A smart gastrointestinal capsule system with UWB positioning function

CN122556895APending Publication Date: 2026-08-14SHENZHEN HUAYUN KERUI TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-09
Publication Date
2026-08-14

AI Technical Summary

Technical Problem

[0003]但是,现有技术在临床应用中仍存在一个核心痛点:无法获取病灶在消化道内的精确三维物理绝对坐标,现有的胶囊内镜多依赖被动射频信号强度(RSSI)进行区域估算,误差通常大于10cm;或依赖纯视觉SLAM算法推算位姿,但在肠道这种柔性、弱纹理环境中极易产生累计漂移误差

Benefits of technology

1、本发明突破性地在胶囊内引入了微型UWB绝对定位技术与体外UWB锚点阵列配合,将病灶的描述从“时间维度(视频第X分X秒)”提升至“三维空间物理维度(体内X,Y,Z坐标)”,定位精度可达5cm以内,不依赖肠道视觉纹理,彻底避免了视觉SLAM算法的漂移,直接为后续手术提供了高价值的导航地图。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122556895A_ABST
    Figure CN122556895A_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent gastrointestinal endoscopy capsule system with UWB positioning function, belonging to the field of medical device design and manufacturing technology, including: an in vivo capsule end, an external positioning and receiving end, and a back-end analysis end. The in vivo capsule end includes an image acquisition module, a UWB transmission module, and a main control and synchronization module. The main control and synchronization module uses a high-precision temperature-compensated crystal oscillator as the global clock reference, synchronously triggering image acquisition and UWB pulse transmission within the same clock cycle, and injecting a unified first timestamp into both, transmitting the timestamped image and UWB pulse to the outside via wireless channels. The external positioning and receiving end receives pulses through a UWB anchor point array and uses a time difference of arrival algorithm to calculate the capsule's three-dimensional coordinates. The back-end analysis end uses the first timestamp as the primary key to spatiotemporally align the coordinates with the image, and after identifying lesions using a large visual model, generates a visual navigation report with three-dimensional coordinate annotations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of medical device design and manufacturing technology, and more specifically to an intelligent gastrointestinal capsule system with UWB positioning function. Background Technology

[0002] Currently, capsule endoscopy, as a painless and non-invasive digestive tract examination tool, is widely used in clinical practice. Existing capsule endoscopy systems typically have image acquisition, wireless data transmission, and external display functions. Some advanced models also integrate pressure sensors and vibration sources, allowing for external control of capsule vibration to change position and acquire more comprehensive images and pressure data.

[0003] However, existing technologies still have a core pain point in clinical applications: they cannot obtain the precise three-dimensional physical absolute coordinates of lesions in the digestive tract. Existing capsule endoscopes mostly rely on passive radiofrequency signal intensity (RSSI) for region estimation, and the error is usually greater than 10cm; or they rely on pure visual SLAM algorithms to calculate pose, but in the flexible and weak texture environment of the intestine, cumulative drift error is very easy to occur.

[0004] Existing systems, when acquiring multidimensional data (such as video streams and external positioning data), suffer from uncontrollable wireless transmission delays and jitter due to the lack of a unified clock source at the underlying hardware level. This results in the location information received by the external system and the video frames not being precisely aligned on the timeline. Consequently, even after doctors have discovered lesions (such as polyps) in videos lasting several hours, it remains difficult to quickly and accurately locate the physical position of the lesions during subsequent interventional treatments.

[0005] Therefore, how to obtain the precise three-dimensional physical absolute coordinates of the lesion in the digestive tract is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of the above problems, the present invention is proposed to provide an intelligent gastrointestinal capsule system with UWB positioning function that overcomes or at least partially solves the above problems.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, embodiments of the present invention provide an intelligent gastrointestinal endoscopy capsule system with UWB positioning function, comprising: an in-vivo capsule end, an external positioning and receiving end, and a back-end analysis end; The in-vivo capsule is used to acquire digital video streams and send UWB pulse signals within the digestive tract; The external positioning and receiving end is used to receive UWB pulse signals and calculate position coordinates, while also receiving digital video streams; The backend analysis module is used to fuse and analyze location coordinates with digital video streams.

[0009] Furthermore, the in vivo capsule includes an image acquisition module, a UWB transmission module, a main control and synchronization module, and a first wireless communication module; The external positioning and receiving end includes a UWB positioning anchor array, a positioning computing unit, and a second wireless communication module; The backend analysis module includes a spatiotemporal alignment module, a visual intelligence analysis module, and a visualization generation module.

[0010] Furthermore, the image acquisition module is used to acquire digital video streams; The UWB transmitting module is used to periodically transmit UWB pulse signals carrying time information. The main control and synchronization modules are connected to the image acquisition module and the UWB transmission module, respectively, to synchronously trigger the image acquisition module to acquire digital video streams and the UWB transmission module to transmit UWB pulses within the same clock cycle, generating a unified first timestamp and associating it with the digital video stream and the UWB pulses. The first wireless communication module is connected to the main control and synchronization module and is used to send image data carrying the first timestamp to the outside. The UWB positioning anchor array is used to receive UWB pulses with a first timestamp; The positioning calculation unit is connected to the UWB positioning anchor array and is used to calculate the three-dimensional spatial coordinates of the capsule based on the UWB pulse with the first timestamp. The second wireless communication module is used to receive a digital video stream carrying a first timestamp; The spatiotemporal alignment module is connected to the positioning calculation unit and the second wireless communication module respectively. Based on the first timestamp, it spatiotemporally aligns and binds the three-dimensional spatial coordinates of the capsule with the corresponding digital video stream. The visual intelligence analysis module is used to receive spatiotemporally aligned data sequences, identify lesions, and extract their corresponding three-dimensional coordinates; The visualization generation module is used to generate a visual navigation report based on the three-dimensional spatial coordinates of the lesion.

[0011] Furthermore, the main control and synchronization module includes a low-power microcontroller and a high-precision temperature-compensated crystal oscillator; the low-power microcontroller uses the high-precision temperature-compensated crystal oscillator as the global reference clock.

[0012] Furthermore, the UWB transmitting module includes a UWB radio frequency transceiver chip and a conformal conformal antenna. The conformal conformal antenna is made of a flexible printed circuit board and is laid close to the inner wall of the capsule shell.

[0013] Furthermore, it also includes a power management module, which employs a miniature silver oxide battery and a DC-DC step-down circuit.

[0014] Furthermore, the positioning calculation unit uses the time difference of arrival algorithm to calculate the capsule's three-dimensional spatial coordinates based on the UWB pulse with the first timestamp.

[0015] Furthermore, a metal shield is provided between the radio frequency front end of the UWB transmitting module and the first wireless communication module.

[0016] Secondly, embodiments of the present invention provide a remote sensing target detection system based on Chebyshev polynomial guided graph modeling, characterized in that it includes: The capsule end inside the body collects digital video streams and sends UWB pulse signals within the digestive tract; The external positioning and receiving end receives UWB pulse signals and calculates position coordinates, while simultaneously receiving digital video streams; The backend analysis unit fuses and analyzes the location coordinates with the digital video stream; Furthermore, the in vivo capsule includes an image acquisition module, a UWB transmission module, a main control and synchronization module, and a first wireless communication module; The external positioning and receiving end includes a UWB positioning anchor array, a positioning calculation unit, and a second wireless communication module. The backend analysis module includes a spatiotemporal alignment module, a visual intelligence analysis module, and a visualization generation module.

[0017] The main control and synchronization modules synchronously trigger the image acquisition module to acquire digital video streams and the UWB transmission module to transmit UWB pulses within the same clock cycle, and generate a unified first timestamp; The first timestamp is written into the digital video stream and the UWB pulse respectively. The digital video stream carrying the first timestamp is sent to the outside through the first wireless communication module, and the UWB pulse carrying the first timestamp is sent to the outside through the UWB transmission module. The UWB positioning anchor array receives UWB pulses, and the positioning calculation unit calculates the capsule's three-dimensional spatial coordinates based on the UWB pulses of the first timestamp. The second wireless communication module receives a digital video stream carrying a first timestamp; The spatiotemporal alignment module performs spatiotemporal alignment of the capsule's three-dimensional coordinates with the corresponding digital video stream based on the first timestamp; The visual intelligence analysis module receives the spatiotemporally aligned data sequence, identifies lesions, and extracts their corresponding three-dimensional coordinates. The visualization generation module generates a visualization navigation report based on the three-dimensional coordinates of the lesion.

[0018] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following: 1. This invention innovatively introduces micro UWB absolute positioning technology within the capsule in conjunction with an external UWB anchor array, elevating the description of lesions from the "time dimension (video at minute X second)" to the "three-dimensional spatial physical dimension (in vivo X, Y, Z coordinates)". The positioning accuracy can reach within 5cm, without relying on intestinal visual texture, completely avoiding the drift of visual SLAM algorithms, and directly providing a high-value navigation map for subsequent surgery.

[0019] 2. This system pioneers a mechanism that utilizes a homogeneous MCU within the capsule to generate a unified timestamp. Within the same cycle of sending the exposure trigger signal to the image sensor, a pulse transmission command is sent to the UWB module, and a unified, native first timestamp is added. Compared to the traditional approach where external devices independently timestamp the received image and location, this system avoids frame-level misalignment caused by differences in transmission rates and external interference between wireless image transmission signals and RF positioning signals at the underlying hardware logic level, achieving high-precision data fusion at the millisecond level.

[0020] 3. The high-frequency UWB module, image transmission module and image acquisition module were successfully made to coexist stably in an extremely small capsule volume, reducing the noise interference caused by high-frequency pulses to the image transmission signal. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 This is a diagram of the overall system structure provided in the embodiments of the present invention; Figure 2 This is a schematic diagram of the data flow provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] This invention discloses an intelligent gastrointestinal capsule system with UWB positioning function, such as... Figure 1 As shown, it includes: an in vivo capsule end, an in vitro positioning and receiving end, and a back-end analysis end; The in vivo capsule includes an image acquisition module, a UWB transmission module, a main control and synchronization module, and a first wireless communication module; The external positioning and receiving end includes a UWB positioning anchor array, a positioning computing unit, and a second wireless communication module; The backend analysis module includes a spatiotemporal alignment module, a visual intelligence analysis module, and a visualization generation module.

[0025] The specific implementation of this invention is as follows: Inside the capsule, which measures only 11mm × 26mm, the physical layout is divided into front and rear sections. The front section is a transparent optical dome, housing the LED light source and CMOS sensor; the middle section houses the MCU motherboard and the first wireless communication module; and the rear section contains the battery. The UWB antenna abandons traditional ceramic antennas, employing a flexible printed circuit board (FPC) fabricated UWB broadband antenna, laid out in a cylindrical shape against the inner wall of the capsule's middle section, ensuring omnidirectional radiation penetration of the human body. A low-pass filter network is added to the power module output, and a metal shield isolates the UWB RF front-end from the 2.4GHz image transmission chip.

[0026] Step 1 (Same-source trigger): The MCU's internal timer uses the TCXO clock as a reference. A hardware interrupt is generated when the timer overflows (e.g., 10Hz).

[0027] Step 2 (Data Packaging): In the interrupt routine, the MCU sends an "image capture" command to the CMOS sensor via the I2C bus, and within the same microsecond, pulls the UWB chip pin high to trigger pulse transmission. The MCU reads the counter value as the native timestamp T. Stamp001 .

[0028] Step 3 (Information Injection): The MCU will inject T Stamp001 The EXIF ​​header of the written frame image file is transmitted through the first wireless module; the UWB pulse also serves as a physical reference to penetrate the human body.

[0029] Step 4 (In Vitro Fusion): The in vitro computing unit calculates the coordinates (X, Y, Z), with the corresponding timestamp being T. Stamp001 The backend workstation does not need to consider wireless transmission channel latency and can directly use the primary key. TStamp001 Perform a JOIN operation on the data table in the database to achieve absolute hard synchronization between the image frame and the spatial coordinates.

[0030] To achieve high-precision spatiotemporal alignment and intelligent assisted diagnosis, the system of this invention, during operation, sequentially undergoes four stages in data flow: source-end time injection, heterogeneous dual-channel transmission, external computation, and backend large model fusion. Figure 2 The specific data flow and processing procedure are as follows: 1. Source-end synchronous triggering and timestamp injection stage (data acquisition) The main control and synchronization module m3 serves as the scheduling core of the system data flow, utilizing a high-precision TCXO to generate a global clock cycle. When the trigger cycle arrives, m3 sends a frame exposure trigger signal to the image acquisition module m1, and simultaneously sends an RF pulse transmission command to the UWB transmission module m2.

[0031] Within this microsecond-level synchronization period, m3 generates a globally unique, incrementing first timestamp T1.

[0032] m1 transmits the acquired image data back to m3, which then writes the timestamp T1 into the EXIF ​​header (or custom frame header protocol) of the image file. Simultaneously, the timestamp T1 is encapsulated into a micro data packet, ready to be transmitted along with the UWB pulse from m2.

[0033] 2. Heterogeneous dual-channel asynchronous transmission stage (data transmission) Video frame data with timestamp T1 is continuously transmitted as a data stream via the 2.4GHz radio frequency channel through the first wireless communication module m4 as a high-bandwidth load.

[0034] A nanosecond-level electromagnetic pulse with a timestamp T1 serves as an extremely low-latency physical beacon, penetrating human tissue through the UWB transmitting module m2 and radiating omnidirectionally into outer space.

[0035] 3. Parallel in vitro computation and reception stage (location analysis) After the multi-node UWB positioning anchor array nx captures the UWB pulse that penetrates the human body, it sends it to the positioning calculation unit n1. n1 extracts the timestamp T1 attached to the pulse and uses the TDoA algorithm to calculate the absolute three-dimensional coordinate data of the capsule in the body at that time point [P(X,Y,Z),T1].

[0036] The second wireless communication module n2 continuously receives video stream data and performs integrity checks to reconstruct the image sequence with the corresponding frame headers. Data At this point, although there is a microsecond to millisecond-level asynchronous delay in physical transmission between the coordinate stream and the video stream, they both logically carry the same T1 primary key.

[0037] 4. Backend hard binding and large-scale intelligent analysis stage (multimodal fusion) The spatiotemporal alignment module s1 receives the coordinate stream from n1 and the video stream from n2, and uses database technology to perform a JOIN operation with timestamp T1 as the unique primary key. This eliminates time jitter caused by the transmission channel, generating a fully spatiotemporally aligned multimodal data set [Frame]. Data ,P(X,Y,Z)。

[0038] The aligned data sets are then fed into the visual large model analysis module s2. s2 first uses embedded visual object detection algorithms (such as YOLO series networks) to analyze the frames. Data Perform rapid reasoning to extract pixel-level location regions (bounding boxes) and classification labels for lesions (such as polyps, bleeding points, and ulcers).

[0039] Subsequently, s2 inputs the original image, lesion region coordinates, and label data as a joint input prompt to a locally deployed visual multimodal large model (such as a visual language large model fine-tuned using LoRA). Based on the multimodal input, the large model automatically outputs structured medical descriptive conclusions.

[0040] Finally, the visualization and reporting module s3 associates and maps the structured conclusions with the corresponding three-dimensional coordinates P(X,Y,Z), highlights the three-dimensional spatial location of the lesion on the standard intestinal three-dimensional reconstruction model, and generates an intuitive and graphic navigation diagnostic report.

[0041] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0042] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A smart gastrointestinal capsule system with UWB positioning function, characterized in that, include: In vivo capsule end, in vitro positioning and receiving end, and back-end analysis end; The in vivo capsule end is used to acquire digital video streams and send UWB pulse signals within the digestive tract; The external positioning and receiving end is used to receive the UWB pulse signal and calculate the position coordinates, and at the same time receive the digital video stream; The backend analysis terminal is used to fuse and analyze the location coordinates with the digital video stream.

2. The system as described in claim 1, characterized in that, include: The in vivo capsule includes an image acquisition module, a UWB transmission module, a main control and synchronization module, and a first wireless communication module; The external positioning and receiving terminal includes a UWB positioning anchor array, a positioning calculation unit, and a second wireless communication module. The backend analysis module includes a spatiotemporal alignment module, a visual intelligence analysis module, and a visualization generation module.

3. The system as described in claim 2, characterized in that, include: The image acquisition module is used to acquire digital video streams; The UWB transmitting module is used to periodically transmit UWB pulse signals carrying time information; The main control and synchronization module is connected to the image acquisition module and the UWB transmission module respectively, and is used to synchronously trigger the image acquisition module to acquire the digital video stream and the UWB transmission module to transmit the UWB pulse within the same clock cycle, generate a unified first timestamp and associate it with the digital video stream and the UWB pulse; The first wireless communication module is connected to the main control and synchronization module and is used to send a digital video stream carrying a first time stamp and a UWB pulse with a first time stamp to the outside of the body; The UWB positioning anchor array is used to receive UWB pulses with a first timestamp; The positioning calculation unit is connected to the UWB positioning anchor array and is used to calculate the three-dimensional spatial coordinates of the capsule based on the UWB pulse with the first timestamp. The second wireless communication module is used to receive the digital video stream carrying the first timestamp; The spatiotemporal alignment module is connected to the positioning calculation unit and the second wireless communication module respectively, and performs spatiotemporal alignment and binding of the capsule's three-dimensional spatial coordinates with the corresponding digital video stream based on the first timestamp; The visual intelligence analysis module is used to receive the spatiotemporally aligned data sequence, identify lesions, and extract their corresponding three-dimensional coordinates. The visualization generation module is used to generate a visualization navigation report based on the three-dimensional spatial coordinates of the lesion.

4. The system as described in claim 2, characterized in that, The main control and synchronization module includes a low-power microcontroller and a high-precision temperature-compensated crystal oscillator; the low-power microcontroller uses the high-precision temperature-compensated crystal oscillator as the global reference clock.

5. The system as described in claim 2, characterized in that, The UWB transmitting module includes a UWB radio frequency transceiver chip and a conformal conformal antenna. The conformal conformal antenna is made of a flexible printed circuit board and is laid close to the inner wall of the capsule shell.

6. The system as described in claim 2, characterized in that, It also includes a power management module, which uses a miniature silver oxide battery and a DC-DC step-down circuit.

7. The system as described in claim 3, characterized in that, include: The positioning calculation unit uses the time difference of arrival algorithm to calculate the three-dimensional spatial coordinates of the capsule based on the UWB pulse with the first timestamp.

8. The system as described in claim 2, characterized in that, It also includes a metal shielding cover between the radio frequency front end of the UWB transmitting module and the first wireless communication module.

9. A positioning method for an intelligent gastrointestinal capsule system with UWB positioning function, used to implement the intelligent gastrointestinal capsule system with UWB positioning function as described in any one of claims 1-8, characterized in that, include: The capsule end inside the body collects digital video streams and sends UWB pulse signals within the digestive tract; The external positioning and receiving end receives the UWB pulse signal and calculates the position coordinates, while simultaneously receiving the digital video stream; The backend analysis unit fuses and analyzes the location coordinates with the digital video stream.

10. The method as described in claim 9, characterized in that, include: The in vivo capsule includes an image acquisition module, a UWB transmission module, a main control and synchronization module, and a first wireless communication module; The external positioning and receiving terminal includes a UWB positioning anchor array, a positioning calculation unit, and a second wireless communication module. The backend analysis module includes a spatiotemporal alignment module, a visual intelligence analysis module, and a visualization generation module. The main control and synchronization modules synchronously trigger the image acquisition module to acquire digital video streams and the UWB transmission module to transmit UWB pulses within the same clock cycle, and generate a unified first timestamp. The first timestamp is written into the digital video stream and the UWB pulse respectively. The digital video stream carrying the first timestamp is sent to the outside through the first wireless communication module, and the UWB pulse carrying the first timestamp is sent to the outside through the UWB transmission module. The UWB positioning anchor array receives the UWB pulses, and the positioning calculation unit calculates the three-dimensional spatial coordinates of the capsule based on the UWB pulses of the first timestamp. The second wireless communication module receives the digital video stream carrying the first timestamp; The spatiotemporal alignment module performs spatiotemporal alignment of the capsule's three-dimensional coordinates with the corresponding digital video stream based on the first timestamp; The visual intelligence analysis module receives the spatiotemporally aligned data sequence, identifies the lesion, and extracts its corresponding three-dimensional coordinates. The visualization generation module generates a visualization navigation report based on the three-dimensional coordinates of the lesion.