Robotic wireless capsule endoscopy: recent advances and upcoming technologies

Abstract

Wireless capsule endoscopy (WCE) offers a non-invasive evaluation of the digestive system, eliminating the need for sedation and the risks associated with conventional endoscopic procedures. Its significance lies in diagnosing gastrointestinal tissue irregularities, especially in the small intestine. However, existing commercial WCE devices face limitations, such as the absence of autonomous lesion detection and treatment capabilities. Recent advancements in micro-electromechanical fabrication and computational methods have led to extensive research in sophisticated technology integration into commercial capsule endoscopes, intending to supersede wired endoscopes. This Review discusses the future requirements for intelligent capsule robots, providing a comparative evaluation of various methods' merits and disadvantages, and highlighting recent developments in six technologies relevant to WCE. These include near-field wireless power transmission, magnetic field active drive, ultra-wideband/intrabody communication, hybrid localization, AI-based autonomous lesion detection, and magnetic-controlled diagnosis and treatment. Moreover, we explore the feasibility for future "capsule surgeons".

Fig. 1. Timeline of major milestones in the development of WCE robots and their future conceptual design.

a Development of WCE devices and technologies^,,,,–. The timeline illustrates the commercial progress of WCE devices, seminal research findings, and clinical advancements from 2001 to 2023. WCE wireless capsule endoscopy, FDA US Food and Drug Administration, CFDA China Food and Drug Administration. b Concept of “capsule surgeon”. Utilizing intelligent detection algorithms, the capsule robot undertakes inspections for potential lesions. Upon identifying abnormal tissue, the capsule can secure its position and kickstart medicinal procedures with its functional components. These elements are manifold and might involve snares, high-frequency electrotomes, or hemostatic clips. AI artificial intelligence. Figure 1, created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license. Partial image elements designed by brgfx - Freepik.com.

Fig. 2. Traditional technology defects and advanced intelligent technologies of WCE.

a Drawbacks of conventional methods. PLP packaged lithium-ion polymers, SMA shape memory alloy, RF radio frequency. b–g Working principles of advanced intelligent technologies. b Near-field wireless power transmission. The patient lies on an operating table with a transmitting coil and a capsule containing the receiving coil in the body. c Magnetic field active drive. The physician manipulates a robotic arm endowed with an external magnet, through a handle, dictating the motility of an internal capsule containing a magnet. d Intrabody communication. A wearable receiver containing multiple electrodes is looped with two strip gold electrodes on the capsule. e Magnet/video hybrid location. Magnetic positioning depends on an external magneto-resistive sensor array, and video positioning employs tissue features such as shape, color, and texture for identification. f AI algorithm achieves lesion classification. g Magnetic field-controlled capsules perform diagnostic and therapeutic functions, including biopsy or drug release at targeted locations. Figure 2, created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Fig. 3. Magnetic field active drive of WCE.

a An external magnet drives the WCE robot safely through the intestine with RRMA. Left: schematic representation of the platform. Top right: force analysis. Bottom right: comparative experiments of intestinal capsule movement. CRMA continuous rotation magnetic actuation, RRMA reciprocally rotating magnetic actuation. b The added electromagnetic coil realizes the high-precision levitation control of the capsule robot. Left: schematic representation of the platform. Right: an image sequence of the capsule model performing magnetic levitating translational motion at a desired square wave trajectory. MLCEC magnetic levitation control electromagnetic coil, OCEC orientation control electromagnetic coil. c The rectangular coil drive system increases the tolerability of patients in decubitus position and has been tested in pig clinical trials. Left: the electromagnetic coil model. Right: in vivo experimental setup using a live pig. HC Helmholtz coil, MC Maxwell coil, RC rectangular coil. Panels reproduced with permission from: a ref. , IEEE; b ref. , IEEE; c ref. , IEEE.

Fig. 4. AI-based comprehensive pathological analysis model.

The process begins with the WCE device capturing gastrointestinal images, which are then preprocessed and used as input for the AI algorithm. The algorithm’s architecture can be categorized into traditional machine learning and deep learning based on varying feature extraction and classification methods. Subsequent image processing is carried out in conjunction with the specific application scenarios and analysis tasks of WCE. Ultimately, the model outputs relevant metrics to determine its performance. ML machine learning, DL deep learning, SVM support vector machine. Figure 4, created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Fig. 5. Magnetic control capsule robots for diagnosis and treatment.

a Micro-needle capillary biopsy capsule robot based on soft Sarrus linkage concept. Left: three-dimensional model design. Right: actual deformation of the robot under magnetic force. b Biopsy capsule robot with scraper assembly. Top left: conceptual design of WCE robot with biopsy module. Bottom left: the proposed biopsy module and its states during locomotion and biopsy operation procedure. Top right: assembled biopsy module. Bottom right: completed prototype of WCE device for biopsy. c Active multiple-sampling capsule for gut microbiome. Top left: overall design. Bottom left: sampling process of the capsule controlled by an external magnetic field. Right: fabricated components and a fully assembled capsule. UM uniform magnetic field, GM gradient magnetic field, PM precessional magnetic field. d Passive magnetic drug delivery capsule. Top left: design of drug delivery capsule endoscope. Bottom left: structure of proposed drug delivery module. Right: assembly and prototype of the WCE. e Active magnetic drug delivery capsule. Top: exterior and section view of the FSCR. Bottom: surface rolling locomotion diagram of the FSCR and EPM. FSCR ferrofluid soft capsule robot. Panels reproduced with permission from: a ref. , Mary Ann Liebert, Inc.; b ref. , IEEE; c ref. , IEEE; d ref. , Springer Nature; e ref. , IEEE.

Fig. 6. Translational strategies of WCE for clinical integration.

a Correlation between WCE clinical indicators and advanced intelligent technologies. Metric 1–5: from gastrointestinal safety to easy manipulation; Tech 1–6: from near-field wireless power transmission to magnetic-controlled diagnosis and treatment. b–f Guidelines and key experiments. b Gastrointestinal safety. SAR specific absorption rate, ICNIRP Non-Ionizing Radiation Protection, IEEE Electrical and Electronics Engineers. c Ethical and regulatory considerations. d Interference with human tissue. e Data diversity. f Easy manipulation. HRC human-robot collaboration, TO teleoperation, VR virtual reality. Individual images adapted with permission from: b ref. , Wiley; c ref. , Springer Nature; f ref. , Mary Ann Liebert, Inc. Figure 6, created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Fig. 7. Key technological advancements required for future developments in WCE.

a Achieving high integration. A plausible approach would be to fit magnets and coils within the capsule for WPT, active motion, positioning, diagnosis, and treatment under an external magnetic field. b Telemedicine. The incorporation of mobile cloud computing coupled with 5G communication could facilitate the telemedicine function. c Expansion of therapeutic functionality of WCE robots. It is not confined to malignant polyp removal and multi-scale robotic coordination for carrying out treatments in areas that are normally inaccessible. EMR endoscopic mucosal resection, ESD endoscopic submucosal dissection. Figure 7, created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.