Segmentation of the pancreatic computed tomography (CT) image in the Internet of Medical Things (IoMT) environment faces dual challenges of feature robustness and accurate recognition of small organs. As a clinically critical but anatomically challenging organ, the pancreas exhibits small volume, high inter-patient variability, and low contrast with surrounding tissues, making automatic pancreas segmentation a representative and difficult task in abdominal CT analysis. Existing automatic, machine-centric segmentation methods often perform unsatisfactorily across different medical institutions due to variations in imaging devices, changes in scanning protocols, as well as the irregular shape and blurred boundaries of the pancreas. To address this problem, this paper proposes a SimCLR-based Contrastive Swin-UNet model (ContSwinU), which integrates contrastive learning with the Swin Transformer architecture to achieve feature robustness learning and high-precision pancreas segmentation. Specifically, ContSwinU leverages SimCLR to learn robust feature representations, thereby enhancing the model’s generalization capability across diverse scenarios. Additionally, by incorporating the hierarchical window attention mechanism of the Swin Transformer, the model effectively balances local texture and global structural information, improving segmentation accuracy for the pancreas as a small organ. Experimental results on a public pancreas CT dataset demonstrate that ContSwinU achieves an IoU of 0.8041, a Dice coefficient of 0.8872, and a recall of 0.9014, significantly outperforming mainstream baseline methods. These results indicate that the proposed framework is well suited for challenging small-organ segmentation tasks and has the potential to be extended to other organs and multi-center IoMT scenarios. This study provides an effective solution for pancreas segmentation in IoMT environments and has substantial clinical application value.

The portal-mesenteric venous system surrounding the pancreas exhibits significant anatomical complexity and variability. Precise segmentation and assessment of these venous structures are crucial for preoperative planning in pancreatic surgery to reduce the risk of intraoperative hemorrhage and postoperative complications. To address this, we present the Venous Anatomy Analyzer for Pancreatic Surgery ($V A^{2} P S$), a novel framework that integrates 3D vein segmentation and Large Vision-Language Model (VLM), to provide surgeons with intuitive visualization and accurate analysis of the portalmesenteric venous system and pancreas. VA${ }^{\mathbf{2}}$PS first performs precise 3D segmentation of the venous system and pancreas from CT scans. Subsequently, it bridges the dimensional gap between segmentation result and VLM input by rendering these 3D structures into optimized 2D views. Finally, conditioned on the visual input and surgical instructions,$\text{VA}^{2}$PS utilizes a VLM reasoning over a built-in repository of clinical knowledge to identify venous anatomical subtypes and generates surgical planning recommendations. Objective and subjective evaluations demonstrate that$\text{VA}^{2} \text{PS}$provides accurate and efficient preoperative assessments for surgeons, holding the potential to improve surgical outcomes. Code and data are available at https://github.com/zengyue1376/VA2PS.

Precise segmentation of thermal infrared images is crucial in domains like surveillance, medical diagnostics, intelligent transportation, accurate guidance and remote sensing. However, current thermal segmentation methods often oversimplify by treating thermal images as grayscale, neglecting vital physical factors such as thermal imaging effects and material information, thereby constraining segmentation precision. To address these limitations, we propose TherNet, a novel thermal infrared segmentation framework integrating thermal imaging effects and material physical information. The study elucidates the impacts of object radiation, inter-object thermal exchange, atmospheric scattering, and camera thermal inertia on thermal infrared imaging, developing four modules to model or rectify these physical processes. To validate the proposed framework, two large-scale infrared datasets were created: TI-Cityscapes for multi-class semantic segmentation in traffic scenes (4,200 frames, 18 classes), and TBRSD for single-object blindroad segmentation (5,180 frames from a pedestrian perspective). The proposed methods achieved SoTA performance across three infrared semantic segmentation datasets and the blind road segmentation dataset, underscoring the pivotal role of leveraging physical properties. TherNet provides innovative perspectives and robust benchmarks for future developments in the domain.

Cerebrovascular segmentation from time-of-flight magnetic resonance angiography (TOF-MRA) and computed tomography angiography (CTA) is essential in providing supportive information for diagnosing and treatment planning of multiple intracranial vascular diseases. Different imaging modalities utilize distinct principles to visualize the cerebral vasculature, which leads to the limitations of expensive annotations and performance degradation while training and deploying deep learning models. In this paper, we propose an unsupervised domain adaptation framework CereTS to perform translation and segmentation of cross-modality unpaired cerebral angiography. Considering the commonality of vascular structures and stylistic textures as domain-invariant and domain-specific features, CereTS adopts a multi-level domain alignment pattern that includes an image-level cyclic geometric consistency constraint, a patch-level masked contrastive constraint and a feature-level semantic perception constraint to shrink domain discrepancy while preserving consistency of vascular structures. Conducted on a publicly available TOF-MRA dataset and a private CTA dataset, our experiment shows that CereTS outperforms current state-of-the-art methods by a large margin.

Deep Image Prior (DIP) is a powerful unsupervised learning image restoration technique. However, DIP struggles when handling complex degradation scenarios involving mixed image artifacts. To address this limitation, we propose a novel technique to enhance DIP’s performance in handling mixed image degradation. Our method leverages additional deep denoiser, which is deployed as a denoising engine in the regularization by denoising (RED) framework. A new objective function is constructed by combining DIP with RED, and solved by the alternating direction method of multiplier (ADMM) algorithm. Our method explicitly learns a more comprehensive representation of the underlying image structure and being robust to different types of degradation. Experimental results demonstrate the effectiveness of our method, showing effective improvements in restoring images corrupted by mixed degradation on several image restoration tasks, such as image inpainting, super-resolution and deblurring.

Cardiovascular image registration is an essential approach to combine the advantages of preoperative 3D computed tomography angiograph (CTA) images and intraoperative 2D X-ray/digital subtraction angiography (DSA) images together in minimally invasive vascular interventional surgery (MIVI). Recent studies have shown that convolutional neural network (CNN) regression model can be used to register these two modality vascular images with fast speed and satisfactory accuracy. However, CNN regression model trained by tens of thousands of images of one patient is often unable to be applied to another patient due to the large difference and deformation of vascular structure in different patients. To overcome this challenge, we evaluate the ability of transfer learning (TL) for the registration of 2D/3D deformable cardiovascular images. Frozen weights in the convolutional layers were optimized to find the best common feature extractors for TL. After TL, the training data set size was reduced to 200 for a randomly selected patient to get accurate registration results. We compared the effectiveness of our proposed nonrigid registration model after TL with not only that without TL but also some traditional intensity-based methods to evaluate that our nonrigid model after TL performs better on deformable cardiovascular image registration.

Fast and accurate ellipse detection is critical in certain computer vision tasks. In this paper, we propose an arc adjacency matrix-based ellipse detection (AAMED) method to fulfill this requirement. At first, after segmenting the edges into elliptic arcs, the digraph-based arc adjacency matrix (AAM) is constructed to describe their triple sequential adjacency states. Curvature and region constraints are employed to make the AAM sparse. Secondly, through bidirectionally searching the AAM, we can get all arc combinations which are probably true ellipse candidates. The cumulative-factor (CF) based cumulative matrices (CM) are worked out simultaneously. CF is irrelative to the image context and can be pre-calculated. CM is related to the arcs or arc combinations and can be calculated by the addition or subtraction of CF. Then the ellipses are efficiently fitted from these candidates through twice eigendecomposition of CM using Jacobi method. Finally, a comprehensive validation score is proposed to eliminate false ellipses effectively. The score is mainly influenced by the constraints about adaptive shape, tangent similarity, distribution compensation. Experiments show that our method outperforms the 12 state-of-the-art methods on 9 datasets as a whole, with reference to recall, precision, F-measure, and time-consumption.

The projection of space circle can be used for relative pose measurement of satellite targets. The accuracy of elliptical parameters plays an important role in the accuracy of pose recovery. However, the quality of space visible and infrared image is poor. The traditional ellipse detection method is mainly based on the edge of pixel‐accuracy‐wise and the ellipse accuracy is low, resulting in pose measurement errors. In this study, a sub‐pixel‐accuracy‐wise edge‐based ellipse fitting method is proposed to improve the ellipse accuracy. To realise this goal, we improved the arc‐based ellipse detection method and designed sub‐pixel‐edge‐based ellipse detection method. Experimental results show that the ellipse accuracy fitted by sub‐pixel edge coordinates is at least 50% higher than that by pixel edge coordinates, especially when the ellipse is incomplete. The author's method is the first one to propose and verify the sub‐pixel edge coordinate which contributes to enhancing ellipse accuracy.

This paper presents a novel design for the soft continuum manipulator based on beehive structure. Our design inspiration derives from the beehive hexagon structure, which has some characteristics of compact layout, low material consumption, thinner thickness of wall, and higher structure strength. The manipulator includes several same modules which composed of many hexagon cell units, the deformation of these cell units synthesize the module macroscopic movement. By installing several modules in series, we can form the manipulator arm. For module fabrication, we adopt 3D printing technology and use the silicone rubber as the flexible material; these measures ensure that the complex shape of the module can be achieved and make the module easier to deform elastically. Based on pneumatic technology, we build the control system and verify its operation via some experiments. In the gravity environment, three-section (about 180mm long at uninflated state) manipulator's maximum bending angle reaches about 90°, its elongation is close to 2:1. We also do a five-section experiment to acquire the manipulator's motion under different section combinations. The experiments verified that the manipulator has both good flexibility and operational capabilities.

This paper presents modular self-reconfigurable robots for grasping an unknown object based on a grasp quality metric. The presented self-reconfiguration approach makes use of basic motions performed by individual robots to change the structure. The method to solve for forward kinematics as well as configuration representation is proposed based on relationship between robotic modules. This method can be used to update the estimated contact points in self-reconfiguration. Incorporating a grasp quality metric of form closure into self-reconfiguration, the structure for grasping can be evaluated until a desired metric is obtained. Grasping and self-reconfiguration simulation experiments are presented, showing the availability and effectiveness of the proposed approach.

Telesurgery is an interdisciplinary field of teleoperated robotics and robot-assisted minimally invasive surgery, and has good promising future. In this work, the structure and work flow of the tele-neurosurgery are addressed at first. Then the key technologies involved, including our newly developed surgical robot, untouched methodology in localizing marks for registration, video overlay and surgical simulation, video transmission and network communication, and safeguards, are discussed, which guarantee the safety and feasibility of the clinical telesurgery. In the end, the first remote clinical stereo tactic neurosurgery is presented.