Precise delineation of the clinical target volume (CTV) and nodal CTV (CTV$_{{\mathit{nd}}}$) is crucial for effective radiotherapy planning in nasopharyngeal carcinoma (NPC). Manual contouring is labor-intensive and subject to substantial inter-observer variability, particularly in regions with complex anatomy and indistinct boundaries. This study presents RT-SAM, a novel framework that adapts the Medical Segment Anything Model 2 (MedSAM-2) for automated CTV (i.e., primary CTV and CTV$_{nd}$) contouring in NPC computed tomography (CT) images. The framework synergistically integrates a generalist foundation model (MedSAM-2) with a domain-specific specialist network (2D U-Net) through three principal contributions: (1) automated generation of multi-modal prompts—comprising mask, bounding box, and point representations—derived from specialist network predictions to guide the generalist model; (2) a Visual-Prompt Fusion Attention (ViPFA) mechanism that optimizes feature-prompt interactions through bidirectional cross-modal attention; and (3) an Uncertainty-Enhanced Prediction Adjustment (UEPA) mechanism that enhances model robustness via confidence-based refinement and selective domain adaptation. Comprehensive evaluation on a multi-center cohort of 256 clinical NPC cases from Sun Yat-sen University Cancer Center and 212 public NPC cases from the SegRap2025 lymph node CTV dataset using 5- fold cross-validation demonstrates that RT-SAM achieves a mean DICE coefficient of 0.796 $\pm$ 0.033 (mean $\pm$ standard deviation), significantly outperforming current state-of-the-art methods. Clinical validation by eight radiation oncologists demonstrates that RT-SAM contours are clinically indistinguishable from expert delineations in blinded Turing assessments, achieve superior quality ratings in 75% of comparisons with mean scores of 2.73 for RT-SAM versus 2.66 for manual expert contours, and attain clinically acceptable ratings in over 97% of cases. These results demonstrate that RT-SAM is a clinically feasible solution for automated CTV contouring, with strong potential to standardize treatment planning and mitigate inter-observer variability in NPC radiotherapy.

Objective: During minimally invasive surgery (MIS), three-dimensional (3D) endoscopes provide valuable 3D perception of the patient's internal structures. However, due to the requirement of two cameras and a relatively large baseline distance, the imaging front-end of the conventional binocular 3D (CB3D) endoscope usually lacks compactness. We aim to develop a novel compact monocular dual-view 3D (MDV3D) endoscope imaging system. Methods: We develop a novel optical design for the MDV3D endoscope that exploits the dichroic prism's reflection capability to its internal light to realize MDV3D imaging, ensuring the 3D endoscope's imaging front-end with high compactness. Additionally, we propose a 3D reconstruction optimization method (MB-BEDE) to address the challenge of insufficient accuracy of 3D surface information posed by the typically micro baseline distance between the two virtual cameras in the MDV3D endoscope. Through seamless integration of our MDV3D endoscope and MB-BEDE method, we can obtain reliable real-time 3D information. Results: Evaluation experiments demonstrate our system's capability to provide accurate 3D surface information. Notably, compared to the CB3D endoscope imaging system occupying two channels in the robotic single-port laparo-endoscopic surgery (SPLS) platform, our system only requires one channel with a 5.60 mm diameter, presenting the advantage of creating more operating space for surgical instruments during robotic SPLS procedures. Conclusion and Significance: The proposed system and method present a novel solution for developing compact and cost-effective 3D endoscope imaging systems in MIS, particularly in robotic SPLS.

Robot-assisted minimally invasive surgery (RAMIS) is preferred in clinical settings due to its associated benefits, including reduced trauma, diminished pain, and shortened hospital stays. Autonomous robotic surgery (ARS) holds promise for further enhancing patient outcomes and alleviating physician fatigue. With the ongoing advancement of ARS and the proliferation of flexible instruments, the imperative for automatic instrument interchange interfaces has become apparent. This study introduces a cone-based interface designed to fulfill the pressing need for automatic instrument interchange, addressing challenges such as disc misalignment between the instrument's actuation part (AP) and transmission part (TP) during exchanges for traditional interface, along with other occasional factors. The proposed cone-based approach transmits motion and torque through static friction across the contact cone surface, which has some tolerance for non-coaxial installations. The relationship between pressure and transmitted torque on the contact cone surface is derived and experimentally validated. Several experiments are conducted to validate the efficacy and the accuracy of the proposed method. Despite requiring additional equipment for applying pressure, the cone-based interface proves to be an effective solution for automatic instrument interchange, even in scenarios involving disc misalignment.

Steerable catheters offer significant advantages over conventional catheters, including enhanced control, stability, and accessibility, which reduce operational complexity, fluoroscopy time, and radiation exposure, positioning them as a promising advancement for vascular interventional procedures. Herein, a novel steerable catheter is presented, featuring a hydraulically actuated, soft, steerable tip that allows for real‐time visualization in X‐ray imaging. To optimize performance, several silicone materials were evaluated for their mechanical properties, resulting in a soft tip design with a diameter of 2.6 mm. The tip incorporates an internal tool channel and supports a large bending angle of 180°. The tip demonstrates an average response time of 1.141 s (±0.750 s), a maximum output force of 0.145 N (±0.001 N), and a maximum radial expansion of 1.121 (±0.006). A steering kinematic model of the catheter tip is developed to simulate its movement. The catheter tip's real‐time shape and position information are obtained through intelligent segmentation and neighborhood‐based endpoint detection methods, assisting the surgeon during superselective procedures. The catheter's visibility and flexibility are validated in a live porcine model, demonstrating its potential for future use in interventional procedures.

Choledochoscopy, a natural orifice transluminal endoscopic surgery, represents an alternative to cholecystolithotomy in comparison to cholecystectomy and open gallbladder-preserving surgery. To reduce surgeon fatigue and increase surgical autonomy, this paper presents an autonomous local navigation (ALN) system for cholecystolithotomy, comprising a novel spring-based choledochoscope (CDS) with a monocular camera, a dataset-independent dynamic adaptive threshold segmentation (DATS) algorithm, and an ALN strategy. During navigation, the DATS detects bile duct bifurcations from the CDS view, selecting the target bifurcation based on bile duct anatomy with a wall-following mechanism. The CDS control method including bending and rotation control is then proposed based on the relationship between the position and direction of the bile duct center and the CDS characteristics. To ensure precise navigation, a closed-loop ALN strategy utilizing three interchangeable proportional controllers is developed, with task-specific proportionality factors. The efficacy of the DATS and CDS is validated through series experiments. Finally, both rigid and soft phantom experiments demonstrate the potential of the ALN system for cholecystolithotomy. Note to Practitioners-This work addresses the growing need for surgical autonomy, particularly in reducing surgeon fatigue during high-volume procedures like cholecystolithotomy. We developed an ALN system featuring a novel spring-based CDS and an ALN strategy to enable efficient and reliable navigation within bile ducts. The system's key features include a DATS algorithm for bile duct bifurcation detection and a control method for CDS maneuvering, improving the feasibility of autonomous navigation in complex lumen environments. While our approach enhances procedural reliability, limitations remain. The system only relies on monocular camera data, and future improvements could incorporate additional sensing technologies such as contact sensing to further improve navigation safety. Although validated in laboratory experiments, further animal experiments is required before clinical application. Beyond cholecystolithotomy, the ALN system could also benefit other procedures involving tubular anatomy, such as bronchoscopy and ureteroscopy, expanding its application potential.

Bronchoscopy, as an essential minimally invasive diagnostic and therapeutic modality, assumes a pivotal role in the early detection of lung cancer. However, the complex anatomy of the airway and the fixed length of the bronchoscope’s bending segment, along with its external propulsion property, pose challenges, including the risk of bleeding. This paper introduces a 4 mm diameter robot-assisted bronchoscope with a spring-based extensible segment. By manipulating two driven rods, the segment can be lengthened or shortened. The advantages of the extensible segment are discussed in two main aspects through theoretical analysis and experimentation. Firstly, the extensible segment enables the bronchoscope to move in a follow-the-leader motion mode or fixed-angle motion mode, navigating through narrow corners that are inaccessible to fixed-length bronchoscopes. It can also be shortened to increase its stiffness when it reaches the target position, creating a stable surgical platform for procedures like biopsies. In addition, a tailored master device has been developed to control the extensible bronchoscope in an isotropic manner. Phantom experiments confirm the feasibility and effectiveness of the extensible bronchoscope.

A Mechanical Finger exoskeleton is a device designed to enhance or restore the function of a finger. With the advancement of science and technology, especially the development of robotics and biomedical engineering, such devices show great potential in the fields of rehabilitation medicine, industrial applications and human-computer interaction. For example, enhancing or restoring human function through wearable mechanical devices, especially in assisting the disabled, improving rehabilitation efficiency and military applications, shows great potential. In this paper, we will use solid works to conduct 3D modeling and Matlab coding to briefly elaborate the folding structure design and analysis of the mechanical exoskeleton finger joint. This study reveals the relevant models and simulations of the mechanical finger exoskeleton, which can increase the bending and stretching of the finger within the natural motion range and improve the flexibility characteristics. It can be adjusted according to the specific situation of patients, reduce medical costs, in the field of robotics itself, improve its flexibility and adaptability, improve task execution efficiency, convenience, lightweight and so on for most fields are a big advantage

Currently, the application rate of virtual reality technology in the fields of medical education and medical treatment is relatively low. In this study, based on the structure of the human stomach, virtual reality technology, sensing technology, big data technology, and cloud computing technology were integrated. A new form of medical education was established to include the online website platform for data storage and analysis and the offline VR glasses for the physical operation. Through the use of 3d Max technology, Unity3D technology, and C# language, we constructed a three-dimensional model of the human stomach to present the stomach in three dimensions. This enables the users to immerse in it to achieve true human-computer interactive learning. This study updates the concept of medical education, effectively improves the quality and efficiency of medical education, and facilitates the development of surgical programs and reduction in the risk of surgery, as well as provides experimental materials for scientific research. In the future, it can be further developed to include the three-dimensional structures of the circulatory system and other major systems in the human body.

Soft robots have garnered considerable interest in recent years in terms of structure and functionality, while commercially viable pressure supply systems with various pressure supply modes for soft robots remain slow to progress. Hence, in this work, a palm-sized multi-mode pressure supply system based on a peristaltic pump is proposed for the first time, and precise pneumatic pressure control is achieved with a model reference adaptive control (MRAC) method based on the system dynamics. Further performance evaluation demonstrates that our system can deliver precise positive (mean error < 0.48 kPa for 050 kPa) and negative pressures (mean error < 0.85 kPa for -500 kPa) with minimal fluctuations, which outperforms the PID method. Application experiments indicate that our system can accommodate various pressure supply requirements, including controllable pressurization and depressurization, and has the potential for both pneumatic and hydraulic applications. In addition, by incorporating the MRAC method into the system, we have achieved excellent stability and high repeatability in controlling the position of a retractable soft pneumatic actuator, thus demonstrating the system's high accuracy and repeatability in pressure output. The phantom experiment of the injection soft robotic system further validates the system's diverse pressure supply capabilities for complex manipulations of soft robots.

The rapid advancement of micro-nano machining technology has led to a decrease in the dimensions of microdevices and microchips, following the principles of Moore’s law. In addition to conventional semiconductor materials like silicon, emerging nanoscale materials such as nanowires, nanotubes, and two-dimensional materials are being considered as promising alternative constituent materials. The mechanical properties of these materials have a significant impact on the performance and service life of these microdevices and microchips. However, conventional mechanical testing methods have difficulty in accurately measuring the properties of these materials at the nanoscale due to limitations in displacement control and microforce sensing. Consequently, there is an urgent need to develop a micromechanical device capable of testing nanoscale solid materials. In this study, we propose a concept based on high-resolution image sequences for the design of an integrated micromechanical device capable of synchronously measuring the force and deformation of tested specimens. The device has been fabricated using ultrafast femtosecond laser etching technology, which offers an efficient and cost-effective approach for manufacturing microstructures and is suitable for processing various materials such as metals and nonmetals. The stiffness of the device plays a crucial role in the design of the micromechanical device, and a stiffness-matching criterion is introduced to ensure appropriate design parameters. The fabricated device is employed to conduct in-situ tension experiments on SiC nanowires and multilayer molybdenum disulfide nanosheet within a scanning electronic microscope, enabling accurate measurement of their strength, modulus, and fracture strain.

The delineation of the Clinical Target Volume (CTV) is a crucial step in the radiotherapy (RT) planning process for patients with nasopharyngeal carcinoma (NPC). However, manual delineation is labor-intensive, and automatic CTV contouring for NPC is difficult due to the nasopharyngeal complexity, tumor variability, and judgement-based criteria. To address the above-mentioned problems, we introduce SAM-RT, the first large vision model (LVM) designed for CTV contouring inNPC. Given the anatomical dependency required for CTV contouring which encapsulates the Gross Tumor Volume (GTV) while minimizing exposure to Organs-at-Risk (OAR)-our approach begins with the fine-tuning of the Segment Anything Model (SAM), using a Low-Rank Adaptation (LoRA) strategy for segmenting GTV and OAR across multi-center and multi-modality datasets. This step ensures SAM-RT initially integrates with anatomical prior knowledge for CTV contouring. To optimize the use of previously acquired knowledge, we introduce Sequential LoRA (SeqLoRA) to improve knowledge retention in SAMRT during the fine-tuning for CTV contouring. We further introduce the Prompt-Visual Cross Merging Attention (ProViCMA) for enhanced image and prompt interaction, and the Gate-Regulated Prompt Adjustment (GaRPA) strategy, utilizing learnable gates to direct prompts for effective CTV task adaptation. Efficient utilization of knowledge across relevant datasets is essential due to sparse labeling of medical images for specific tasks. To achieve this, SAM-RT is trained using an information-querying approach. SAM-RT incorporates various prior knowledge: 1) Reliance of CTV on GTV and OAR, and 2) Eliciting expert knowledge in CTV contouring. Extensive quantitative and qualitative experiments validate our designs.

Autonomous Robotic Surgery (ARS) is gaining traction for its potential to ease surgeon workload, enhance surgical consistency, and improve patient outcomes. The accurate perception of the surgical environment, particularly soft tissue deformation, is critical for ARS success. However, existing depth estimation techniques fall short by focusing solely on the distance to the target surface, neglecting post-contact soft tissue deformation. In response to this challenge, a novel approach has been proposed, centered on predicting tissue stiffness to address motion errors arising from soft tissue movement after contact in endoscopic robot systems. In liver puncture scenarios, a network featuring two encoders is employed to predict puncture needle force and movement information based on endoscopic images. Utilizing Hooke’s law, the tissue stiffness is calculated from the predicted data. Subsequently, this forecasted information is utilized to determine the displacement at the targeted operative site resulting from tissue elastic deformation and movement during tissue-instrument interaction. Preliminary experimental results affirm the efficacy of the proposed method in predicting stiffness, demonstrating its potential for facilitating precise tissue-instrument interaction. These results highlight the method’s promise for future applications in the field of ARS.

Robotic ultrasound imaging systems (RUSs) have captured significant interest owing to their potential to facilitate autonomous ultrasound imaging. However, existing RUSs built upon robotic systems oriented towards conventional manufacturing struggle to navigate the variable and dynamic clinical environments. We introduce a portable and lightweight RUS designed to enhance adaptability for ultrasound imaging tasks. The proposed system features multiple parallel rings and bearings, affording it four degrees-of-freedom for precise posture control. Further enhancing its adaptability, the actuators are isolated from the mechanism and connected by a cable-sheath mechanism, resulting in a mere 519g lightweight structure that attaches to the body. Quantitative assessments indicate that within a vast workspace of 981 cm(3), the posture control precision of the probe is measured at 1.32 +/- 0.1 mm and [ 1.8 +/- 1.1 degrees , 1.9 +/- 2.2(degrees) , 0.8 +/- 0.8(degrees) ]. The maximum compression force measured for the probe is 14.5 N. The quantitative evaluation results show that the system can attach to various parts of the human body for image acquisition. In addition, the proposed system excels in performing stable scanning procedures even in rapidly changing dynamic environments. Our system can realize imaging tasks with a much lighter structure and has the potential to be applied to more complex scenarios.

As the micromanipulator of surgical robots works in a narrow space, it is difficult to install any position sensors at the end, so the position control and position detection cannot be accurately performed. A position estimator based on the parameter autonomous selection model is proposed to estimate the end position indirectly. First, a single joint principle prototype and a position estimator model are established through the 4DOF driving scheme of the micromanipulator and the cable-driven model. Second, the proposed parameter change model is combined with the parameter selection method to form a parameter autonomous selection model. Finally, a position estimator based on the parameter autonomous selection model is established. The experimental results show the maximum estimation error of the position estimator is 0.1928 deg. Compared with other position estimation methods, the position estimator proposed in this paper has higher accuracy and better robustness, which lays a foundation for the full closed-loop control of micromanipulator position.

Minimally invasive surgery (MIS) is commonly used in some robotic-assisted surgery (RAS) systems. However, many RAS lack the strength and tactile sensation of surgical tools. Therefore, researchers have developed various force sensing techniques in robot-assisted minimally invasive surgery (RMIS). This paper provides a systematic classification and review of force sensing approaches in the field of RMIS, with a particular focus on direct and indirect force sensing. In this survey, the relevant literature on various sensing principles, haptic sensor design standards, and sensing technologies between 2000 and 2022 is reviewed. This survey can also serve as a roadmap for future developments by identifying the shortcomings of the field and discussing the emerging trends in force sensing methods.

The Poisson multi-Bernoulli mixture (PMBM) filter is an effective tracking framework for tracking multiple extended objects. However, methods based on this framework typically assume that the object's shape is an ellipse, which cannot adequately describe the object's shape. When two objects are spatially close, problems such as difficulty correctly distinguishing the object trajectory and insufficient utilization of the object shape's feature information arise. To address the aforementioned issues, this paper describes the shape of the object in greater detail using the multi-ellipse model. On this premise, the extended object shape will be divided into multiple Gaussian inverse Wishart components, and the likelihood will be calculated. Furthermore, a new partitioning method is proposed to divide the measurements into several most linearly dependent components, which aids in estimating the object shape composed of multiple ellipses. The simulation results show that the new algorithm outperforms the original PMBM algorithm in terms of accuracy.

Safe human-robot interaction and excellent motion flexibility ensure a wide range of potential applications for soft robots in industry and medicine. Although many efforts have been made to empower soft robots with variable stiffness, solutions for miniaturization, wide-range, and rapid-response stiffness have yet to be realized. To maintain the combination of motion flexibility and operation stability, in this work a soft variable stiffness manipulator with a diameter of 10.7 mm based on a low melting point alloy (LMPA) solution is proposed. Inspired by reinforced concrete structures, LMPA structures can be freely designed and optimized by a novel 2D to 3D fabrication method for the first time. The crossed pattern LMPA was selected as the variable stiffness skeleton of the manipulator through comparative experiments. Under hot and cold water circulation, the rigid-soft transition time is reduced to 10–15 seconds due to structure optimization. Tests have shown that the omnidirectional steering manipulator can easily be bent over $180^{\circ }$ with high repeatability (standard deviation of $1.37^{\circ }$ ) in the soft state and lock the shape at any bend angle with a stiffness of about 415.45 N/m in the rigid state. The rapid and wide variation in stiffness of the soft manipulator can be used as a small platform for minimally invasive surgical tools in confined spaces.

Accurately detecting laparoscopic surgical instruments is crucial to ensuring the success of robot-assisted laparoscopic surgery. In this paper, we propose a laparoscopic surgical instrument detection framework called ToolNet-X to address the issues of low accuracy and inadequate real-time performance in current detection methods. The algorithm effectively fuses feature maps of different scales using weighted feature fusion and increasing fusion links. In addition, we introduce gated convolution and recursive operations to enhance the network’s global modeling capabilities. The experimental results show that ToolNet-X achieved a precision of 97.8%, a mAP@0.5 (mean Average Precision) of 96.4%, and a mAP@0.5:0.95 of 59.398% on the M2CAI16-Tool-Locations dataset. Furthermore, we evaluate the performance of the YoloV5 and YoloV7 algorithms in surgical instrument detection and improve the latter using the same approach proposed in this paper, demonstrating significant improvements for both algorithms.

Natural orifice transluminal endoscopic surgery (NOTES) associated with less pain, shorter hospital stay, and outstanding cosmetic results has drawn considerable attention from the academic community. For NOTES, pneumoperitoneum is an essential technology to obtain intraperitoneal surgical space, which, however, can have some serious adverse hemodynamic effects. In this paper, we propose a novel intra-abdominal wall-lifting device that can replace the pneumoperitoneum functionally for transvaginal NOTES (vNOTES). The device consists mainly of two layers of elastic strips, namely, the outer strips that are used to lift the abdominal wall for surgical exposure and the inner strip with a camera mounted that can rotate to provide a wide surgical field of view. With the design, gasless vNOTES can be carried out with 82.74% of the operative field provided by pneumoperitoneum, eliminating all complications associated with gas insufflation. The design principle and kinematic analysis of the device are discussed in detail and validated by experiments. Finally, a prototype is fabricated to demonstrate the feasibility of the proposed concept.

Minimally invasive surgery is gaining wide- spread attention in the laboratory and the clinic due to its advantages of less trauma, less pain, and faster recovery. Miniature continuum instruments with a large central channel are needed for reducing damage to healthy tissue and delivering multiple surgical tools in microsurgery such as fetal surgery and maxillary sinus surgery. In this article, we proposed a novel miniature spring-based cable-driven continuum manipulator characterized by its offset neutral bending plane and its simple and clever structure. These features enable the manipulator to have several advantages such as easy miniaturization, bending stability and uniformity, a large central channel, smaller cable tension required, and low cost. A piecewise constant curvature model, with the offset neutral bending plane taken into account, is presented for the kinematic control of the proposed manipulator. To explore the relationship between the cable tension and manipulator bending angle, the statics model based on the equivalent model principle, is exhaustively discussed. The accuracy of kinematics and statics models, as well as the design superiority, was verified experimentally. Furthermore, teleoperation experiments of peg-transfer tasks and porcine tissue laser ablation are conducted to verify the feasibility and effectiveness of the proposed manipulator in different forms. Final, design variations are showcased to enrich manipulator design and functionality.