A Novel Combined Level Set Model for Carpus Segmentation from Magnetic Resonance Images with Prior Knowledge aligned in Polar Coordinate System

SAA-SDM: Neural Networks Faster Learned to Segment Organ Images

In the field of medicine, rapidly and accurately segmenting organs in medical images is a crucial application of computer technology. This paper introduces a feature map module, Strength Attention Area Signed Distance Map (SAA-SDM), based on the principal component analysis (PCA) principle. The module is designed to accelerate neural networks' convergence speed in rapidly achieving high precision. SAA-SDM provides the neural network with confidence information regarding the target and background, similar to the signed distance map (SDM), thereby enhancing the network's understanding of semantic information related to the target. Furthermore, this paper presents a training scheme tailored for the module, aiming to achieve finer segmentation and improved generalization performance. Validation of our approach is carried out using TRUS and chest X-ray datasets. Experimental results demonstrate that our method significantly enhances neural networks' convergence speed and precision. For instance, the convergence speed of UNet and UNET +  + is improved by more than 30%. Moreover, Segformer achieves an increase of over 6% and 3% in mIoU (mean Intersection over Union) on two test datasets without requiring pre-trained parameters. Our approach reduces the time and resource costs associated with training neural networks for organ segmentation tasks while effectively guiding the network to achieve meaningful learning even without pre-trained parameters.

Wrist Bone Motion during Flexion-Extension and Radial-Ulnar Deviation: An MRI Study

The wrist joint plays a vital role in activities of daily living. Clinical applications, e.g., therapeutic planning, prosthesis design, and wrist biomechanical analysis, require a detailed understanding of wrist maneuvers and motion. The lack of soft tissue information, motion analysis on limited carpal bones, etc., restrain the investigation of wrist kinematics. In this study, we established 3D models of carpal bones with their cartilages, and revealed the helical axes (HA) of all eight carpal bones for the first time. Both left and right hands at different positions of flexion-extension (FE) and radial-ulnar deviation (RUD) from five subjects were in-vivo imaged through a magnetic resonance imaging device. We segmented all of the bones, including cartilage information in the wrist joint, after which we explored the kinematics of all carpal bones with the HA method. The results showed that the HA of all carpal bones for FE bounded tightly and was mainly located slightly above the radius. During the RUD, carpal bones in the distal row rotated along with wrist movement while the scaphoid, lunate, and triquetrum primarily flexed and extended. Further results reported that the carpal bones translated greater in RUD than in FE. With the generation of more delicate wrist models and thorough investigations of carpal motion, a better understanding of wrist kinematics was obtained for further pathologic assessment and surgical treatment.

Reply to Nikolaidis, P.T.; Afonso, J. Comment on "Eschweiler et al. Anatomy, Biomechanics, and Loads of the Wrist Joint. Life 2022, 12, 188"

Pantelis Nikolaidis and Jose Afonso published a letter [...].

Smoothing the Undersampled Carpal Bone Model with Small Volume and Large Curvature: A Feasibility Study

The carpal bones are eight small bones with irregularities and high curvature on their surfaces. The 3D model of the carpal bone serves as the foundation of further clinical applications, e.g., wrist kinematic behavior. However, due to the limitation of the Magnetic Resonance Imaging (MRI) technique, reconstructed carpal bone models are discretely undersampled, which has dramatic stair-step effects and leads to abnormal meshes on edges or surfaces, etc. Our study focuses on determining the viability of various smoothing techniques for a carpal model reconstructed by in vivo gathered MR images. Five algorithms, namely the Laplacian smoothing algorithm, the Laplacian smoothing algorithm with pre-dilation, the scale-dependent Laplacian algorithm, the curvature flow algorithm, and the inverse distance algorithm, were chosen for evaluation. The assessment took into account the Relative Volume Difference and the Hausdorff Distance as well as the surface quality and the preservation of morphological and morphometric properties. For the five algorithms, we analyzed the Relative Volume Difference and the Hausdorff Distance for all eight carpal bones. Among all the algorithms, the scale-dependent Laplacian method processed the best result regarding surface quality and the preservation of morphological and morphometric properties. Based on our extensive examinations, the scale-dependent Laplacian algorithm is suitable for the undersampled carpal bone model with small volume and large curvature.

Automatic Segmentation for Favourable Delineation of Ten Wrist Bones on Wrist Radiographs Using Convolutional Neural Network

Purpose: This study aimed to develop and validate an automatic segmentation algorithm for the boundary delineation of ten wrist bones, consisting of eight carpal and two distal forearm bones, using a convolutional neural network (CNN). Methods: We performed a retrospective study using adult wrist radiographs. We labeled the ground truth masking of wrist bones, and propose that the Fine Mask R-CNN consisted of wrist regions of interest (ROI) using a Single-Shot Multibox Detector (SSD) and segmentation via Mask R-CNN, plus the extended mask head. The primary outcome was an improvement in the prediction of delineation via the network combined with ground truth masking, and this was compared between two networks through five-fold validations. Results: In total, 702 images were labeled for the segmentation of ten wrist bones. The overall performance (mean (SD] of Dice coefficient) of the auto-segmentation of the ten wrist bones improved from 0.93 (0.01) using Mask R-CNN to 0.95 (0.01) using Fine Mask R-CNN (p < 0.001). The values of each wrist bone were higher when using the Fine Mask R-CNN than when using the alternative (all p < 0.001). The value derived for the distal radius was the highest, and that for the trapezoid was the lowest in both networks. Conclusion: Our proposed Fine Mask R-CNN model achieved good performance in the automatic segmentation of ten overlapping wrist bones derived from adult wrist radiographs.