Digestive endoscopy and autonomous navigation system thereof

Abstract

The invention provides a digestive endoscope and an autonomous navigation system thereof. The digestive endoscope comprises a head end which comprises a multispectral camera, a miniature laser radar and a flexible force sensor and is used for collecting sensing data; the bionic snakelike driving structure is used for performing active segmented bending so as to enable the digestive endoscope to move in the digestive tract; the bionic snakelike driving structure has a first advancing mode and a second advancing mode, and the digestive endoscope meets the requirement that the bionic snakelike driving structure moves in the first advancing mode and / or the second advancing mode at different sections where active section bending is conducted on the bionic snakelike driving structure. The digestive endoscopy provided by the invention can realize comprehensive perception of the digestive tract environment, and improves the examination efficiency and safety.

Description

Technical Field

[0001] The present invention relates to the field of machine control, and in particular to a digestive endoscope and an autonomous navigation system thereof. Background Art

[0002] With the advancement of medical technology, endoscopy has become an important tool for the diagnosis and treatment of digestive system diseases. However, the operation of traditional endoscopes (such as colonoscopy) is highly dependent on the doctor's skills and experience. For novice doctors, there is a long learning curve and it can easily cause patients to feel pain during the operation or misdiagnosis. Especially when dealing with complex cases, such as patients with long intestines, postoperative adhesions, or anatomical variations, it is difficult for the endoscope to reach the target area (such as the ileocecal region), resulting in an examination failure rate as high as 10% to 20%. In addition, the pursuit of efficiency also faces safety issues: shortening the examination time may increase the risk of perforation (with an incidence of 0.1%-0.3%), while cautious operation will prolong diagnosis and treatment time.

[0003] Currently, most technologies in the industry focus on AI-assisted diagnosis, such as polyp detection, but navigation still requires manual operation. While there have been attempts to develop semi-automated solutions, magnetically controlled capsule endoscopes, for example, can only perform passive imaging and are unable to provide active treatment. Some robotic endoscopes require full remote control by a physician, failing to achieve true autonomous navigation. Furthermore, major bottlenecks faced by existing technologies include insufficient real-time 3D reconstruction accuracy, low flexible drive efficiency, and a lack of clinical validation of safety control algorithms. Summary of the Invention

[0004] An embodiment of the present invention provides a digestive endoscope and an autonomous navigation system thereof, which can solve at least one of the technical problems mentioned above.

[0005] According to a specific embodiment of the present invention, in a first aspect, the present invention provides a digestive endoscope, comprising: The head end includes a multispectral camera, a micro laser radar and a flexible force sensor for collecting sensory data; the bionic serpentine drive structure is used to perform active segmented bending to enable the digestive endoscope to displace in the digestive tract; the bionic serpentine drive structure has a first forward mode and a second forward mode, and the digestive endoscope satisfies the requirement of displacing in the first forward mode and / or the second forward mode at different segments where the bionic serpentine drive structure performs active segmented bending; wherein the first forward mode includes a first propulsion action in which the front end is fixed and the middle end is bent to move the rear end forward, and a second propulsion action in which the rear end is fixed and the middle end is extended to move the front end forward, and the first forward mode satisfies the requirement of controlling the bionic serpentine drive structure to alternate between the first propulsion action and the second propulsion action; the second forward mode satisfies the requirement of controlling the bionic serpentine drive structure to propel forward in a spiral manner.

[0006] According to a specific embodiment of the present invention, in a second aspect, the present invention provides an autonomous navigation system for a digestive endoscope, which is used to provide decision control for the digestive endoscope provided in the first aspect, including: an environmental perception module, which is used to obtain the perception data collected by the head end, and construct a three-dimensional model of the digestive tract in real time based on the perception data; an autonomous decision-making algorithm module, which is configured with a reinforcement learning model and a target database, and is used to control the reinforcement learning model to perform operation simulation on the three-dimensional model of the digestive tract according to the target database to obtain an optimal planned path; wherein, the optimal planned path is the path with the shortest endoscopic operation path and the lowest operation difficulty score; the autonomous decision-making algorithm module is also used to segment the optimal planned path based on the adaptation scenarios supported by each of the first forward mode and the second forward mode, and determine the optimal forward mode adapted to each segment in the optimal planned path, and control the bionic serpentine drive structure to displace each segment in the optimal planned path according to the optimal forward mode.

[0007] In some embodiments, the environmental perception module constructs a three-dimensional model of the digestive tract in the following manner: based on the visual data collected by the multispectral camera with light of different bands, and based on the differences in visual data between the infrared band, the ultraviolet band and the visible light band, the mucosal layer is screened and the lesion area is identified to obtain target visual data; based on the micro laser radar, the distance between the head end and the digestive tract wall is measured, and point cloud data is generated; based on the flexible force sensor, the contact force data between the digestive endoscope and the digestive tract wall is monitored; and based on the target visual data, the point cloud data, and the contact force data, a three-dimensional model of the digestive tract is constructed.

[0008] In some embodiments, the environmental perception module constructs a three-dimensional model of the digestive tract based on the target visual data, the point cloud data, and the contact force data in the following manner: based on a data fusion algorithm, the target visual data is aligned with the point cloud data to obtain an initial three-dimensional structure of the digestive tract; for the structural part of the bionic serpentine drive structure that has entered the digestive tract, the structural part is subjected to force analysis based on the contact force data to determine a contact point surface information set between the bionic serpentine drive structure and the digestive tract, wherein the contact point surface information set includes at least two possible contact situations between the bionic serpentine drive structure and the digestive tract; based on the initial three-dimensional structure, the contact point surface information is screened in the contact point surface information set to obtain target contact point surface information, and the target contact point surface information has the highest matching degree with the initial three-dimensional structure; based on the target contact point surface information and the contact force data, the initial three-dimensional structure is structurally adjusted and data supplemented to obtain the three-dimensional model of the digestive tract.

[0009] In some embodiments, the autonomous decision-making algorithm module determines the optimal forward mode in the following manner: the default configuration of the optimal forward mode is the second forward mode; if the minimum diameter of the digestive tract space of the target segment is less than a preset diameter threshold, the optimal forward mode of the target segment is modified to the first forward mode.

[0010] In some embodiments, the autonomous decision-making algorithm module determines the optimal forward mode in the following manner: the default configuration of the optimal forward mode is the second forward mode; if the spatial curvature of the digestive tract of the target segment is greater than a preset curvature threshold, the optimal forward mode of the target segment is modified to the first forward mode.

[0011] In some embodiments, the system also includes a safety closed-loop control module; the safety closed-loop control module is used to: obtain and monitor the contact force data between the endoscope and the digestive tract wall monitored by the flexible force sensor; in response to the force value represented by the contact force data being greater than or equal to a preset force threshold, control the bionic serpentine drive structure to switch to a forward mode at the latest segment of travel, and / or adjust the movement amplitude to maintain the forward mode.

[0012] In some embodiments, controlling the bionic serpentine drive structure to switch the forward mode at the latest segment of travel includes: in response to the forward mode adapted at the latest segment being the second forward mode, controlling the bionic serpentine drive structure to switch the forward mode at the latest segment to the first forward mode.

[0013] In some embodiments, the adjustment of the motion amplitude to maintain the forward mode includes: in response to the forward mode adapted at the latest segment being the first forward mode, reducing the distance between the front end and the rear end; in response to the forward mode adapted at the latest segment being the second forward mode, reducing the spiral diameter of the second forward mode.

[0014] In some embodiments, the safety closed-loop control module is used to: predict a pressure threshold based on the contact force data and the three-dimensional model of the digestive tract, and adjust the propulsion force of the bionic serpentine drive structure in real time based on the prediction result of the pressure threshold prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0016] Figure 1 A structural block diagram of a digestive endoscope is shown; Figure 2 The figure shows a structural block diagram of a digestive endoscope autonomous navigation system. DETAILED DESCRIPTION

[0017] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] The terms used in the embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The singular forms "a," "an," "the," and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, and unless the context clearly indicates otherwise, "a plurality" generally includes at least two.

[0019] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0020] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0021] It should also be noted that the terms "include," "comprises," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a product or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such product or device. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the product or device comprising the element.

[0022] It should be noted in particular that any symbols and / or numbers in the specification that are not marked in the accompanying drawings are not drawing marks.

[0023] The optional embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0024] The embodiment provided in this application is an embodiment of a digestive endoscope.

[0025] The following combination Figure 1 The embodiments of the present application are described in detail.

[0026] Figure 1 A structural diagram of a digestive endoscope 1 is shown. Figure 1 As shown, the digestive endoscope 11 includes a head end 11 and a bionic snake-shaped driving structure 12.

[0027] The head end 11 is used as the front end of the digestive endoscope 1 moving in the digestive tract to perform functions such as data collection and machine operation.

[0028] In some embodiments, the head end 11 includes a multispectral camera, a micro laser radar, and a flexible force sensor for collecting perception data.

[0029] In other embodiments, the head end 11 also includes operating components such as a robotic arm, a mechanical knife, etc. that can perform simple machine control and operation.

[0030] On this basis, the head end 11 provides the operator and the system processor with sensory data in the digestive tract on the one hand, and on the other hand can respond to control and execute machine actions based on the operating components to complete surgical actions such as preoperative marking, cutting, and clamping.

[0031] In an embodiment of the present invention, the bionic serpentine drive structure 12 is used to perform active segmented bending to enable the digestive endoscope 1 to move in the digestive tract. The bionic serpentine drive structure 12 serves as the "body end" connected to the head end 11, providing movement and support for the digestive endoscope 1. Segmented bending means that the bionic serpentine drive structure 12 can achieve separate control of multiple segments, thereby maintaining or changing posture within the range of different segment structures, and there is no mutual influence between the different segment structures during the process. On this basis, when the digestive endoscope 1 moves in the digestive tract relying on the bionic serpentine drive structure 12, each time the digestive endoscope 1 completes a displacement, the structural posture within the segment remains relatively static, and the digestive endoscope 1 as a whole appears to be able to observe movement only at the nearest travel point. This structure has significant advantages in the movement of the digestive endoscope 1. When the completed action does not compress the digestive tract, the digestive endoscope 1 can perform a new action without considering whether the completed action will cause additional impact on the action to be completed, thereby greatly reducing the difficulty of endoscopic exploration over long travel distances.

[0032] In an embodiment of the present invention, the bionic serpentine drive structure 12 has a first forward mode and a second forward mode, and the digestive endoscope 1 satisfies the requirement of actively segmented bending of the bionic serpentine drive structure 12 and displaces in the first forward mode and / or the second forward mode respectively at different segments.

[0033] The first forward mode includes a first propulsion action in which the front end is fixed and the middle end is bent to move the rear end forward, and a second propulsion action in which the rear end is fixed and the middle end is stretched to move the front end forward. The first forward mode satisfies the control of the bionic serpentine drive structure 12 to alternate between the first propulsion action and the second propulsion action. This mode imitates the peristalsis of the intestine of an organism, that is, the contents are pushed forward by wave-like muscle contraction. In the context of endoscopy, this means that the front end of the endoscope will imitate this wave-like motion and move forward in a gentler manner. This method is particularly suitable for situations that require delicate control and avoidance of damage to surrounding tissues, and therefore tends to be an obstacle avoidance mode. It can help the endoscope safely pass through narrow or curved paths.

[0034] The second forward mode controls the bionic serpentine drive structure 12 to propel forward in a spiral motion. Compared to peristaltic propulsion, spiral rotation is a more direct and rapid propulsion method. In this mode, the front end of the endoscope rotates in a spiral motion, like a drill bit. This method provides greater penetration and propulsion efficiency, making it suitable for situations requiring rapid traversal of longer distances. Therefore, spiral rotation is often considered a highly efficient propulsion method.

[0035] In this invention, peristaltic propulsion is more suitable for obstacle avoidance because it allows for more precise control of the endoscope's movement, minimizing damage to surrounding tissue. Meanwhile, spiral rotation is a highly efficient advancement mode, enabling faster navigation through different parts of the digestive tract while ensuring safety. These two modes can be automatically switched based on the specific conditions of the digestive tract to achieve optimal navigation results.

[0036] In the embodiment of the present invention, the digestive endoscope 1 achieves efficient data collection and precise operation execution, such as preoperative marking, cutting and clamping, by including a head end 11 equipped with a multispectral camera, a micro laser radar and a flexible force sensor, and a bionic snake-like drive structure 12 with active segmented bending capability. Its unique two forward modes of "peristaltic propulsion" and "spiral rotation" are respectively suitable for meticulous obstacle avoidance and rapid traversal of long distances, which not only reduces potential damage to surrounding tissues, but also greatly reduces the difficulty of endoscope exploration in complex digestive tract environments. This flexible and multifunctional design significantly improves the effectiveness and safety of the digestive endoscope 1 in the diagnosis and treatment process.

[0037] Among them, the digestive endoscope 1 integrates a multispectral camera, a micro laser radar and a flexible force sensor through the head end 11, thereby achieving a comprehensive perception of the digestive tract environment. The design of the bionic snake-shaped drive structure 12 enables the digestive endoscope 1 to move flexibly inside the digestive tract and has two different forward modes: the first forward mode is advanced by alternating the movement of bending the rear part with bending the front end and extending the front end with fixing the rear end, while the second forward mode uses a spiral motion to move forward. This design not only improves the navigation ability of the digestive endoscope 1 in complex anatomical structures, but also can automatically switch the forward mode according to the specific situation, thereby improving the efficiency and safety of the inspection. The evidence is that animal experiments have shown that the success rate of autonomously reaching the ileocecal region is as high as 98%, which is 6 percentage points higher than manual operation.

[0038] Based on the same concept, the present invention also provides a digestive endoscope autonomous navigation system 2 for providing decision control for the digestive endoscope 1 provided in the first aspect. If there are any unclear points in the following embodiments, please refer to any of the above embodiments.

[0039] Figure 2 A structural diagram of a digestive endoscope 1 is shown. Figure 2 As shown, the digestive endoscope autonomous navigation system 2 includes an environment perception module 21 and an autonomous decision-making algorithm module 22, and in some embodiments further includes a safety closed-loop control module 23.

[0040] The environmental perception module 21 is used to obtain the perception data collected by the head end 11 and construct a three-dimensional model of the digestive tract in real time based on the perception data.

[0041] Autonomous decision-making algorithm module 22 is configured with a reinforcement learning model and a target database. It controls the reinforcement learning model to simulate operations on the three-dimensional digestive tract model according to the target database to obtain the optimal planned path. The target database contains best practice case data derived from clinical practice data of expert digestive endoscopy. The optimal planned path is the one that provides the shortest endoscopic operation path and the lowest operational difficulty score.

[0042] The autonomous decision-making algorithm module 22 is also used to segment the optimal planned path based on the adaptation scenarios supported by the first forward mode and the second forward mode, and determine the optimal forward mode adapted for each segment in the optimal planned path, and control the bionic serpentine drive structure 12 to move according to the optimal forward mode in each segment in the optimal planned path.

[0043] The bionic serpentine drive structure 12, based on shape memory alloy (SMA), mimics the serpentine motion of biological snakes, enabling active segmented bending. This design significantly improves the device's maneuverability within narrow and tortuous intestinal tracts and supports both peristaltic propulsion and spiral rotation. Depending on the specific anatomical region, the system automatically switches to the most appropriate propulsion mode, ensuring flexible and efficient endoscope operation in complex anatomical locations such as the hepatic flexure of the colon.

[0044] The autonomous decision-making algorithm combines reinforcement learning (RL) with a database of expert experience, enabling the system to dynamically plan the optimal path and adjust its movement pattern when necessary. By simulating different operation scenarios and incorporating clinical data, the algorithm selects the most appropriate approach for each section of the digestive tract, thereby improving the success rate and efficiency of the operation. Furthermore, in complex areas, the system can automatically adjust its movement mode to ensure a smooth examination process.

[0045] In an embodiment of the present invention, the digestive endoscope autonomous navigation system 2 integrates an environmental perception module 21 and an autonomous decision-making algorithm module 22. The former is used to construct a three-dimensional model of the digestive tract in real time, while the latter plans the optimal path based on a reinforcement learning model and a target database. By incorporating best practice case data into the target database, the system can select the most appropriate advance mode for each segment, significantly improving the accuracy and efficiency of the operation. For example, in complex anatomical areas, the system can automatically adjust the advance mode to adapt to narrow or curvy areas, reducing the difficulty of operation and the risk of complications.

[0046] In the embodiment of the present invention, the environment perception module 21 constructs a three-dimensional model of the digestive tract in the following manner.

[0047] For example, the environmental perception module 21 uses visual data collected by a multispectral camera using light of different wavelengths, and based on the differences between the infrared and ultraviolet bands and the visible light band, it screens out the mucosal layer and identifies the lesion area, thereby obtaining target visual data. A micro-lidar is used to measure the distance between the tip 11 and the digestive tract wall and generate point cloud data. A flexible force sensor is used to monitor the contact force data between the digestive endoscope and the digestive tract wall. A three-dimensional model of the digestive tract is constructed based on the target visual data, point cloud data, and contact force data.

[0048] The system provided by this invention effectively solves the problems of mucosal layer screening and lesion area identification by combining data from a multispectral camera, a micro-lidar, and a flexible force sensor, while simultaneously generating a highly accurate three-dimensional model of the digestive tract. By processing visual data differences between light bands, image clarity and lesion identification accuracy are improved. Furthermore, point cloud data and contact force data further enhance the model's accuracy, ensuring the safety and effectiveness of endoscopic procedures.

[0049] In an embodiment of the present invention, the environment perception module 21 constructs a three-dimensional model of the digestive tract based on target visual data, point cloud data, and contact force data in the following manner.

[0050] For example, the environmental perception module 21 aligns the target visual data with the point cloud data based on the data fusion algorithm to obtain the initial three-dimensional structure of the digestive tract. For the structural part of the bionic serpentine drive structure 12 that has entered the digestive tract, the structural part is subjected to force analysis based on the contact force data to determine the contact point surface information set between the bionic serpentine drive structure 12 and the digestive tract. The contact point surface information set contains at least two possible contact situations between the bionic serpentine drive structure 12 and the digestive tract. Based on the initial three-dimensional structure, the contact point surface information is screened in the contact point surface information set to obtain the target contact point surface information. The target contact point surface information has the highest matching degree with the initial three-dimensional structure. Based on the target contact point surface information and the contact force data, the initial three-dimensional structure is structurally adjusted and the data is supplemented to obtain a three-dimensional model of the digestive tract.

[0051] In summary, through the fusion analysis of target visual data, point cloud data, and contact force data, the present system can construct a more accurate three-dimensional model of the digestive tract. In particular, when determining the contact between the bionic serpentine drive structure 12 and the digestive tract wall, the force-directed analysis method helps to screen the optimal contact point surface information, thereby optimizing the model's structural adjustment process. This method not only improves the model's reliability but also provides a solid foundation for subsequent autonomous decision-making.

[0052] In some embodiments, the autonomous decision-making algorithm module 22 determines the optimal advancement mode based on the digestive tract space diameter. For example, the default optimal advancement mode is configured as the second advancement mode. If the minimum diameter of the digestive tract space of the target segment is less than a preset diameter threshold, the optimal advancement mode for the target segment is modified to the first advancement mode.

[0053] In other embodiments, the autonomous decision-making algorithm module 22 determines the optimal advancement mode based on the spatial curvature of the digestive tract. For example, the optimal advancement mode is configured as the second advancement mode by default. If the spatial curvature of the digestive tract of the target segment exceeds a preset curvature threshold, the optimal advancement mode for the target segment is modified to the first advancement mode.

[0054] Of course, in some other embodiments, the determination method may also take into account both the diameter of the digestive tract space and the curvature of the digestive tract space. The specific method can be found in the above embodiments and will not be repeated here.

[0055] In order to ensure the safety of patients, the system of the present invention is also equipped with a safety closed-loop control module 23.

[0056] The safety closed-loop control module 23 is used to obtain and monitor the contact force data between the endoscope and the digestive tract wall monitored by the flexible force sensor. In response to the force value represented by the contact force data being greater than or equal to the preset force threshold, the bionic serpentine drive structure 12 is controlled to switch to the forward mode at the latest segment of the movement, and / or adjust the movement amplitude to maintain the forward mode. The module uses tactile feedback and pressure threshold prediction technology to monitor and adjust the propulsion force in real time to ensure that the risk of tissue damage is less than 0.05%. When an abnormally high contact force is detected, the system will immediately take measures to reduce the pressure, thereby avoiding potential injury risks. This mechanism not only improves the safety of the operation, but also provides doctors with greater confidence.

[0057] In this embodiment of the present invention, the introduction of a safety closed-loop control module 23 significantly enhances system safety. It not only monitors the contact force between the endoscope and the digestive tract wall but also adjusts the advancement mode or reduces the amplitude of movement when necessary, thereby protecting the patient from unnecessary injury. For example, if an abnormally high contact force is detected, the system immediately takes measures to reduce the pressure, ensuring safety during the operation.

[0058] On the one hand, the bionic serpentine drive structure 12 is controlled to switch its forward mode at the latest segment of travel, including controlling the bionic serpentine drive structure 12 to switch the forward mode at the latest segment to the first forward mode in response to the second forward mode being the selected forward mode at the latest segment. When the selected forward mode at the latest segment is the second forward mode, the system can promptly switch it back to the first forward mode. This function is particularly useful in confined areas. This reduces pressure on surrounding tissue, avoids potential perforation risks, and maintains high operational efficiency.

[0059] On the other hand, the bionic serpentine drive structure 12 is controlled to adjust the motion amplitude to maintain the forward mode, including reducing the distance between the front and rear ends in response to the forward mode adapted at the latest segment being the first forward mode. In response to the forward mode adapted at the latest segment being the second forward mode, the spiral diameter of the second forward mode is reduced. In the case of adapting to the first forward mode, reducing the distance between the front and rear ends can more precisely control the movement of the device, making it suitable for more confined spaces. For the second forward mode, the same purpose is achieved by reducing the spiral diameter. These adjustment strategies are all intended to minimize discomfort and potential risks to the patient.

[0060] In addition, the safety closed-loop control module 23 is used to predict the pressure threshold based on the contact force data and the digestive tract three-dimensional model, and to adjust the propulsion force of the bionic serpentine drive structure 12 in real time based on the prediction result of the pressure threshold prediction.

[0061] On this basis, pressure threshold prediction based on contact force data and a three-dimensional model of the digestive tract enables the system to dynamically adjust propulsion force, ensuring both efficiency and safety throughout the entire process. This real-time feedback mechanism not only prevents tissue damage caused by excessive pressure but also ensures smooth passage of the endoscope through various anatomical structures, significantly improving the overall quality of diagnosis and treatment and patient satisfaction.

[0062] The present invention aims to provide an innovative, full-digestive tract intelligent autonomous navigation endoscopy system to address multiple challenges in traditional endoscopic procedures. The system integrates multiple advanced technologies, including an environmental perception module 21, a bionic serpentine drive structure 12, an autonomous decision-making algorithm, and safe closed-loop control. The environmental perception module 21 consists of a multispectral camera, a miniature laser radar, and a flexible force sensor. These components work together to construct a high-precision three-dimensional model of the digestive tract in real time. By fusing visual data from different wavelengths with precise distance measurements, this module not only resolves common problems such as mucus reflections but also significantly improves image clarity and the accuracy of lesion identification.

[0063] In practical applications, this invention has demonstrated excellent performance. Animal experiments showed that in a porcine colon model, the system achieved a 98% success rate in autonomously reaching the ileocecal region, 6 percentage points higher than manual operation. The average examination time was reduced from 13 minutes to 8 minutes, a 40% reduction in time consumption. Furthermore, the AI-assisted diagnostic accuracy rate reached 95%, a significant improvement over the 85% accuracy of traditional endoscopes. This demonstrates the system's significant potential for improving diagnostic efficiency and accuracy.

[0064] Although the above-mentioned technological innovations have brought many advantages, they still face several challenges in practical applications. For example, in order to ensure the biocompatibility and operability of the system, the diameter of the sensor and actuator must be controlled within 3mm, while taking into account their power consumption and reliability. In addition, it is also necessary to overcome the impact of individual anatomical differences on navigation strategies, especially when dealing with special cases such as a lengthy sigmoid colon, to ensure that the system can flexibly respond to various situations. Finally, considering the special nature of autonomous navigation systems, it is necessary to formulate clear standards for defining medical responsibilities and design an effective emergency takeover mechanism to meet the requirements of relevant certifications such as FDA / CE. By solving these technical and practical difficulties, the present invention is expected to become the standard configuration of the next generation of digestive endoscopes1, providing patients with a safer and more efficient diagnosis and treatment experience.

[0065] The whole digestive tract intelligent autonomous navigation endoscopy system of the present invention has broad market prospects and can meet the needs of medical institutions at different levels. In the high-end medical market, the target customers are the endoscopy centers of approximately 50,000 tertiary hospitals in the world. These institutions have extremely high requirements for equipment performance, so the price of a single device is set between 500,000 and 1 million US dollars, aiming to replace traditional colonoscopy and provide more efficient and safe diagnosis and treatment services. In addition, in response to the sinking demand of the primary medical market, by launching a simplified version of the system, while retaining the core navigation function, the price is reduced to 200,000 US dollars to solve the problem of insufficient experience of primary doctors and improve the level of primary medical services. For overseas emerging markets, especially in areas with high incidence of digestive system diseases such as India and Southeast Asia, combining the telemedicine model for promotion will not only help improve local medical diagnostic capabilities, but also adapt to local economic conditions.

[0066] The system's competitive advantages lie primarily in its technological advantages, clinical value, and business model expansion. Clinically, the system significantly shortens examination times and reduces complication rates, directly saving hospitals labor costs and reducing the cost of a single procedure by 30%.

[0067] This invention breaks through the technical bottleneck of existing endoscopic autonomous navigation through the innovative combination of bionic motion control, edge computing artificial intelligence and multimodal perception. It not only meets the actual clinical needs but also has strong commercial potential. It is expected to become the standard configuration of the next generation of digestive endoscopes1.

[0068] Although operations are described in a particular order in the drawings, this should not be understood as requiring that the operations be performed in the particular order shown or in serial order, or that all shown operations be performed to obtain the desired results. In certain circumstances, multitasking and parallel processing may be advantageous.

[0069] The methods, apparatus, devices, and storage media of the present invention can be implemented using standard programming techniques, utilizing rule-based logic or other logic to implement the various method steps. It should also be noted that the terms "apparatus" and "module" as used herein and in the claims are intended to include implementations using one or more lines of software code and / or hardware implementations and / or devices for receiving input.

[0070] Any steps, operations or procedures described herein may be performed or implemented using one or more hardware or software modules, either alone or in combination with other devices. In one embodiment, the software modules are implemented using a computer program product comprising a computer-readable medium containing computer program code, which can be executed by a computer processor to perform any or all of the steps, operations or procedures described.

[0071] The foregoing description of the implementation of the present invention has been given for the purpose of illustration and description. The foregoing description is not intended to be exhaustive nor to limit the invention to the exact form disclosed, and various variations and modifications may exist in accordance with the above teachings, or may be obtained from the practice of the invention. These embodiments are selected and described in order to illustrate the principles of the present invention and its practical application, so that those skilled in the art can utilize the present invention in various embodiments and various modifications suitable for the specific purpose conceived. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method and will not be elaborated here.

[0072] It is further understood that, unless otherwise specified, “connection” includes a direct connection where there are no other components between the two elements, and also includes an indirect connection where there are other elements between the two elements.

[0073] It should be further understood that, although operations are described in a particular order in the accompanying drawings in the embodiments of the present invention, this should not be construed as requiring that the operations be performed in the particular order shown or in a serial order, or that all of the operations shown be performed to obtain the desired results. In certain circumstances, multitasking and parallel processing may be advantageous.

[0074] Other embodiments of the present invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the field of the invention not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0075] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the scope of the appended rights. The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate from the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A digestive endoscope, characterized in that: include: The headend, which includes a multispectral camera, a micro-lidar, and a flexible force sensor, is used to collect perception data; A bionic serpentine drive structure for performing active segmented bending to enable the digestive endoscope to move within the digestive tract; The bionic serpentine drive structure has a first forward mode and a second forward mode, and the digestive endoscope is adapted to move in the first forward mode and / or the second forward mode at different sections where the bionic serpentine drive structure performs active segmented bending. The first forward mode includes a first propulsion action in which the front end is fixed and the middle end is bent to move the rear end forward, and a second propulsion action in which the rear end is fixed and the middle end is extended to move the front end forward. The first forward mode satisfies the requirement of controlling the bionic serpentine drive structure to alternate between the first propulsion action and the second propulsion action. The second forward mode satisfies the requirement of controlling the bionic serpentine drive structure to advance in a spiral manner.

2. A digestive endoscope autonomous navigation system, characterized in that: The digestive endoscope according to claim 1 comprises: An environmental perception module, configured to obtain the perception data collected by the headend and construct a three-dimensional model of the digestive tract in real time based on the perception data; an autonomous decision-making algorithm module, configured with a reinforcement learning model and a target database, for controlling the reinforcement learning model to perform an operation simulation on the three-dimensional digestive tract model according to the target database to obtain an optimal planned path; wherein the optimal planned path is the path with the shortest endoscopic operation path and the lowest operation difficulty score; The autonomous decision-making algorithm module is also used to segment the optimal planned path based on the adaptation scenarios supported by the first forward mode and the second forward mode, determine the optimal forward mode adapted for each segment in the optimal planned path, and control the bionic serpentine drive structure to move according to the optimal forward mode in each segment of the optimal planned path.

3. The system according to claim 2, characterized in that The environmental perception module constructs a three-dimensional model of the digestive tract in the following manner: Based on the visual data collected by the multispectral camera using light of different wavelengths, and based on the difference in visual data between the infrared band, the ultraviolet band, and the visible light band, the mucosal layer is screened out and the lesion area is identified to obtain target visual data; Measuring the distance between the head end and the digestive tract wall based on the micro laser radar and generating point cloud data; Monitoring contact force data between the digestive endoscope and the digestive tract wall based on the flexible force sensor; A three-dimensional model of the digestive tract is constructed based on the target visual data, the point cloud data, and the contact force data.

4. The system according to claim 3, characterized in that The environment perception module constructs a three-dimensional model of the digestive tract based on the target visual data, the point cloud data, and the contact force data in the following manner: Based on a data fusion algorithm, the target visual data is registered with the point cloud data to obtain an initial three-dimensional structure of the digestive tract; performing a force direction analysis on the structural portion of the bionic serpentine drive structure that has entered the digestive tract based on the contact force data to determine a contact point surface information set between the bionic serpentine drive structure and the digestive tract, the contact point surface information set including at least two possible contact situations between the bionic serpentine drive structure and the digestive tract; Based on the initial three-dimensional structure, contact point surface information is screened in the contact point surface information set to obtain target contact point surface information, where the target contact point surface information has the highest matching degree with the initial three-dimensional structure; Based on the target contact point surface information and the contact force data, the initial three-dimensional structure is adjusted and data is supplemented to obtain the three-dimensional model of the digestive tract.

5. The system according to claim 2, wherein: The autonomous decision-making algorithm module determines the optimal forward mode in the following manner: The default configuration optimal forward mode is the second forward mode; If the minimum diameter of the digestive tract space of the target segment is smaller than the preset diameter threshold, the optimal advancing mode of the target segment is modified to the first advancing mode.

6. The system according to claim 2, wherein: The autonomous decision-making algorithm module determines the optimal forward mode in the following manner: The default configuration optimal forward mode is the second forward mode; If the spatial curvature of the digestive tract of the target segment is greater than a preset curvature threshold, the optimal advancing mode of the target segment is modified to the first advancing mode.

7. The system according to claim 2, wherein: The system also includes a safety closed-loop control module; The safety closed-loop control module is used for: Acquiring and monitoring contact force data between the endoscope and the digestive tract wall monitored by the flexible force sensor; In response to the force value represented by the contact force data being greater than or equal to a preset force threshold, the bionic serpentine drive structure is controlled to switch to a forward mode at the latest segment of travel, and / or to adjust the movement amplitude to maintain the forward mode.

8. The system according to claim 7, characterized in that The controlling the bionic serpentine drive structure to switch to a forward mode at the latest segment of travel includes: In response to the forward mode adapted at the latest segment being the second forward mode, the bionic serpentine drive structure is controlled to switch the forward mode at the latest segment to the first forward mode.

9. The system according to claim 7, wherein: The adjustment of the movement amplitude to maintain the forward mode includes: In response to the forward mode adapted at the latest segment being the first forward mode, reducing the distance between the front end and the rear end; In response to the forward mode adapted at the latest segment being the second forward mode, a helical diameter of the second forward mode is reduced.

10. The system according to claim 7, wherein: The safety closed-loop control module is used for: A pressure threshold is predicted based on the contact force data and the digestive tract three-dimensional model, and the propulsion force of the bionic serpentine drive structure is adjusted in real time based on the prediction result of the pressure threshold.

Images