This paper presents a design and optimization of a dexterous tendon-driven robotic hand that each finger is under-actuated. The robotic hand has 9 DOFs and 6 motors with a modular design concept of three fingers. A theoretical analysis had been carried out to verify the mechanical structure of the proposed hand, optimize the location of the pulleys for efficient transmission and workspace by exam the kinematics model of the designed hand. Experimental results show good grasping performance of the proposed hand. It demonstrates that our three-fingered robotic hand can grasp these objects with different shapes.

Prostate cancer is one of the male cancers with high mortality rate in the world. The mainstream diagnostic method is puncture biopsy. Accurate detection and segmentation of the prostate in magnetic resonance (MR) and transrectal ultrasonography (TRUS) images is an important prerequisite for image-guided puncture biopsy. Manual or semiautomatic detection and segmentation methods and traditional automatic detection and segmentation methods are difficult to achieve real-time and accurate detection and segmentation under the conditions of poor imaging effect or less labeled data. In this article, a real-time and accurate automatic detection and segmentation algorithm dense residual convolutional (DRC) U-Net for 3-D MR and TRUS images of prostate is proposed for prostate puncture biopsy. A novel convolution module that integrates residual connections, dense connections, and deep separable convolutions is proposed for multiscale feature extraction and fusion. At the same time, deep supervision mechanism and a variety of attention mechanisms are integrated in the network to improve training efficiency and ensure segmentation effect. The experimental results on the public MR dataset Promise12 and the private TRUS dataset show that the real-time performance and accuracy of this method are better than the existing 3-D medical image detection and segmentation methods, such as 3-D U-Net and V-Net, and also have some advantages compared with the most advanced 3-D medical image detection and segmentation methods, which proves that the proposed network structure has important clinical significance and practical value for image-guided puncture biopsy.

Humanoid robots with anthropomorphic hands can assist or replace humans in performing complex tasks in dynamic scenarios, such as industrial manufacturing. This article presents a rigid-soft coupling anthropomorphic hand system aimed at improving grasping performance by focusing on dexterity, adaptability, and system integration. The hand adopts a bionic design inspired by the human hand, featuring rigid-soft fingers fabricated using a multimaterial layered manufacturing approach and optimized thumb parameters to maximize the enveloping grasp range. A lightweight, embedded control system is implemented alongside a gesture-coordinated control method based on disturbance observer, further enhancing functionality. Weighing approximately 600 g and resembling adult hand dimensions, the hand demonstrates human-like dexterity. Its ability to leverage environmental constraints for grasping reduces dependence on perception and planning, enabling secure and efficient interactions. This study offers meaningful insights into deploying humanoid robots for intricate tasks in dynamic environments.