are obtained. By constructing a two-dimensional reconstructed unfolded map of the target organ, the completeness of the target organ examined by the examinee can be objectively reflected. Different colors can be used to highlight areas not detected by the examinee. This not only allows for data storage, facilitating review by trainees and improving training efficiency, but also allows for the storage of unfolded maps and scoring reports of all examinees to form a database. This allows for longitudinal comparison of the progress curves of the same person practicing multiple times, or horizontal comparison of differences between different individuals.

[0067] Additionally, in one embodiment, prior to step S224, the following steps are included:

[0068] Step S223: Obtain a two-dimensional reconstruction standard image of the interior of the complete model of the target organ; the standard image is obtained by professional personnel inspecting and photographing the target of the complete model of the target organ according to the preset operation procedure, obtaining a professional inspection video of the preset target, and performing two-dimensional reconstruction of the interior of the complete model of the target organ based on the professional inspection video of the preset target.

[0069] The aforementioned professionals can be senior endoscopists or examiners whose professional skills meet the preset requirements. The standard image obtained by reconstructing the interior of a complete model of the target organ using the professional examination video based on the preset target can be obtained by processing the professional examination video using a 2D reconstruction algorithm to obtain the 2D reconstructed standard image of the interior of the complete model of the target organ detected by the person being evaluated. The aforementioned professionals can be one or more. When there are multiple professionals, each professional can examine and film the target organ's complete model according to a preset operating procedure, generating multiple professional examination videos. Then, using these multiple professional examination videos, a 2D reconstructed image of the interior of the target organ's complete model can be generated for each individual professional ...

[0070] In one embodiment, step S224, based on the two-dimensional reconstruction unfolded image of the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard image of the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, obtains the monitoring results of the endoscopic examination operation of the person to be evaluated, including:

[0071] Step S2242: Based on the two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person to be evaluated, the total number of grid regions divided, the number of lesion markers set, the grid region number monitoring results and lesion marker monitoring results of the endoscopic examination operation of the person to be evaluated are determined.

[0072] In this step, the total number of grid regions mentioned above can be the number of grid regions after the original model of the target organ is gridded during the construction of the complete model of the target organ. The grid region numbering monitoring result can be the proportion of the number of correctly detected grid regions in the two-dimensional reconstruction unfolded image to the total number of grid regions. The lesion marker monitoring result can be the proportion of the number of lesion markers detected by the person being evaluated in the two-dimensional reconstruction unfolded image of the complete model of the target organ to the number of set lesion markers.

[0073] The above-mentioned monitoring results for determining the grid region numbering of the endoscopic examination operation of the person under evaluation can be based on the ratio between the number of correct grid markers in the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person under evaluation and the total number of grid regions.

[0074] The above-mentioned two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person under evaluation, and the number of lesion markers set, determine the lesion marker monitoring results of the endoscopic examination operation of the person under evaluation. This can be determined by the ratio between the number of correct lesion markers in the two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person under evaluation and the number of lesion markers set.

[0075] Step S2244: Based on the two-dimensional reconstruction unfolded map of the interior of the complete model of the target organ detected by the person to be evaluated, and the two-dimensional reconstruction standard map of the interior of the complete model of the target organ, determine the image quality monitoring results of the endoscopic examination operation of the person to be evaluated.

[0076] The aforementioned image quality monitoring results can be monitoring results related to the brightness and sharpness of the image. The aforementioned determination of image quality monitoring results for the endoscopic examination operation of the person being evaluated, based on the two-dimensional reconstructed unfolded image of the complete model of the target organ detected by the person being evaluated, and the two-dimensional reconstructed standard image of the complete model of the target organ, can be based on determining the brightness and sharpness of each grid region in the two-dimensional reconstructed unfolded image. Then, the brightness and sharpness of each grid region in the two-dimensional reconstructed standard image of the complete model of the target organ are used as the standard brightness and standard sharpness of each grid region. Based on the brightness and sharpness of each grid region in the two-dimensional reconstructed unfolded image, and the standard brightness and standard sharpness of each grid region, the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstructed unfolded image are determined. Finally, based on the brightness and sharpness monitoring results of each grid region in the two-dimensional reconstructed unfolded image, and the preset brightness monitoring result weights and sharpness monitoring result weights, the image quality monitoring results of each grid region in the two-dimensional reconstructed unfolded image are determined.

[0077] Step S2246: Based on the duration of the examination operation performed by the person to be evaluated, the preset standard operation duration, and the preset operation duration threshold, determine the operation time monitoring result of the endoscopic examination operation of the person to be evaluated.

[0078] In this step, the operation time monitoring results of the endoscopic examination operation of the person to be evaluated are determined based on the duration of the examination operation performed by the person to be evaluated, the preset standard operation time, and the preset operation time threshold. The specific calculation process is as follows:

[0079] ;

[0080] Among them, T ideal T is the preset operation time threshold. actual This is the preset standard operation time.

[0081] It should be noted that the preset operation time threshold and the preset standard operation time can be specifically set based on actual needs or actual application scenarios. This embodiment does not make specific limitations here, as long as the operation time monitoring results of the endoscopic examination operation of the person to be evaluated can be determined through the preset operation time threshold and the preset standard operation time.

[0082] Step S2248: Based on the monitoring results of the grid area numbering, lesion marker monitoring, image quality monitoring, and operation time monitoring of the endoscopic examination operation of the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0083] The monitoring results of the endoscopic examination operation of the person to be evaluated, which are based on the grid area number monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated, can be obtained by weighting and summing the grid area number monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the person to be evaluated based on preset weights, preset weights, preset weights, and preset weights, respectively.

[0084] Steps S2242 to S2248 above obtain the monitoring results of the endoscopic examination operation of the person being evaluated by using a two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person being evaluated, a standard two-dimensional reconstruction diagram of the complete model of the target organ, and the duration of the examination operation performed by the person being evaluated. By determining the monitoring results of the person being evaluated's endoscopic examination operation, the monitoring results of the endoscopic examination operation can be generated simply by acquiring the examination operation video and examination duration of a preset target, without the need for virtual reality equipment. This method is simple, easy to implement, low-cost, and easy to promote. It solves the problem that existing endoscopic examination operation monitoring methods require virtual reality equipment, are costly, and are not conducive to widespread adoption.

[0085] In another embodiment, step S2242, based on the two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person to be evaluated, the total number of divided grid regions, the number of set lesion markers, and determining the grid region numbering monitoring results and lesion marker monitoring results of the endoscopic examination operation of the person to be evaluated, includes:

[0086] Step S1: Based on the ratio between the number of correct grid markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated and the total number of divided grid regions, determine the grid region numbering monitoring result of the endoscopic examination operation of the person to be evaluated.

[0087] The ratio between the number of correct grid markers in the two-dimensional reconstructed unfolded diagram of the target organ detected by the evaluated person and the total number of divided grid regions is used to determine the grid region numbering monitoring results for the endoscopic examination operation of the evaluated person. , can be represented as:

[0088] ;

[0089] in, This represents the number of correctly identified grid regions in the 2D reconstruction unfolded image. This represents the total number of grid regions.

[0090] Step S2: Based on the ratio between the number of correct lesion markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated and the number of lesion markers set, determine the lesion marker monitoring results of the endoscopic examination operation of the person to be evaluated.

[0091] In this step, the ratio between the number of correct lesion markers in the two-dimensional reconstructed unfolded image of the complete model of the target organ detected by the person being evaluated, and the number of lesion markers set, is used to determine the lesion marker monitoring result of the endoscopic examination procedure of the person being evaluated. , can be represented as:

[0092] ;

[0093] Where, N true N represents the number of lesion markers set. detected This represents the number of lesion markers detected in the two-dimensional reconstruction unfolded image.

[0094] Steps S1 to S2 above determine the grid area number monitoring result of the endoscopic examination operation of the person being evaluated by comparing the ratio between the number of correct grid markers in the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person being evaluated and the total number of divided grid areas. Then, based on the ratio between the number of correct lesion markers in the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person being evaluated and the number of lesion markers set, the lesion marker monitoring result of the endoscopic examination operation of the person being evaluated is determined. The determination of the grid area number monitoring result and the lesion marker monitoring result facilitates the subsequent determination of the monitoring result of the endoscopic examination operation of the person being evaluated.

[0095] Further, in one embodiment, step S2244, based on the two-dimensional reconstructed unfolded image of the interior of the complete model of the target organ detected by the person to be evaluated, and the two-dimensional reconstructed standard image of the interior of the complete model of the target organ, determines the image quality monitoring result of the endoscopic examination operation of the person to be evaluated, including:

[0096] Step S3: Based on the two-dimensional reconstruction unfolded map of the interior of the complete model of the target organ detected by the person to be evaluated, determine the brightness and sharpness of each grid area in the two-dimensional reconstruction unfolded map.

[0097] Step S4: The brightness and sharpness of each grid region in the standard two-dimensional reconstruction of the interior of the complete model of the target organ are used as the standard brightness and standard sharpness of each grid region.

[0098] Step S5: Based on the brightness and sharpness of each grid region in the 2D reconstruction unfolded image, as well as the standard brightness and standard sharpness of each grid region, determine the brightness monitoring results and sharpness monitoring results of each grid region in the 2D reconstruction unfolded image.

[0099] The above-mentioned brightness monitoring results for each grid region in the two-dimensional reconstruction unfolded image are determined based on the brightness of each grid region and the standard brightness of each grid region. The specific process is as follows: the histogram similarity of the V channel (brightness) of the HSV (HueSaturation Value, three components of hue, saturation and brightness) color space is used to determine the brightness monitoring results for each grid region in the two-dimensional reconstruction unfolded image.

[0100] The specific calculation formula is as follows:

[0101] ;

[0102] in, This is the brightness monitoring result of the i-th grid region in the 2D reconstruction unfolded image. The results are the histogram similarity calculation. Let V be the histogram value of the color space V channel (luminance) of the i-th grid region in the 2D reconstructed unfolded image. This represents the histogram value of the V channel (luminance) in the color space of the i-th grid region in the 2D reconstructed standard image.

[0103] The histogram similarity calculation results mentioned above can be calculated using cosine similarity or Bach coefficient.

[0104] The process of calculating histogram similarity using cosine similarity is as follows:

[0105] ;

[0106] in, Let k be the value of the k-th bin (interval) in the brightness histogram of the i-th grid region in the 2D reconstructed unfolded image. Let K be the value of the k-th bin (interval) in the brightness histogram of the i-th grid region in the 2D reconstruction standard image. The above interval refers to a bar in the histogram, and K represents the total number of bars in the histogram.

[0107] In addition, the process of calculating histogram similarity using the Bach coefficient is as follows:

[0108] ;

[0109] Using the Bach coefficient to calculate histogram similarity would provide better robustness.

[0110] The above-mentioned sharpness monitoring results for each grid region in the 2D reconstruction unfolded image are determined based on the sharpness of each grid region and the standard sharpness of each grid region. The specific process is as follows: the sharpness monitoring results for each grid region in the 2D reconstruction unfolded image are calculated using the Laplace gradient method.

[0111] The specific calculation process is as follows:

[0112] ;

[0113] in, This represents the sharpness monitoring result for the i-th grid region in the 2D reconstructed unfolded image. Var() is an operator for calculating local variance, used to measure sharpness. To determine the sharpness of the i-th grid region in the 2D reconstruction unfolded image, Let represent the sharpness of the i-th grid region in the 2D reconstruction standard image. Let be the Laplacian operator (second derivative) of image I, used for edge detection.

[0114] It should be noted that, in order to avoid the sharpness monitoring results of each grid area in the 2D reconstruction unfolded image being too high due to overexposure or oversharpening, the sharpness monitoring results of each grid area in the 2D reconstruction unfolded image can be normalized.

[0115] The specific normalization process is as follows:

[0116] like >1, then take =1.

[0117] Step S6: Based on the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstruction unfolded image, as well as the preset weights of the brightness monitoring results and sharpness monitoring results, determine the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image.

[0118] The process of determining the image quality monitoring results for each grid region in the 2D reconstruction unfolded image based on the brightness and sharpness monitoring results of each grid region, as well as the preset weights for the brightness and sharpness monitoring results, is as follows:

[0119] ;

[0120] in, The image quality monitoring result for the i-th grid region is... The preset weights for the clarity monitoring results, The preset weights are used for the brightness monitoring results.

[0121] Step S7: Based on the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image, determine the image quality monitoring results of the endoscopic examination operation of the person to be evaluated.

[0122] The above-mentioned image quality monitoring results S for the endoscopic procedures performed on the personnel to be evaluated were determined. quality The specific process is as follows:

[0123] ;

[0124] Where N is the total number of grid regions.

[0125] Preferably, different weights can be set for clarity monitoring results and different weights for brightness monitoring results for different regions. In addition, the weights for clarity monitoring results and brightness monitoring results for different regions can be adjusted according to specific circumstances.

[0126] Specifically, the weight of the sharpness monitoring result for the i-th grid region The calculation process is as follows:

[0127] ;

[0128] in, Let α be the regional importance coefficient of the clarity monitoring result for the i-th grid region. For example, during cystoscopy, α = 1.2 for the triangular area and α = 1.0 for other areas, where α is the regional importance coefficient of the clarity monitoring result for the grid region.

[0129] Specifically, the weight of the brightness monitoring result for the i-th grid region The calculation process is as follows:

[0130] ;

[0131] in, is the regional importance coefficient of the brightness monitoring result of the i-th grid region.

[0132] Steps S3 to S7 above determine the image quality monitoring results of the endoscopic examination operation of the person being evaluated by using the two-dimensional reconstruction unfolded map of the interior of the complete model of the target organ detected by the person being evaluated and the two-dimensional reconstruction standard map of the interior of the complete model of the target organ. The determination of the image quality monitoring results facilitates the subsequent determination of the monitoring results of the endoscopic examination operation of the person being evaluated based on the image quality monitoring results.

[0133] In one embodiment, step S2248, based on the monitoring results of the grid area numbering, lesion marker monitoring, image quality monitoring, and operation time monitoring of the endoscopic examination operation of the person to be evaluated, obtains the monitoring results of the endoscopic examination operation of the person to be evaluated, including:

[0134] Step S8: Based on the preset weights of the grid area number monitoring results, the preset weights of the lesion marker monitoring results, the preset weights of the image quality monitoring results, and the preset weights of the operation time monitoring results, the grid area number monitoring results, the lesion marker monitoring results, the image quality monitoring results, and the operation time monitoring results of the endoscopic examination operation of the person to be evaluated are weighted and summed to obtain the monitoring results of the endoscopic examination operation of the person to be evaluated.

[0135] In this step, the monitoring results S of the endoscopic examination procedures of the person to be evaluated are obtained above. total The calculation process is as follows:

[0136] ;

[0137] in, Assign weights to the monitoring results for each pre-defined grid area. The pre-defined weights for monitoring results of lesion markers, The preset weights for image quality monitoring results, The preset weights are assigned to the monitoring results of operation time. , , and The sum of is 1. It should be noted that... , , and The value can be adaptively adjusted based on specific teaching needs, teaching objectives, or application scenarios. , , and The specific values ​​are not specifically limited in this embodiment.

[0138] In one embodiment, after step S220, the following is included:

[0139] Step S230: Input the monitoring results of the endoscopic examination operation of the person to be evaluated into the preset reinforcement learning network, and output suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

[0140] The monitoring results of the endoscopic examination procedures of the personnel under evaluation are the monitoring results before weighted summation, including the monitoring results of grid region numbering, lesion marker monitoring, image quality monitoring, and operation time monitoring. The pre-built reinforcement learning network is a pre-constructed network model capable of outputting suggestions for improvement of the endoscopic examination procedures of the personnel under evaluation based on the input monitoring results. These suggestions for improvement may include one or more of the following: grid training, lesion training, image training, and time training. When a suggestion for grid training is output, it indicates that the proportion of each region captured in the monitoring results of the personnel under evaluation's endoscopic examination procedures is low, requiring enhanced training on grid coverage. When a suggestion for lesion training is output, it indicates that the proportion of pre-defined lesion markers examined in the monitoring results of the personnel under evaluation's endoscopic examination procedures is low, requiring enhanced training on lesion marker examination. When a suggestion for image training is output, it indicates that the image quality of the generated two-dimensional reconstructed unfolded image in the monitoring results of the personnel under evaluation's endoscopic examination procedures is poor, requiring enhanced training on image quality during the examination. Because the image quality acquired during the examination can be fed back to the image quality of the 2D reconstructed unfolded image generated from the examination video with the preset target, the image quality of the generated 2D reconstructed unfolded image can be improved by enhancing the image quality during the examination. Additionally, when the output time training suggestion is given, it indicates that the operation time of the endoscopic examination for the person being evaluated is not ideal, and training to strengthen the control of the operation time is needed.

[0141] The aforementioned pre-defined reinforcement learning network is a minimalist and interpretable Q-learning framework used to automatically analyze examinees' weaknesses and provide suggestions for further training after an exam. The core components of this network include: state space, action space, reward function, update process, state discretization function, and policy function. The output is a state-action value matrix.

[0142] The state space described above is represented by a 4-dimensional vector:

[0143] ;

[0144] Wherein, S is the state space, which is the set of all possible states; G is the grid coverage rate (monitoring result of grid area numbering), which can be represented by values ​​in the range {0, 1} from poor to excellent; D is the lesion detection rate (monitoring result of lesion markers), which can be represented by values ​​in the range {0, 1} from poor to excellent; H is the image quality score (image quality monitoring result), which can be represented by values ​​in the range {0, 1} from poor to excellent; and T is the time efficiency score (monitoring result of time consumption), which can be represented by values ​​in the range {0, 1} from poor to excellent.

[0145] The state space described above uses a 4-bit binary vector (G, D, H, T) to discretize the results of an exam into 16 combinations to determine whether "grid coverage, lesion detection, image quality, and time efficiency" meet the standards.

[0146] The aforementioned action space A includes four discrete actions:

[0147] ;

[0148] Among them, a g To improve the action of grid cover training, a d To improve the movements in lesion detection training, a h To improve image quality during training, a t To improve the movements during operation time training.

[0149] The reward function described above is directly generated from the "real weaknesses" identified by experts. If the action precisely targets that weakness (i.e., the expert's suggestion is correct), the reward function increments by 1; otherwise, if the expert's suggestion is incorrect, the reward function decrements by 1. The specific reward function is as follows:

[0150] ;

[0151] Here, R(s, a) represents the immediate reward function obtained by the system after performing action a in state s, which is the evaluation feedback on whether performing action a in state s is correct. In this step, the above evaluation feedback has two results: +1 indicates that action a exactly hits the "real weakness" marked by the expert, indicating that the suggestion is correct; -1 indicates that action a does not hit the real weakness, indicating that the suggestion is wrong.

[0152] The above update process can be expressed by the formula:

[0153] ;

[0154] Here, Q(s, a) is the state-action value function, representing the expected improvement in exam score if action a is suggested while currently in state s. A positive Q(s, a) indicates that the training action is expected to significantly improve performance, while a negative Q(s, a) indicates that the training action is expected to be insufficient to compensate for weaknesses. The larger the absolute value of Q(s, a), the higher the priority. α is the learning rate, α∈[0,1], used to control the update step size; α defaults to 0.1. γ is the discount factor, γ∈[0,1], representing the future reward decay; γ defaults to 0.9. r represents the immediate reward, s′ represents the next state, and a′ represents the next action. Q(s′, a′) is the optimal Q-value for the next state, representing the expected improvement in monitoring results if action a′ is suggested while currently in state s′.

[0155] The immediate reward function is as follows:

[0156] ;

[0157] Where r(s, a) represents the reward value for the current state s if action a is suggested. This indicates the true weaknesses in the exam record, obtained based on expert annotations.

[0158] The above state discretization function for:

[0159] ;

[0160] in, The mapping function representing the inspection operation record τ to the state, where τ represents the operation trajectory of a single inspection. x is g, d, h or t, where g represents the value corresponding to state space G, d represents the value corresponding to state space D, h represents the value corresponding to state space H, and t represents the value corresponding to state space T.

[0161] The strategy function mentioned above is a greedy function. The greedy function π(s) can be expressed as:

[0162] ;

[0163] in, The exploration rate is set to 0.1 by default.

[0164] By setting a greedy function, the action with the largest Q value is selected with a high probability, while the action is selected randomly with a low probability. This avoids local optima, adapts to changes in the standard, and mitigates small sample bias.

[0165] When the monitoring results of the endoscopic examination operation of the person to be evaluated are input into the preset reinforcement learning network, the preset reinforcement learning network calculates the status based on the received monitoring results, generates vulnerability data, and generates a vulnerability report.

[0166] The calculation process for the weak point data w(τ) is as follows:

[0167] W(τ)={a∈A|Q(s, a)≤min a′ Q(s, a′) + θ};

[0168] Here, Q(s, a′) represents the expected value of how much the exam score can be improved in the future if action a′ is suggested, given the current state s. The tolerance threshold for identifying weaknesses is set to 0.2 by default.

[0169] The above output is a state-action value matrix, as shown in Table 1:

[0170]

[0171] The characteristics of the aforementioned pre-defined reinforcement learning network are as follows:

[0172] 1. It can be implemented in a very simple way.

[0173] The size of the state space is The action space size is |A| = 4. Therefore, only 16 × 4 = 64 floating-point numbers need to be stored.

[0174] 2. Enables efficient training.

[0175] The computational cost of a single update is O(1), and the data required for convergence is 100-200 exam records.

[0176] 3. Clear physical meaning: Negative Q value indicates an item that needs improvement, and the magnitude of the absolute value of Q value represents the priority of improvement.

[0177] Step S240: Based on the monitoring results of the endoscopic examination operation of the person to be evaluated and the suggestions for improvement of the endoscopic examination operation of the person to be evaluated, generate a monitoring result report of the endoscopic examination operation of the person to be evaluated; the monitoring result report includes the scoring points, deduction points and suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

[0178] Steps S230 to S240 involve inputting the monitoring results of the endoscopic examination operation of the person to be evaluated into a preset reinforcement learning network, outputting suggestions for improvement of the endoscopic examination operation of the person to be evaluated, and then generating a monitoring result report of the endoscopic examination operation of the person to be evaluated based on the monitoring results and suggestions for improvement. The generation of the monitoring result report allows the examination operator to see their shortcomings and areas for improvement through the report, thereby achieving teaching and training.

[0179] The present embodiment will now be described and illustrated through preferred embodiments.

[0180] Figure 3 This is a flowchart of a preferred embodiment of an endoscopic examination operation monitoring method provided in this application. Figure 3 As shown, this endoscopic examination operation monitoring method includes the following steps:

[0181] Step S301: Set the original model of the target organ;

[0182] Step S302: Mesh the inner surface of the original model and label each mesh region obtained after meshing to obtain the mesh label of each mesh region;

[0183] Step S303: Adsorb the lesion marker onto the surface of the target grid area to obtain a complete model of the target organ;

[0184] Step S304: During the process of the person to be evaluated inserting the endoscope into the complete model of the preset target organ and performing the examination operation on the preset target, the video of the examination operation and the duration of the examination operation are acquired; the preset target includes at least one of the following: key parts, lesion markers, and inner wall.

[0185] Step S305: Based on the inspection operation video, generate a two-dimensional reconstruction unfolded diagram of the interior of the complete model of the target organ detected by the person to be evaluated.

[0186] Step S306: Based on the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard diagram of the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

[0187] Step S307: Input the monitoring results of the endoscopic examination operation of the person to be evaluated into the preset reinforcement learning network, and output suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

[0188] Step S308: Based on the monitoring results of the endoscopic examination operation of the person to be evaluated and the suggestions for improvement of the endoscopic examination operation of the person to be evaluated, generate a monitoring result report of the endoscopic examination operation of the person to be evaluated; the monitoring result report includes the scoring points, deduction points, and suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

[0189] Steps S301 to S308 described above involve acquiring video footage and the duration of an examination of a pre-defined target organ by inserting an endoscope into a complete model of the organ. Based on this video and duration, a monitoring result for the endoscopic examination is generated. This method, which examines a pre-defined target organ with pre-defined lesion markers on its inner surface, generates monitoring results simply by acquiring the video and duration of the examination. It eliminates the need for virtual reality equipment, making it simple, cost-effective, and easily implemented. This solves the problem that existing endoscopic examination monitoring methods require virtual reality equipment, resulting in higher costs and hindering widespread adoption.

[0190] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0191] Based on the same inventive concept, this embodiment also provides an endoscopic examination operation monitoring device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. The terms "module," "unit," "subunit," etc., used below refer to combinations of software and / or hardware that achieve a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0192] In one embodiment, Figure 4 This is a structural block diagram of an endoscopic examination operation monitoring device provided in one embodiment of this application, as shown below. Figure 4 As shown, the endoscopic examination operation monitoring device includes:

[0193] The monitoring information acquisition module 42 is used to acquire the video of the examination operation and the duration of the examination operation of the person to be evaluated on the preset target organ during the process of the person to be evaluated inserting the endoscope into the complete model of the preset target organ to perform the examination operation on the preset target; the preset target includes at least one of key parts, lesion markers and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers.

[0194] And the monitoring result determination module 44 is used to generate the monitoring results of the endoscopic examination operation of the person to be evaluated based on the video of the inspection operation of the person to be evaluated on the preset target and the duration of the inspection operation.

[0195] The aforementioned endoscopic examination operation monitoring device acquires video footage and the duration of the examination operation performed by the examinee inserting an endoscope into a complete model of a preset target organ. Based on this data, it generates monitoring results for the examinee's endoscopic examination operation. This device operates on a complete model of a preset target organ with preset lesion markers on its inner surface. By simply acquiring video footage and the duration of the examination operation, it can generate monitoring results without the need for virtual reality equipment. It is simple, low-cost, and easily promoted. This solves the problem that existing endoscopic examination operation monitoring methods require virtual reality equipment, resulting in higher costs and hindering widespread adoption.

[0196] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0197] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0198] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0199] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for monitoring endoscopic examination procedures, characterized in that, The method includes: During the process of the person to be evaluated inserting the endoscope into a complete model of a preset target organ and performing an examination on the preset target, the video of the person to be evaluated's examination of the preset target and the duration of the examination are acquired; the preset target includes at least one of key parts, lesion markers, and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers; Based on the video of the person to be evaluated's examination of the preset target and the duration of the examination, the monitoring results of the person's endoscopic examination are generated.

2. The endoscopic examination operation monitoring method according to claim 1, characterized in that, The complete model of the target organ is obtained by dividing the inner surface of the original model of the target organ into grids, labeling each grid region obtained after grid division, obtaining grid markers for each grid region, and adsorbing the lesion markers onto the surface of the target grid region.

3. The endoscopic examination operation monitoring method according to claim 2, characterized in that, The step of generating endoscopic examination operation monitoring results for the person to be evaluated based on the examination operation video and examination operation duration of the person to be evaluated for the preset target includes: Based on the inspection operation video, a two-dimensional reconstruction unfolded diagram of the interior of the target organ detected by the person to be evaluated is generated. Based on the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated, the two-dimensional reconstruction standard diagram of the complete model of the target organ, and the duration of the examination operation performed by the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

4. The endoscopic examination operation monitoring method according to claim 3, characterized in that, Before obtaining the monitoring results of the endoscopic examination operation of the person under evaluation based on the two-dimensional reconstruction unfolded map inside the complete model of the target organ detected by the person under evaluation, the two-dimensional reconstruction standard map inside the complete model of the target organ, and the duration of the examination operation performed by the person under evaluation, the following steps are included: A standard two-dimensional reconstruction of the interior of the complete model of the target organ is obtained. The standard image is obtained by professional personnel inspecting and photographing the target of the complete model of the target organ according to a preset operating procedure, obtaining a professional inspection video of the target, and performing two-dimensional reconstruction of the interior of the complete model of the target organ based on the professional inspection video of the target.

5. The endoscopic examination operation monitoring method according to claim 3, characterized in that, The monitoring results of the endoscopic examination operation of the person under evaluation are obtained based on the two-dimensional reconstruction unfolded map of the complete model of the target organ detected by the person under evaluation, the standard two-dimensional reconstruction map of the complete model of the target organ, and the duration of the examination operation performed by the person under evaluation. These results include: Based on the two-dimensional reconstruction unfolded diagram of the interior of the target organ detected by the person to be evaluated, the total number of grid regions divided, the number of lesion markers set, the grid region numbering monitoring results and lesion marker monitoring results of the endoscopic examination operation of the person to be evaluated are determined. Based on the two-dimensional reconstruction unfolded image of the interior of the complete model of the target organ detected by the person to be evaluated, and the two-dimensional reconstruction standard image of the interior of the complete model of the target organ, the image quality monitoring result of the endoscopic examination operation of the person to be evaluated is determined. Based on the duration of the examination operation performed by the person to be evaluated, the preset standard operation duration, and the preset operation duration threshold, the operation time monitoring result of the endoscopic examination operation of the person to be evaluated is determined. Based on the monitoring results of grid area numbering, lesion marker monitoring, image quality monitoring, and operation time monitoring of the endoscopic examination operation of the person to be evaluated, the monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

6. The endoscopic examination operation monitoring method according to claim 5, characterized in that, The two-dimensional reconstruction unfolded diagram of the interior of the target organ based on the complete model detected by the person to be evaluated, the total number of divided grid regions, the number of set lesion markers, and the monitoring results of the grid region numbering and lesion marker monitoring of the endoscopic examination operation of the person to be evaluated include: The grid region numbering monitoring result of the endoscopic examination operation of the person under evaluation is determined based on the ratio between the number of correct grid markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person under evaluation and the total number of grid regions. The lesion marker monitoring result of the endoscopic examination operation of the person to be evaluated is determined by the ratio between the number of correct lesion markers in the two-dimensional reconstruction unfolded diagram of the complete model of the target organ detected by the person to be evaluated and the number of lesion markers set.

7. The endoscopic examination operation monitoring method according to claim 5, characterized in that, The determination of image quality monitoring results for the endoscopic examination operation of the person under evaluation, based on the two-dimensional reconstructed unfolded map of the interior of the complete model of the target organ detected by the person under evaluation, and the two-dimensional reconstructed standard map of the interior of the complete model of the target organ, includes: Based on the two-dimensional reconstruction unfolded image of the interior of the complete model of the target organ detected by the person to be evaluated, the brightness and sharpness of each grid region in the two-dimensional reconstruction unfolded image are determined; The brightness and sharpness of each grid region in the standard two-dimensional reconstruction image inside the complete model of the target organ are used as the standard brightness and standard sharpness of each grid region. Based on the brightness and sharpness of each grid region in the two-dimensional reconstruction unfolded image, as well as the standard brightness and standard sharpness of each grid region, the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstruction unfolded image are determined. Based on the brightness monitoring results and sharpness monitoring results of each grid region in the two-dimensional reconstruction unfolded image, as well as the preset weights of the brightness monitoring results and sharpness monitoring results, the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image are determined. Based on the image quality monitoring results of each grid region in the two-dimensional reconstruction unfolded image, the image quality monitoring results of the endoscopic examination operation of the person to be evaluated are determined.

8. The endoscopic examination operation monitoring method according to claim 5, characterized in that, The monitoring results of the endoscopic examination operation of the person to be evaluated are obtained based on the grid area numbering monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated, including: Based on preset weights for grid area number monitoring results, preset weights for lesion marker monitoring results, preset weights for image quality monitoring results, and preset weights for operation time monitoring results, the weighted sums of the grid area number monitoring results, lesion marker monitoring results, image quality monitoring results, and operation time monitoring results of the endoscopic examination operation of the person to be evaluated are obtained.

9. The method for monitoring endoscopic examination procedures according to any one of claims 1-8, characterized in that, After generating the monitoring results of the endoscopic examination operation of the person to be evaluated based on the examination operation video and examination operation duration of the preset target, the process includes: The monitoring results of the endoscopic examination operation of the person to be evaluated are input into a preset reinforcement learning network, and suggestions for improvement of the endoscopic examination operation of the person to be evaluated are output. Based on the monitoring results of the endoscopic examination operation of the person to be evaluated, and the suggestions for improvement of the endoscopic examination operation of the person to be evaluated, a monitoring result report of the endoscopic examination operation of the person to be evaluated is generated; the monitoring result report includes the scoring points, deduction points, and suggestions for improvement of the endoscopic examination operation of the person to be evaluated.

10. An endoscopic examination operation monitoring device, characterized in that, The device includes: The monitoring information acquisition module is used to acquire video footage and duration of the examination operation performed by the person to be evaluated on the preset target organ during the process of inserting the endoscope into a complete model of the preset target organ to perform an examination operation on the preset target; the preset target includes at least one of key parts, lesion markers, and inner wall; the inner surface of the complete model of the target organ is provided with preset lesion markers; And a monitoring result determination module, used to generate the monitoring results of the endoscopic examination operation of the person to be evaluated based on the video of the examination operation of the person to be evaluated on the preset target and the duration of the examination operation.

Images