We introduce a one-shot segmentation method to alleviate the burden of manual annotation for medical images. The main idea is to treat one-shot segmentation as a classical atlas-based segmentation problem, where voxel-wise correspondence from the atlas to the unlabelled data is learned. Subsequently, segmentation label of the atlas can be transferred to the unlabelled data with the learned correspondence. However, since ground truth correspondence between images is usually unavailable, the learning system must be well-supervised to avoid mode collapse and convergence failure. To overcome this difficulty, we resort to the forward-backward consistency, which is widely used in correspondence problems, and additionally learn the backward correspondences from the warped atlases back to the original atlas. This cycle-correspondence learning design enables a variety of extra, cycle-consistency-based supervision signals to make the training process stable, while also boost the performance. We demonstrate the superiority of our method over both deep learning-based one-shot segmentation methods and a classical multi-atlas segmentation method via thorough experiments.

Underwater glider (UG) is the only marine equipment in the field of ocean observation that can realize unmanned autonomous, all-weather and wide-area continuous 3-D observation. The rapid individualized design (RID) and full lifecycle management (FLM) make two difficulties to be solved in the industrialization development of UGs since the UG system involves interdisciplinary multiphysics and there exists multisource uncertainties in complex ocean environment. With the application of new information technologies in industry, the above problems can be effectively solved by the digital twin (DT)-driven methodology. In this article, an architecture of DT-driven RID and FLM for UGs is first put forward based on our previous theoretical research and technique development. Then, the proposed architecture of DT-driven methodology is researched in detail in terms of its digital modeling, design optimization, virtual verification and practical application. Finally, Petrel developed by China is introduced, and a typical preliminary application of DT-driven methodology is presented based on petrel to verify its feasibility. The architecture proposed in this article is also appropriate for other types of autonomous underwater vehicles.

The current study highlights the development of a novel and high-accuracy Fiber Bragg Grating (FBG)-enabled tri-axial distal force sensor for minimally invasive surgical palpation. This tri-axial sensor designs the primary flexure/elastomer consisting of axial and radial force-sensitive structures in a serial configuration, based on the method of freedom and constraint topology (FACT) from the perspective of the mechanism. This method provides general guidelines for designing multi-dimensional sensor design, achieving excellent sensitivity and a large measurement range, depressing crosstalks and couplings among different axes, and keeping the sensitivity of the three directions at the same order of magnitude. Five tightly suspended optical fibers embedded with one FBG element each have been assembled with the proposed flexure. They are configured with one arranged at the axial flexure’s central line and four distributed along the circumference of the radial flexure. This tight suspension configuration generates uniform and constant strain distribution on the FBG element, thus improving resolution and repeatability and further avoiding FBG chirping. Finite element modeling (FEM)-based simulation has been conducted for performance investigation and design optimization to improve the sensor sensitivity. Static calibration experiments and in-vitro and ex-vivo palpation experiments have been performed to investigate the proposed design’s performances and feasibility. The optimized sensor prototype achieves excellent resolution values of 1.18mN and 1.81mN in the x- and y-directions within [−5N, 5N], and 2.61mN in the z-direction within [0, 5N], realizing the same order of sensitivity magnitude at each axis.

Upper gastrointestinal (UGI) tract pathology is common worldwide. With recent advancements in robotics, innovative diagnostic and treatment devices have been developed and several translational attempts made. This review paper aims to provide a highly pictorial critical review of robotic gastroscopes, so that clinicians and researchers can obtain a swift and comprehensive overview of key technologies and challenges. Therefore, the paper presents robotic gastroscopes, either commercial or at a progressed technology readiness level. Among them, we show tethered and wireless gastroscopes, as well as devices aimed for UGI surgery. The technological features of these instruments, as well as their clinical adoption and performance, are described and compared. Although the existing endoscopic devices have thus far provided substantial improvements in the effectiveness of diagnosis and treatment, there are certain aspects that represent unwavering predicaments of the current gastroenterology practice. A detailed list includes difficulties and risks, such as transmission of communicable diseases (e.g., COVID-19) due to the doctor–patient proximity, unchanged learning curves, variable detection rates, procedure-related adverse events, endoscopists’ and nurses’ burnouts, limited human and/or material resources, and patients’ preferences to choose non-invasive options that further interfere with the successful implementation and adoption of routine screening. The combination of robotics and artificial intelligence, as well as remote telehealth endoscopy services, are also discussed, as viable solutions to improve existing platforms for diagnosis and treatment are emerging.

This paper presents a clamping force sensor based on fiber Bragg grating (FBG) to provide interaction force feedback for laparoscopic surgery. The proposed sensor mainly consists of a force-sensitive clamping flexure and a tightly suspended optical fiber with an FBG inscribed. The force-sensitive clamping flexure utilizes a bridge-type structure with an excellent load-bearing capacity and anti-interference ability to linearly convert the vertical force or displacement input exerted on the grasping surface into translational deformation along the flexure central line. The FBG fiber is arranged along the flexure central axis to sense the vertical force-induced horizontal strain. This two-point pasting configuration can achieve a uniform fiber strain distribution and an enhanced sensitivity. This assembly arrangement produces a linear relationship between the applied vertical force and the force-induced horizontal strain value sensed by the FBG. The force sensitivity remains the constant and it less affected by the grasping positions due to the high stiffness and deformation conversion, overcoming the difficulty that the traditional clamping force sensor designs are sensitive to the grasping position due to the variable grasping positions and areas on the soft tissues during clamping. The finite element method (FEM)-based simulation has been utilized for design optimization and performance investigation to guide the sensor design. The simulation sensitivity values have been determined as a close value of 52pm/N regarding the three different grasping positions at middle, left and right parts. The optimized sensor design has been integrated into a manual surgical grasper and achieve experimental sensitivity values of 56.2pm/N, 51.1pm/N and 47.5pm/N for the three different loading positions within [0, 10N]. Both dynamic loading experiments and clamping experiments on ex-vivo tissues and in-vivo animal for tissue resection were implemented to validate the effectiveness of the proposed design.