Vascular interventional surgery (VIS) robots are capable of performing various operations such as drug delivery, coil filling, balloon stent expansion, percutaneous catheter ablation, thrombectomy, and more. Some reviews of VIS robots have focused on slave robots and master controllers. However, to the best of our knowledge, there is no review paper on the comprehensive description of all technologies of the VIS robots. Therefore, this paper comprehensively analyzes the key technologies of VIS robot. First, according to the type of VIS, the progress of the VIS robots and some typical applications are summarized. Second, the research status and existing problems of four key technologies related to the VIS robots are discussed: surgical instrument delivery system, force perception and haptic feedback technology, surgical navigation with multimodal image fusion and virtual surgical systems. Finally, we analyze the obstacles to VIS robot development and outline the future directions of active research areas.

Segmentation of intracranial aneurysms (IAs) is an important step for the diagnosis and treatment of IAs. However, the process by which clinicians manually recognize and localize IAs is overly labor intensive. This study aims to develop a deep-learning-based framework (defined as FSTIF-UNet) towards IAs segmentation in un-reconstructed 3D Rotational Angiography (3D-RA) images. 3D-RA sequences from 300 patients with IAs from Beijing Tiantan Hospital are enrolled. Inspired by radiologists' clincial skills, a Skip-Review attention mechanism is proposed to repeatedly fuse the long-term spatiotemporal features of several images with the most obvious IA's features (sellected by a pre-detection network). Then, a Conv-LSTM is used to fuse the short-term spatiotemporal features of the selected 15 3D-RA images from the equally-spaced viewing angles. The combination of the two modules realizes the full-scale spatiotemporal information fusion of the 3D-RA sequence. FSTIF-UNet achieves DSC, IoU, Sens, Haus, and F1-Score of 0.9109, 0.8586, 0.9314, 1.358 and 0.8883, respectively, and time taken for network segmentation is 0.89s/case. The results show significant improvement in IA segmentation performance with FSTIF-UNet compared with baseline networks (with DSC from 0.8486 - 0.8794). The proposed FSTIF-UNet establishes a practical method to assist the radiologists in clinical diagnosis.

Multirotor drones are widely used in fields such as environmental monitoring, agricultural inspection, and package delivery, but they still face numerous challenges in durability and aerial operation capabilities. To address these issues, this paper presents a biomimetic leg-claw mechanism (LCM) inspired by the biomechanics of birds. The claw of the LCM adopts a bistable gripper design that can rapidly close through external impact or actively close via the coordination of internal mechanisms. Additionally, its foldable, parallelogram-shaped legs bend under external forces, stretching the main tendon. A ratchet and pawl mechanism at the knee joint locks the leg in the bent position, thereby enhancing the gripping force of the claw. This paper calculates and experimentally verifies the degrees of freedom in different states, the forces required to open and close the gripper, the application scenarios of active and passive grasping, and the maximum load capacity of the mechanism. Furthermore, perching experiments demonstrate that the LCM enables the drone to perch stably on objects of varying diameters.