System based on AI glasses and suitable for endoscope and display method

Abstract

The invention provides a system suitable for an endoscope based on AI glasses and a display method, and relates to the technical field of crossing of medical instruments and artificial intelligence wearable equipment. The system comprises an endoscope main body, a signal processing and transmitting module and an AI glasses terminal, the endoscope body collects real-time pictures of a target part, the signal processing and transmitting module carries out encoding processing on the pictures and then wirelessly sends the pictures, the AI glasses terminal receives, decodes and displays the pictures, meanwhile, target recognition and auxiliary prompting are achieved through the built-in AI processing unit, and integration of visual operation and AI auxiliary diagnosis and treatment is achieved. According to the display method, efficient and accurate endoscope operation is achieved through the steps of image collection, code transmission, decoding display, AI auxiliary recognition and brightness self-adaptive adjustment. The defects that a traditional endoscope display device is limited, transmission is unstable, and the AI fusion degree is low are overcome, the operation view of a doctor is liberated, the diagnosis and treatment precision and efficiency are improved, and the system is suitable for endoscope operation scenes such as hospital operating rooms and endoscope centers.

Description

Technical Field

[0001] This invention relates to the field of medical devices and artificial intelligence wearable devices, specifically to a system and display method based on AI glasses suitable for endoscopes. Background Technology

[0002] In clinical medicine, endoscopes (such as disposable gastrointestinal catheterization endoscopes and flexible bronchoscopes) have become core equipment for the diagnosis and catheterization of gastrointestinal diseases. Their disposable nature effectively solves the problems of high risk of cross-infection and high disinfection costs associated with traditional reusable endoscopes. However, in practical applications, existing disposable endoscopes still have many shortcomings in real-time display and auxiliary diagnostic functions that urgently need to be addressed:

[0003] Display devices have significant limitations: the real-time images from traditional disposable endoscopes rely on large screen monitors in the operating room or handheld terminals. Doctors need to frequently look up at the large screen or look down at the handheld terminal during the operation, resulting in a fragmented field of vision. They cannot take into account both the patient's body and the image information, which directly affects the accuracy of catheter placement and examination. At the same time, the handheld terminal is easily affected by shaking during operation, resulting in blurred images and further reducing the reliability of the operation.

[0004] Insufficient wireless transmission performance: Some existing technologies attempt to achieve wireless image transmission via Bluetooth or WiFi, but Bluetooth transmission suffers from high latency and low resolution, failing to meet the requirements of real-time diagnosis and treatment for image synchronization; WiFi transmission is easily interfered with by the complex electromagnetic environment of the operating room, resulting in poor signal stability, frequent image stuttering and interruption, which seriously affects the diagnosis and treatment process.

[0005] Low integration of AI technology: Existing medical AI glasses are mostly used in simple scenarios such as drug recognition and physiological parameter monitoring. They have not yet been deeply integrated with disposable endoscopes and cannot perform functions such as intelligent recognition of target structures and path planning prompts for the internal images of the human body captured by the endoscope. They still need to rely on the clinical experience of doctors to operate, which is not user-friendly for novice doctors and is prone to operational risks due to human judgment errors.

[0006] The equipment integration design is unreasonable: the signal transmission component of existing disposable endoscopes is mostly integrated with the operating handle, resulting in a large and heavy handle, which can easily cause fatigue for doctors during long-term operation and reduce flexibility. At the same time, the communication protocol between the transmission component and the display device lacks a unified standard, has poor compatibility, and is difficult to be compatible with different brands and types of display devices, thus limiting the scope of clinical application.

[0007] In summary, existing disposable endoscopes have significant shortcomings in terms of display methods, transmission stability, AI-assisted functions, and device integration. There is an urgent need for a technical solution that can achieve lightweight wireless display, stable transmission, and AI-assisted diagnosis and treatment in an integrated manner to improve the convenience, accuracy, and safety of endoscopic operations. Summary of the Invention

[0008] The purpose of this invention is to overcome the above-mentioned defects of the prior art and provide a system and display method for endoscopes based on AI glasses. It achieves stable wireless transmission of high-definition images through an independent signal processing and transmission module. Combining the wearable characteristics and edge computing capabilities of AI glasses, it realizes immersive display of real-time endoscope images and integration of AI-assisted diagnosis and treatment, freeing up the doctor's field of vision, reducing the difficulty of operation, and improving the efficiency and safety of diagnosis and treatment.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0010] In a first aspect, the present invention provides a system based on AI glasses suitable for endoscopes, comprising:

[0011] The endoscope body is used to insert into the target area and acquire real-time images of that area;

[0012] The signal processing and transmission module establishes a data connection with the endoscope body to receive the real-time image, perform encoding, compression, and anti-interference processing, and then transmit it wirelessly.

[0013] The AI ​​glasses terminal wirelessly pairs and communicates with the signal processing and transmission module to receive and decode the encoded real-time images, and presents them visually through the built-in display component. The AI ​​glasses terminal also has a built-in AI processing unit that performs target structure recognition, region marking, and operation assistance prompts on the real-time images, realizing the integration of visualization of endoscopic operations and AI-assisted diagnosis and treatment.

[0014] As a further improvement to the technical solution of the present invention, the endoscope body includes a disposable flexible lens end, a flexible catheter, and a handheld operating handle, wherein the flexible lens end is fixedly connected to the handheld operating handle through the flexible catheter;

[0015] The soft lens end has a built-in high-definition camera, a fill light component, and an angle adjustment component. The high-definition camera is used to capture real-time images, the fill light component is used to provide a shooting light source, and the angle adjustment component is used to adjust the shooting angle.

[0016] The handheld operating handle has control buttons on its surface and a data transmission line inside. The control buttons are electrically connected to the fill light component and the angle adjustment component to control the working status of the components. The data transmission line is connected to the signal processing and transmission module through an end interface.

[0017] As a further improvement to the technical solution of the present invention, the soft lens end is made of medical-grade disposable polymer material and has a waterproof and anti-fog coating on the surface; the high-definition camera has a resolution of ≥1080P and a frame rate of ≥30fps; the fill light component is an LED fill light, and the angle adjustment component can achieve ±180° deflection adjustment.

[0018] As a further improvement to the technical solution of the present invention, the signal processing and transmission module is an independent transmitter box, which integrates a video encoding chip, a dual-frequency wireless communication module, an anti-interference processing unit, and a power supply component.

[0019] The video encoding chip is used to compress the real-time video into H.265 encoding format;

[0020] The dual-band wireless communication module includes a 2.4G / 5G WiFi module and a Bluetooth 5.3 module, which can be selected or automatically switched for wireless transmission;

[0021] The anti-interference processing unit is used to filter electromagnetic interference signals to ensure that the transmission delay is ≤50ms.

[0022] The power supply component is a rechargeable lithium battery with a capacity of ≥5000mAh and a battery life of ≥4 hours.

[0023] As a further improvement to the technical solution of the present invention, the external dimensions of the transmitter box are ≤10cm×8cm×3cm, and it adopts a lightweight and portable design; the transmitter box is detachably connected to the end interface of the handheld operating handle of the endoscope body via a Type-C connection cable.

[0024] As a further improvement to the technical solution of the present invention, the AI ​​glasses terminal includes a micro-display optical module, an edge AI processor, a wireless receiving module, an environmental perception component, and an audio output component;

[0025] The micro-display optical module is a holographic waveguide lens used to project the decoded real-time image, with a display resolution ≥720P and a viewing angle ≥40°.

[0026] The edge AI processor has a computing power of ≥8 TOPS and is used to run a preset target recognition model to perform semantic segmentation and recognition on real-time images.

[0027] The environmental sensing component is an ambient light sensor, used to detect the intensity of ambient light.

[0028] The audio output component is a bone conduction headset, used to output voice prompts.

[0029] As a further improvement to the technical solution of the present invention, the preset target recognition model is a digestive tract structure semantic segmentation model, and the tissue regions used for recognition include the esophagus, stomach, and duodenum; the target location and danger area of ​​the tube placement are marked; the light intensity signal detected by the ambient light sensor is used to automatically adjust the brightness of the holographic waveguide lens, with an adjustment range of 100-1000 cd / m²; the voice prompt information includes tube placement path deviation and danger area warning.

[0030] As a further improvement to the technical solution of the present invention, the AI ​​glasses terminal is also equipped with a device linkage interface, which supports establishing a communication connection with an external display device to realize the synchronous sharing of real-time images. The external display device is an operating room screen or a medical monitor.

[0031] Secondly, the present invention provides a display method for endoscopes based on AI glasses, comprising the following steps:

[0032] S1: Assemble and debug the system. Connect the handheld operating handle of the endoscope body to the signal processing and transmission module via a connecting cable. Wirelessly pair the AI ​​glasses terminal with the signal processing and transmission module and calibrate the screen resolution and display parameters.

[0033] S2: Insert the flexible lens end of the endoscope body into the target area, and activate the high-definition camera and lighting component through the control button to capture real-time images of the target area;

[0034] S3: The handheld operating handle transmits the real-time image to the signal processing and transmission module, which compresses it into H.265 format through the video encoding chip, filters electromagnetic interference through the anti-interference processing unit, and then transmits it through the dual-frequency wireless communication module.

[0035] S4: The AI ​​glasses terminal receives the compressed image through a wireless receiving module, decodes it, and then visualizes it through a holographic waveguide lens.

[0036] S5: The edge AI processor runs a preset target recognition model, performs semantic segmentation on the real-time image, identifies the target structure and marks the target location and danger zone, and outputs corresponding voice prompts through bone conduction headphones;

[0037] S6: The ambient light sensor detects the ambient light intensity in real time and automatically adjusts the brightness of the holographic waveguide lens until the operation is complete.

[0038] As a further improvement to the technical solution of the present invention, in step S3, the dual-band wireless communication module preferentially uses the 5G WiFi module for transmission; when a transmission delay ≥ 50ms is detected, it is determined that the electromagnetic interference exceeds the preset threshold, and the module is automatically switched to Bluetooth 5.3 for low-power transmission, at which time the screen resolution is adaptively adjusted to 480P.

[0039] The technical solution of the present invention has the following advantages over the prior art:

[0040] This invention achieves integrated real-time endoscopic image acquisition, stable wireless transmission, and AI-assisted diagnosis and treatment through the collaborative design of the endoscope body, signal processing and transmission module, and AI glasses terminal. Its core benefits are: it completely breaks through the limitations of traditional endoscopes that rely on large screens or handheld terminals; the wearable nature of the AI ​​glasses allows doctors to clearly view real-time images of target areas while directly looking at the patient, avoiding visual field fragmentation and significantly improving operational flexibility and accuracy; the independent design and anti-interference technology of the signal processing and transmission module ensure low-latency transmission of high-definition images, effectively overcoming electromagnetic interference in the operating room environment and meeting the needs of real-time clinical diagnosis and treatment; the edge computing capabilities of the AI ​​glasses terminal enable intelligent recognition of target structures, marking of dangerous areas, and voice prompts for path deviations, reducing doctors' reliance on clinical experience and improving operational safety and success rates; simultaneously, the lightweight design and disposable lens structure reduce operational fatigue, eliminate the risk of cross-infection, and support linkage with external medical equipment without requiring modifications to existing facilities, adapting to various clinical scenarios and possessing broad application value. Attached Figure Description

[0041] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0042] Figure 1 This is a schematic diagram of the overall structure of an AI glasses-based endoscope system according to an embodiment of the present invention;

[0043] Figure 2 This is a flowchart illustrating a display method for endoscopes based on AI glasses, as described in an embodiment of the present invention.

[0044] In the attached diagram: 1-Endoscope body (including disposable flexible lens end); 2-Handheld operating handle; 3-Type-C connection cable; 4-Signal processing and transmission module (transmitter box); 5-AI glasses terminal; 6-External display device (operating room large screen). Detailed Implementation

[0045] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0046] The present invention will be further described in detail below with reference to the accompanying drawings.

[0047] Reference Figure 1 In a first aspect, the present invention provides a system based on AI glasses suitable for endoscopes, comprising:

[0048] Endoscope body 1, used to insert into the target area and acquire real-time images of that area;

[0049] The signal processing and transmission module 4 establishes a data connection with the endoscope body 1, and is used to receive the real-time image, perform encoding compression and anti-interference processing, and then transmit it through wireless communication.

[0050] AI glasses terminal 5 wirelessly pairs and communicates with the signal processing and transmission module 4 to receive and decode the encoded real-time image, and present it visually through the built-in display component. The AI ​​glasses terminal 5 also has a built-in AI processing unit that performs target structure recognition, region marking and operation assistance prompts on the real-time image, realizing the integration of visualization of endoscopic operation and AI-assisted diagnosis and treatment.

[0051] Figure 1 In this system, the flexible lens end of the endoscope body 1 is used to insert into the target site of the patient and is connected to the handheld operating handle 2 via a flexible catheter; the handheld operating handle 2 establishes a data connection with the signal processing and transmission module 4 via a connecting cable 3; the signal processing and transmission module 4 sends the encoded image to the AI ​​glasses terminal 5 via a wireless signal; the holographic waveguide lens of the AI ​​glasses terminal 5 displays the real-time image, and at the same time performs structural recognition and voice prompts through the edge AI processor; the system can be connected to an external linkage display device 6 through the device linkage interface of the AI ​​glasses terminal 5 to achieve synchronous image sharing.

[0052] It should be noted that the system of the present invention operates through the coordinated operation of three core modules. After the endoscope body 1 is inserted into the target area, its built-in acquisition component acquires the real-time image of that area. The signal processing and transmission module 4 establishes a data connection with the endoscope body 1, receives the real-time image, compresses it through an encoding chip, filters environmental interference signals through an anti-interference unit, and finally wirelessly transmits the processed image data through a dual-frequency wireless communication module. After the AI ​​glasses terminal 5 completes wireless pairing with the signal processing and transmission module 4, it receives and decodes the encoded data, visualizes the image through its built-in display component, and simultaneously runs a preset recognition model to identify and mark the target structure in the real-time image, and outputs operation assistance prompts through an audio component, ultimately realizing the integration of visualization of endoscopic operation and AI-assisted diagnosis and treatment. This invention, through modular collaborative design, breaks through the limitations of traditional endoscopes that rely on large screens or handheld terminals for display, achieving a deep integration of real-time image acquisition, stable transmission, and AI assistance. The wearable nature of the AI ​​glasses allows doctors to avoid frequently switching fields of view, balancing patient observation and image viewing, significantly improving operational flexibility. Anti-interference processing and dual-frequency transmission design ensure the stability and real-time performance of image transmission, while AI assistance reduces the reliance on experience in operation, thus improving overall diagnostic accuracy and safety, and adapting to various clinical endoscopic operation scenarios.

[0053] In some embodiments, the endoscope body 1 includes a disposable flexible lens tip, a flexible catheter, and a handheld operating handle 2, wherein the flexible lens tip is fixedly connected to the handheld operating handle 2 via the flexible catheter;

[0054] The soft lens end has a built-in high-definition camera, a fill light component, and an angle adjustment component. The high-definition camera is used to capture real-time images, the fill light component is used to provide a shooting light source, and the angle adjustment component is used to adjust the shooting angle.

[0055] The handheld operating handle 2 has control buttons on its surface and a data transmission line inside. The control buttons are electrically connected to the fill light component and the angle adjustment component to control the working status of the components. The data transmission line is connected to the signal processing and transmission module 4 through an end interface.

[0056] It should be noted that the endoscope body 1 consists of a disposable flexible lens end, a flexible catheter, and a handheld operating handle 2. The flexible lens end is connected to the operating handle via the flexible catheter, forming a complete acquisition and control link. The high-definition camera on the flexible lens end is responsible for acquiring images of the target area, the supplementary lighting component provides a light source adapted to the shooting environment, and the angle adjustment component can adjust the shooting angle according to operational needs. The control buttons on the surface of the handheld operating handle 2 are electrically connected to the supplementary lighting component and the angle adjustment component. The control signals are triggered by the buttons to adjust the working status of the components. The transmission line inside the handle transmits the acquired image data to the signal processing and transmission module 4 through the end interface. This invention adopts a combination design of disposable lens + reusable handle, which avoids cross-infection through the disposable structure and reduces the cost of equipment use. The multi-component integrated design of the flexible lens end ensures the clarity and flexibility of image acquisition, the button layout of the operating handle facilitates real-time adjustment by doctors, and the built-in design of the data transmission line improves the safety and stability of equipment use. The overall structure is compact and easy to operate, adapting to the complex clinical endoscopic operation needs.

[0057] In some embodiments, the soft lens end is made of medical-grade disposable polymer material and has a waterproof and anti-fog coating on its surface; the high-definition camera has a resolution of ≥1080P and a frame rate of ≥30fps; the fill light component is an LED fill light, and the angle adjustment component can achieve ±180° deflection adjustment.

[0058] It should be noted that the soft lens is made of medical-grade disposable polymer material, and its waterproof and anti-fog coating can resist the effects of humid environments such as the digestive tract, preventing image blurring. The high-definition camera captures images at a resolution of ≥1080P and a frame rate of ≥30fps, ensuring clear and smooth image details. The LED fill light provides an adjustable shooting light source to adapt to the needs of capturing images under different lighting conditions. The angle adjustment component achieves ±180° deflection through an internal transmission structure, allowing flexible adjustment of the shooting angle to cover different observation angles of the target area. The design of medical-grade disposable materials and waterproof and anti-fog coating completely eliminates the risk of cross-infection while ensuring reliability in humid environments. The high-definition camera's parameter configuration ensures high-definition and smooth images, supporting doctors in accurately diagnosing conditions. The ±180° angle adjustment range enhances operational flexibility, accurately capturing key details of the target area, and the adaptability design of the fill light component further optimizes the image acquisition quality.

[0059] In some embodiments, the signal processing and transmission module 4 is an independent transmitter box, which integrates a video encoding chip, a dual-frequency wireless communication module, an anti-interference processing unit, and a power supply component.

[0060] The video encoding chip is used to compress the real-time video into H.265 encoding format;

[0061] The dual-band wireless communication module includes a 2.4G / 5G WiFi module and a Bluetooth 5.3 module, which can be selected or automatically switched for wireless transmission;

[0062] The anti-interference processing unit is used to filter electromagnetic interference signals to ensure that the transmission delay is ≤50ms.

[0063] The power supply component is a rechargeable lithium battery with a capacity of ≥5000mAh and a battery life of ≥4 hours.

[0064] It should be noted that the signal processing and transmission module 4 is an independent transmitter box. Its built-in video encoding chip compresses the received raw image into H.265 format, reducing bandwidth usage while maintaining image quality. The dual-band wireless communication module includes a 2.4G / 5G WiFi module and a Bluetooth 5.3 module, which can be used selectively or automatically switched depending on the transmission environment to meet the transmission needs of different scenarios. The anti-interference processing unit filters electromagnetic interference signals, controlling transmission latency to ≤50ms. A rechargeable lithium battery provides ≥4 hours of battery life, ensuring the needs of extended clinical operations. The independent design separates the module from the endoscope handle, reducing the handle's size and weight and improving operational flexibility. The H.265 encoding format achieves a balance between image quality and transmission efficiency, while the dual-band transmission module enhances scenario adaptability. The anti-interference processing unit effectively resists the influence of the complex electromagnetic environment of the operating room, ensuring low-latency transmission required for real-time diagnosis and treatment. The high-capacity lithium battery design meets the needs of long surgeries, eliminating the need for frequent charging and improving the convenience of clinical use.

[0065] In some embodiments, the transmitter box has external dimensions of ≤10cm×8cm×3cm and adopts a lightweight and portable design; the transmitter box is detachably connected to the end interface of the handheld operating handle 2 of the endoscope body 1 via a Type-C connection cable 3.

[0066] It should be noted that the transmitter box adopts a lightweight design, with dimensions of ≤10cm×8cm×3cm, making it easy to carry and place in clinical settings. A detachable connection is achieved via a Type-C cable 3 to the end interface of the handheld operating handle 2 on the end of the endoscope body 1. The connection process is simple and convenient, and the Type-C interface is characterized by strong universality and stable transmission, ensuring reliable data transmission. The compact size and lightweight design enhance the portability of the device, facilitating flexible deployment in operating rooms, endoscopy centers, and other similar settings. The detachable Type-C connection simplifies system assembly and disassembly processes, reducing operational difficulty. Furthermore, the universality of the Type-C interface enhances the device's adaptability, making it compatible with various endoscope handles that conform to interface standards, thus improving the device's reusability and promotional value.

[0067] In some embodiments, the AI ​​glasses terminal 5 includes a micro-display optical module, an edge AI processor, a wireless receiving module, an environmental perception component, and an audio output component;

[0068] The micro-display optical module is a holographic waveguide lens used to project the decoded real-time image, with a display resolution ≥720P and a viewing angle ≥40°.

[0069] The edge AI processor has a computing power of ≥8 TOPS and is used to run a preset target recognition model to perform semantic segmentation and recognition on real-time images.

[0070] The environmental sensing component is an ambient light sensor, used to detect the intensity of ambient light.

[0071] The audio output component is a bone conduction headset, used to output voice prompts.

[0072] It should be noted that the micro-display optical module of the AI ​​glasses terminal 5 uses holographic waveguide lenses to project the decoded real-time image to the doctor's field of vision at a resolution of ≥720P and a viewing angle of ≥40°, achieving an immersive display. The edge AI processor runs a preset target recognition model with a computing power of ≥8 TOPS to perform semantic segmentation and recognition on the real-time image, quickly extracting target structural information. The ambient light sensor detects the ambient light intensity in real time, providing data support for adjusting the screen brightness. The bone conduction headphones, as an audio output component, output operation assistance prompts in voice form, without affecting communication between doctors and medical teams. The high resolution and wide viewing angle design of the holographic waveguide lenses ensure clear visibility and coverage of the image, allowing doctors to clearly view the image while looking directly at the patient. The high-performance edge AI processor ensures the speed and accuracy of target recognition, providing computing power support for real-time assistance. The ambient light sensor enables adaptive adjustment of screen brightness, improving visual comfort under different lighting conditions. The audio output method of the bone conduction headphones avoids the problem of ear blockage, ensuring the dual needs of doctor-patient communication and operation prompts.

[0073] In some embodiments, the preset target recognition model is a digestive tract structure semantic segmentation model, and the tissue regions used for recognition include the esophagus, stomach, and duodenum; the tube placement target location and danger areas are marked; the light intensity signal detected by the ambient light sensor is used to automatically adjust the brightness of the holographic waveguide lens, with an adjustment range of 100-1000 cd / m²; the voice prompt information includes tube placement path deviation and danger area warning.

[0074] It should be noted that the preset target recognition model adopts a digestive tract structure semantic segmentation model. Through deep learning algorithms, it performs pixel-level recognition of tissue regions such as the esophagus, stomach, and duodenum in real-time images, accurately distinguishing target locations from danger zones and marking them accordingly. An ambient light sensor transmits the detected light intensity signal to the control unit, which automatically adjusts the brightness of the holographic waveguide lens according to a preset threshold, maintaining the brightness within a comfortable range of 100-1000 cd / m². The bone conduction headphones output voice prompts such as catheter placement path deviation and danger zone warnings based on the AI ​​recognition results and operational deviations, guiding doctors to adjust their procedures. The accurate recognition capability of the digestive tract structure semantic segmentation model allows doctors to quickly locate target positions and avoid danger zones, reducing reliance on clinical experience, especially suitable for novice doctors. The adaptive brightness adjustment function ensures clear visibility in both strong and low light environments, improving operational comfort and accuracy. Targeted voice prompts provide real-time guidance for doctors to adjust their procedures, reducing operational deviations and medical risks, further improving the safety and success rate of catheter placement.

[0075] In some embodiments, the AI ​​glasses terminal 5 is also provided with a device linkage interface, which supports establishing a communication connection with an external display device 6 to realize the synchronous sharing of real-time images. The external display device 6 is an operating room large screen or a medical monitor.

[0076] It should be noted that the device linkage interface of the AI ​​glasses terminal 5 supports communication with external display devices 6 such as operating room screens and medical monitors. Through the interface protocol, the decoded real-time image is synchronously transmitted to the external device, enabling shared display and meeting the needs of collaborative diagnosis and treatment by medical teams. The device linkage function breaks the limitations of a single display, allowing all members of the medical team to view the real-time image through the external display device 6, facilitating communication, collaboration, and exchange of opinions among team members. Linkage can be achieved without modifying existing medical facilities, reducing the cost of clinical application modifications and improving the system's adaptability and practicality, making it particularly suitable for collaborative diagnosis and treatment in complex surgical scenarios.

[0077] Reference Figure 2 Secondly, the present invention provides a display method for endoscopes based on AI glasses, comprising the following steps:

[0078] S1: Assemble and debug the system. Connect the handheld operating handle 2 of the endoscope body 1 to the signal processing and transmission module 4 via the connecting cable. Wirelessly pair the AI ​​glasses terminal 5 with the signal processing and transmission module 4 and calibrate the screen resolution and display parameters.

[0079] S2: Insert the flexible lens end of the endoscope body 1 into the target area, and activate the high-definition camera and supplementary lighting component through the control button to capture real-time images of the target area;

[0080] S3: The handheld operating handle 2 transmits the real-time image to the signal processing and transmission module 4, which compresses it into H.265 format through the video encoding chip, and then filters electromagnetic interference through the anti-interference processing unit before sending it through the dual-frequency wireless communication module.

[0081] S4: AI glasses terminal 5 receives the compressed image through a wireless receiving module, decodes it, and then visualizes it through a holographic waveguide lens.

[0082] S5: The edge AI processor runs a preset target recognition model, performs semantic segmentation on the real-time image, identifies the target structure and marks the target location and danger zone, and outputs corresponding voice prompts through bone conduction headphones;

[0083] S6: The ambient light sensor detects the ambient light intensity in real time and automatically adjusts the brightness of the holographic waveguide lens until the operation is complete.

[0084] In practice, the system is first assembled and debugged. The handheld operating handle 2 is connected to the signal processing and transmission module 4, and the AI ​​glasses terminal 5 is wirelessly paired with the signal processing and transmission module 4 to complete the calibration of the screen resolution and display parameters. Then, the soft lens end is inserted into the target area, and the acquisition component is started by controlling the button to acquire the real-time image of the target area. The handheld operating handle 2 transmits the image data to the signal processing and transmission module 4. After encoding, compression and anti-interference processing, it is transmitted by the dual-frequency wireless communication module. The AI ​​glasses terminal 5 receives the data and decodes it, and presents the image through the holographic waveguide lens. The edge AI processor runs the recognition model to perform semantic segmentation of the image, mark the target location and danger area, and output voice prompts through the bone conduction headphones. The ambient light sensor detects the light intensity in real time and automatically adjusts the screen brightness until the operation is completed.

[0085] The display method of this invention ensures the standardization and consistency of system operation through standardized step design, reducing the probability of operational errors; the end-to-end design from acquisition, transmission, display to AI assistance achieves seamless connection of each link, ensuring the continuity and real-time nature of diagnosis and treatment; the integration of adaptive brightness adjustment and AI voice prompts further improves the convenience and accuracy of operation. The overall process is scientific and reasonable, and is suitable for various clinical endoscopic operation scenarios such as gastrointestinal tube placement.

[0086] In some embodiments, in step S3, the dual-band wireless communication module preferentially uses the 5G WiFi module for transmission; when a transmission delay ≥ 50ms is detected, it is determined that the electromagnetic interference exceeds a preset threshold, and it automatically switches to the Bluetooth 5.3 module for low-power transmission, at which time the screen resolution is adaptively adjusted to 480P.

[0087] It should be noted that during image transmission, the dual-band wireless communication module prioritizes the 5G WiFi module for transmission to ensure high image quality and low latency. The signal processing and transmission module 4 monitors transmission latency in real time. When a latency ≥50ms is detected, it determines that electromagnetic interference exceeds a preset threshold and automatically switches to the Bluetooth 5.3 module for low-power transmission. Simultaneously, the image resolution is adaptively adjusted to 480P to ensure transmission continuity. The design prioritizing 5G WiFi transmission ensures the high image quality and low latency requirements in normal environments, meeting the core requirements of real-time diagnosis and treatment. The latency monitoring and automatic switching mechanism enables the system to quickly adapt to the complex electromagnetic environment of the operating room, avoiding image stuttering or interruption due to interference and ensuring operational continuity. The adaptive resolution adjustment achieves a balance between transmission stability and image quality, providing a clear and discernible image even in interference environments, ensuring the smooth conduct of diagnosis and treatment.

[0088] To provide a clearer understanding of the invention, the invention is further described below:

[0089] 1. An AI-based endoscope system, comprising three core modules: endoscope body 1, signal processing and transmission module 4, and AI glasses terminal 5. These modules work together via physical connection and wireless communication, with the specific structure as follows:

[0090] (1) Endoscope body 1

[0091] As the core of image acquisition, the endoscope body 1 adopts a combination design of disposable lens and reusable handle, including disposable flexible lens end, flexible catheter and handheld operating handle 2:

[0092] Disposable soft lens tip: Made of medical-grade disposable polymer material, it can be disposed of directly as medical waste after use, avoiding cross-infection. Its front end features a built-in high-definition camera (resolution ≥1080P, frame rate ≥30fps), an LED fill light (providing a stable shooting light source), and an angle adjustment component (allowing ±180° deflection to adapt to different shooting needs). The lens surface has a waterproof and anti-fog coating, suitable for humid environments such as the digestive tract, ensuring clear images.

[0093] Flexible catheter: Made of medical flexible material, with embedded data transmission lines and angle adjustment control lines. One end is fixedly connected to the flexible lens end, and the other end is connected to the handheld operating handle 2 to realize signal transmission and angle control signal transmission.

[0094] Handheld operating handle 2: It adopts an ergonomic design and has control buttons on the surface (including camera start / stop button, fill light brightness adjustment button, angle adjustment button). It integrates data transmission lines internally and has a Type-C interface at the end. It can be detachably connected to the signal processing and transmission module 4 through a connecting cable to realize wired transmission of real-time image data.

[0095] (2) Signal processing and transmission module 4

[0096] This is an independent transmitter box with a lightweight and portable design (size ≤10cm×8cm×3cm). It features a built-in video encoding chip, a dual-band wireless communication module (2.4G / 5G WiFi + Bluetooth 5.3), an anti-interference processing unit, and a rechargeable lithium battery (capacity ≥5000mAh, battery life ≥4 hours).

[0097] Video encoding chip: used to compress the raw real-time video transmitted by the endoscope body 1 into H.265 encoding format, reducing the transmission bandwidth usage while ensuring image quality.

[0098] Dual-band wireless communication module: Supports both 5G WiFi and Bluetooth 5.3 transmission modes. The 5G WiFi mode is suitable for short-range, high-quality transmission needs in operating rooms, while the Bluetooth 5.3 mode is suitable for low-power, low-interference scenarios and can automatically switch according to the transmission environment.

[0099] Anti-interference processing unit: Built-in electromagnetic interference filtering algorithm can filter electromagnetic signal interference generated by medical equipment in the operating room, ensuring that the image transmission delay is ≤50ms, meeting the needs of real-time diagnosis and treatment.

[0100] Power supply components: It adopts a rechargeable lithium battery, supports Type-C interface charging, and can meet more than 4 hours of continuous operation on a single charge, making it suitable for long-term surgical scenarios.

[0101] (3) AI glasses terminal 5

[0102] As the core of display and AI assistance, it integrates a micro-display optical module, an edge AI processor, a wireless receiving module, an ambient light sensor, and bone conduction headphones.

[0103] Micro-display optical module: It adopts holographic waveguide lens, which has high light transmittance (≥85%) and wide viewing angle (≥40°). It can project the decoded real-time image into the doctor's field of vision, with a display resolution of ≥720P, realizing an integrated field of vision for direct viewing of the patient and viewing the image.

[0104] Edge AI processor: computing power ≥8 TOPS, pre-trained with a semantic segmentation model of digestive tract structure, capable of pixel-level semantic segmentation of tissues such as esophagus, stomach, and duodenum in real-time images, accurately identifying the target location of tube placement (such as cardia and pylorus) and dangerous areas (such as areas with dense blood vessels and ulcer areas).

[0105] Wireless receiving module: Adapted to the dual-band wireless communication module of signal processing and transmission module 4, it supports 5G WiFi and Bluetooth 5.3 signal reception, and realizes stable decoding of image data.

[0106] Ambient light sensor: Real-time detection of ambient light intensity in the operating room, automatically adjusting the brightness of the holographic waveguide lens (adjustment range 100-1000 cd / m²) to avoid strong or weak light environments affecting image visibility.

[0107] Bone conduction headphones: used to output AI-assisted voice prompts, including "The lens is off the midline of the esophagus, please adjust to the left" and "The area ahead is densely populated with blood vessels, proceed with caution," etc., without blocking the ears and without affecting communication between doctors and medical teams.

[0108] Device linkage interface: Supports communication with external devices such as operating room large screens and medical monitors to achieve real-time image synchronization and sharing, meeting the needs of team collaborative diagnosis and treatment.

[0109] 2. A display method for endoscopes based on AI glasses, applied to the above system, with the following specific steps:

[0110] (1) System assembly and debugging (step S1)

[0111] The disposable flexible lens end is fixedly connected to the handheld operating handle 2 through a flexible conduit to ensure a firm connection without any looseness;

[0112] Connect the end interface of the handheld operating handle 2 to the signal processing and transmission module 4 (transmitter box) using the Type-C cable 3, and check the stability of the line connection;

[0113] The signal processing and transmission module 4 is turned on, and the 5G WiFi transmission mode is started by default. At the same time, the AI ​​glasses terminal 5 is turned on. The pairing request with the signal processing and transmission module 4 is initiated through the touch panel of the glasses. After successful pairing, the AI ​​glasses terminal 5 automatically calibrates the screen resolution (default 720P) and display brightness (default 500cd / m²).

[0114] Adjust the angle adjustment component and the fill light component to ensure that the high-definition camera captures clear images, the angle can be adjusted flexibly, and the brightness of the fill light is adjustable.

[0115] (2) Image capture (step S2)

[0116] The doctor holds the operating handle and inserts the disposable soft lens into the target digestive tract area through the patient's mouth or nose. The doctor activates the high-definition camera and LED fill light through the control buttons on the operating handle. The brightness of the fill light can be adjusted according to the operation needs (adjustable from 1 to 5 levels). The high-definition camera captures real-time images of the inside of the digestive tract at a stable frame rate of 30fps.

[0117] (3) Encoding and wireless transmission (step S3)

[0118] The handheld operating handle 2 transmits the acquired raw real-time image to the signal processing and transmission module 4 via an internal data transmission line. The video encoding chip compresses the raw image into H.265 format. The anti-interference processing unit runs an electromagnetic interference filtering algorithm to filter out electromagnetic signal interference in the operating room. Then, the dual-band wireless communication module sends the encoded image data to the AI ​​glasses terminal 5. If the signal processing and transmission module 4 detects that the 5G WiFi transmission delay is ≥50ms (i.e., the electromagnetic interference exceeds the preset threshold), it automatically switches to the Bluetooth 5.3 module for transmission. At this time, the image resolution is adaptively adjusted to 480P to ensure transmission continuity.

[0119] (4) Image decoding and display (step S4)

[0120] The AI ​​glasses terminal 5's wireless receiving module receives the encoded image data. After decoding by the built-in decoding chip, the image is projected onto the holographic waveguide lens. Doctors can view a clear real-time image of the digestive tract while looking directly at the patient without having to look up or down. The viewing angle is ≥40°, which can cover the doctor's core field of vision.

[0121] (5) AI-assisted recognition and voice prompts (step S5)

[0122] The edge AI processor runs a digestive tract structure semantic segmentation model to perform real-time semantic segmentation on the decoded real-time video:

[0123] Identify tissue regions such as esophageal mucosa, cardia, gastric fundus, and duodenum, and mark the target placement location with a green box;

[0124] Identify dangerous areas such as areas with dense blood vessels and ulcers, and mark them with red boxes;

[0125] The system analyzes the deviation between the advancement path of the soft lens and the target position in real time. When the deviation exceeds 3mm, the bone conduction headphones automatically issue a voice prompt to guide the doctor to adjust the operation direction.

[0126] (6) Brightness adaptive adjustment and operation end (step S6)

[0127] An ambient light sensor detects the ambient light intensity in the operating room in real time. When the detected light intensity is ≥800 lux, the brightness of the holographic waveguide lens is automatically reduced to 100-300 cd / m². When the detected light intensity is ≤200 lux, the brightness of the image is automatically increased to 700-1000 cd / m² to ensure that the image is always clearly visible. After the catheter placement or examination is completed, the doctor turns off the signal processing and transmission module 4 and the AI ​​glasses terminal 5, and disposes of the disposable soft lens end and flexible catheter as medical waste. The handheld operating handle 2 is disinfected and kept for later use. The signal processing and transmission module 4 is rechargeable and can be used repeatedly.

[0128] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments:

[0129] Example 1: System Hardware Configuration and Assembly

[0130] In this embodiment, the specific parameters of each component of the system are as follows:

[0131] Endoscope body 1: The soft lens tip is made of medical-grade polycarbonate material, 30cm in length and 5mm in diameter; the high-definition camera has a resolution of 1080P and a frame rate of 30fps; the LED fill light consists of 3 surface-mount LEDs with adjustable brightness levels 1-5; the angle adjustment component can achieve ±90° vertical and ±90° horizontal rotation; the handheld operating handle 2 is 15cm in length and weighs 80g, with 4 control buttons on its surface (power button, camera start / stop button, fill light adjustment button, and angle adjustment button).

[0132] Signal processing and transmission module 4: Dimensions 8cm×6cm×2cm, weight 120g; video encoding chip uses HiSilicon Hi3519V500; dual-band wireless communication module uses Qualcomm QCA9377 (supports 5G WiFi 802.11ac and Bluetooth 5.3); anti-interference processing unit uses adaptive frequency hopping algorithm; lithium battery capacity 5000mAh, supports 18W fast charging.

[0133] AI Glasses Terminal 5: Holographic waveguide lens with 88% light transmittance, 720P display resolution, and 45° viewing angle; Edge AI processor uses Horizon Journey 2 (8 TOPS computing power); Ambient light sensor uses BH1750; Bone conduction headphones with 100dB sensitivity and 20-20000Hz frequency response; Device linkage interface supports both HDMI and USB-C protocols.

[0134] System assembly steps:

[0135] Remove the disposable flexible lens end, align its tail connector with one end of the flexible conduit, and tighten it to ensure a waterproof seal.

[0136] Connect the other end of the flexible catheter to the front interface of the handheld operating handle 2, insert it, rotate to lock it, and check whether the angle adjustment component is flexible.

[0137] Insert one end of the Type-C cable 3 into the tail port of the handheld operating handle 2 and the other end into the input port of the signal processing and transmission module 4. A "click" sound indicates that the connection is in place.

[0138] Turn on the power button of the signal processing and transmission module 4. A solid green indicator light indicates normal operation. Turn on the AI ​​glasses terminal 5 by pressing and holding the power button for 3 seconds to power on. Enter the settings interface through the touch panel, select "Device Pairing", search for the device name of the signal processing and transmission module 4, and click Connect. After successful pairing, the indicator light will flash blue.

[0139] Example 2: Clinical application of gastrointestinal tube placement

[0140] This embodiment applies to a gastrointestinal tube placement procedure in a hospital's gastroenterology department. The specific procedure is as follows:

[0141] Preoperative preparation: The patient fasts and abstains from water for 6 hours and is placed in a semi-recumbent position; the doctor wears AI glasses terminal 5 to complete system assembly and debugging, ensuring that the screen display is normal and the AI ​​recognition function is available;

[0142] Lens insertion: The doctor holds the operating handle and slowly inserts the soft lens end through the patient's mouth, then turns on the high-definition camera and LED fill light (adjusted to level 3 brightness) to capture real-time images of the inside of the esophagus;

[0143] Transmission and display: The signal processing and transmission module 4 compresses the captured image into H.265 format and transmits it to the AI ​​glasses terminal 5 via 5G WiFi. After decoding, it is projected onto the holographic waveguide lens, allowing the doctor to clearly see the texture of the esophageal mucosa.

[0144] AI-assisted operation: The edge AI processor identifies the esophageal midline and marks it with a green line. When the soft lens deviates from the midline by 2mm, the bone conduction headphones prompt "Slight deviation from the midline, adjust to the right". The doctor adjusts the angle with the operating handle to bring the lens back to the midline. When it reaches the cardia, the AI ​​automatically marks a green box and prompts "The cardia has been reached, ready to enter the stomach cavity".

[0145] Danger Zone Warning: When the camera approaches the area with dense blood vessels at the fundus of the stomach, the AI ​​marks the area with a red box and provides a voice prompt, "The area ahead is densely covered with blood vessels. Slow down the advance speed." The doctor then reduces the advance speed to avoid damaging the blood vessels.

[0146] Catheter placement completed: After the lens reaches the target position (middle of the stomach body), the doctor fixes the lens angle with the operating handle and completes the catheter placement operation with the guide wire. The operation takes 8 minutes, which is 3 minutes shorter than the traditional method.

[0147] Equipment handling: After the operation is completed, turn off the system power, disconnect the connecting wires, put the disposable flexible lens end and flexible catheter into the medical waste recycling bag, wipe the handheld operating handle 2 with chlorine disinfectant and set it aside, and charge the signal processing and transmission module 4 for the next use.

[0148] Alternative embodiments

[0149] Alternative transmission method: If the electromagnetic interference in the operating room is extremely severe, both 5G WiFi and Bluetooth 5.3 transmission will be affected. The signal processing and transmission module 4 and the AI ​​glasses terminal 5 can be connected via wired fiber optic cable to achieve zero-latency transmission and adapt to high-precision surgical scenarios.

[0150] AI Function Expansion: The AI ​​glasses terminal 5 can integrate a physiological parameter monitoring module, which collects the patient's heart rate and blood oxygen saturation through non-contact sensors. If the parameters are abnormal, it will immediately issue a voice warning, realizing the integration of video monitoring and physiological monitoring.

[0151] Transmitter integration replacement: The signal processing and transmission module 4 can be integrated with the operating room monitor, and the monitor's power supply and display screen can be used to achieve image backup, further simplifying the equipment layout and reducing clinical use costs.

[0152] The technical solutions provided by the embodiments of the present invention have the following significant beneficial effects:

[0153] Integrated operation and vision, significantly improved accuracy: The holographic waveguide lens of AI glasses terminal 5 integrates the real-time image with the doctor's field of vision. Doctors can take care of both the patient and the image without frequently switching their field of vision, completely solving the problem of field of vision fragmentation caused by traditional display devices. The accuracy of catheter placement and examination is improved by ≥30%, and the rate of operational errors is reduced.

[0154] Stable and reliable wireless transmission meets real-time requirements: The dual-frequency wireless communication design of the independent signal processing and transmission module 4, along with anti-interference algorithms, ensures that the image transmission delay is ≤50ms and the resolution can reach up to 1080P. It effectively resists electromagnetic interference in the operating room and solves the defects of high latency and poor stability of traditional Bluetooth and WiFi transmission, meeting the stringent requirements of real-time clinical diagnosis and treatment.

[0155] AI-assisted diagnosis and treatment lowers the operational threshold: The semantic segmentation and recognition and bone conduction voice prompt functions of the edge AI processor enable automatic recognition of digestive tract structures, marking of target locations and danger areas, and path deviation prompts, reducing the reliance on doctors' clinical experience and increasing the success rate of catheter placement by ≥20%, which is especially suitable for novice doctors and primary care hospitals.

[0156] The device features a lightweight design and optimized user experience: the signal processing and transmission module 4 is separated from the handheld operating handle 2, reducing the handle weight by ≥40%, making doctors less prone to fatigue during long-term operation and significantly improving operational flexibility; the disposable soft lens end design completely eliminates the risk of cross-infection and meets medical and health requirements.

[0157] With strong scene adaptability and high compatibility, the system does not require modification of the existing hospital operating room facilities and environment. It can be linked with external devices such as operating room large screens and monitors, and is suitable for various clinical scenarios such as intensive care units, gastroenterology departments, and operating rooms, and has broad application value.

[0158] The technical solutions provided by the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the embodiments of the present invention. The descriptions of the embodiments above are only for helping to understand the principles of the embodiments of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A system based on AI glasses for use with endoscopes, characterized in that, include: The endoscope body is used to insert into the target area and acquire real-time images of that area; The signal processing and transmission module establishes a data connection with the endoscope body to receive the real-time image, perform encoding, compression, and anti-interference processing, and then transmit it wirelessly. The AI ​​glasses terminal wirelessly pairs and communicates with the signal processing and transmission module to receive and decode the encoded real-time images, and presents them visually through the built-in display component. The AI ​​glasses terminal also has a built-in AI processing unit that performs target structure recognition, region marking, and operation assistance prompts on the real-time images, realizing the integration of visualization of endoscopic operations and AI-assisted diagnosis and treatment.

2. The system for endoscopes based on AI glasses according to claim 1, characterized in that, The endoscope body includes a disposable flexible lens tip, a flexible catheter, and a handheld operating handle. The flexible lens tip is fixedly connected to the handheld operating handle through the flexible catheter. The soft lens end has a built-in high-definition camera, a fill light component, and an angle adjustment component. The high-definition camera is used to capture real-time images, the fill light component is used to provide a shooting light source, and the angle adjustment component is used to adjust the shooting angle. The handheld operating handle has control buttons on its surface and a data transmission line inside. The control buttons are electrically connected to the fill light component and the angle adjustment component to control the working status of the components. The data transmission line is connected to the signal processing and transmission module through an end interface.

3. The system for endoscopes based on AI glasses according to claim 2, characterized in that, The soft lens is made of medical-grade disposable polymer material and has a waterproof and anti-fog coating on its surface; the high-definition camera has a resolution of ≥1080P and a frame rate of ≥30fps; the fill light component is an LED fill light, and the angle adjustment component can achieve ±180° deflection adjustment.

4. The system for endoscopes based on AI glasses according to claim 1, characterized in that, The signal processing and transmission module is an independent transmitter box, which integrates a video encoding chip, a dual-frequency wireless communication module, an anti-interference processing unit, and a power supply component. The video encoding chip is used to compress the real-time video into H.265 encoding format; The dual-band wireless communication module includes a 2.4G / 5G WiFi module and a Bluetooth 5.3 module, which can be selected or automatically switched for wireless transmission; The anti-interference processing unit is used to filter electromagnetic interference signals to ensure that the transmission delay is ≤50ms. The power supply component is a rechargeable lithium battery with a capacity of ≥5000mAh and a battery life of ≥4 hours.

5. The system for endoscopy based on AI glasses according to claim 4, characterized in that, The transmitter box has external dimensions of ≤10cm×8cm×3cm and adopts a lightweight and portable design; the transmitter box is detachably connected to the end interface of the handheld operating handle of the endoscope body via a Type-C connection cable.

6. The system for endoscopy based on AI glasses according to claim 1, characterized in that, The AI ​​glasses terminal includes a micro-display optical module, an edge AI processor, a wireless receiving module, an environmental perception component, and an audio output component; The micro-display optical module is a holographic waveguide lens used to project the decoded real-time image, with a display resolution ≥720P and a viewing angle ≥40°. The edge AI processor has a computing power of ≥8 TOPS and is used to run a preset target recognition model to perform semantic segmentation and recognition on real-time images. The environmental sensing component is an ambient light sensor, used to detect the intensity of ambient light. The audio output component is a bone conduction headset, used to output voice prompts.

7. The system for endoscopy based on AI glasses according to claim 6, characterized in that, The preset target recognition model is a digestive tract structure semantic segmentation model, and the tissue regions used for recognition include the esophagus, stomach, and duodenum; the tube placement target location and danger areas are marked; the light intensity signal detected by the ambient light sensor is used to automatically adjust the brightness of the holographic waveguide lens, with an adjustment range of 100-1000 cd / m²; the voice prompt information includes tube placement path deviation and danger area warning.

8. The system for endoscopy based on AI glasses according to claim 1, characterized in that, The AI ​​glasses terminal is also equipped with a device linkage interface, which supports establishing communication connections with external display devices to achieve synchronous sharing of real-time images. The external display devices are operating room screens or medical monitors.

9. A display method for endoscopes based on AI glasses, characterized in that, Applied to the endoscope-based AI glasses system as described in any one of claims 1-8 Includes the following steps: S1: Assemble and debug the system. Connect the handheld operating handle of the endoscope body to the signal processing and transmission module via a connecting cable. Wirelessly pair the AI ​​glasses terminal with the signal processing and transmission module and calibrate the screen resolution and display parameters. S2: Insert the flexible lens end of the endoscope body into the target area, and activate the high-definition camera and lighting component through the control button to capture real-time images of the target area; S3: The handheld operating handle transmits the real-time image to the signal processing and transmission module, which compresses it into H.265 format through the video encoding chip, filters electromagnetic interference through the anti-interference processing unit, and then transmits it through the dual-frequency wireless communication module. S4: The AI ​​glasses terminal receives the compressed image through a wireless receiving module, decodes it, and then visualizes it through a holographic waveguide lens. S5: The edge AI processor runs a preset target recognition model, performs semantic segmentation on the real-time image, identifies the target structure and marks the target location and danger zone, and outputs corresponding voice prompts through bone conduction headphones; S6: The ambient light sensor detects the ambient light intensity in real time and automatically adjusts the brightness of the holographic waveguide lens until the operation is complete.

10. The display method for endoscopes based on AI glasses according to claim 9, characterized in that, In step S3, the dual-band wireless communication module prioritizes the use of the 5G WiFi module for transmission; when a transmission delay of ≥50ms is detected, it is determined that the electromagnetic interference exceeds the preset threshold, and it automatically switches to the Bluetooth 5.3 module for low-power transmission. At this time, the screen resolution is adaptively adjusted to 480P.

Images