Three-dimensional reconstruction of nerve fascicle is important in the analysis of biological characteristics in thearm. The topology of fascicle has been used by doctors to investigate the nerve direction and the relationshipbetween the individual nerve fascicle. However, there still does not exist an ideal internal fascicle and 3D model inthe human peripheral nerve. Accurate segmentation of fascicle from CT images is a crucial step to obtain reliable3D nerve fascicle model. Traditional method in the fascicle segmentation is not efficient due to time consuming,manual work and poor generalization capacity. In this study, we proposed an efficient deep segmentation networkand then reconstruct 3D nerve fascicle model. The proposed network explores the intra-slice contextual featureswith convolutional long short-term memory for accurate fascicle segmentation, and model long-range semanticinformation among image slices. Transfer learning technique is integrated with ResNet34, and the discriminativecapability of intermediate features are further improved. The proposed network architecture is efficient, exibleand suitable for separating the adhesive fascicle. Our approach is the first deep learning method for nervessegmentation. The proposed approach achieves state-of-the-art performance on our dataset, where the meanDice of our method is 95.4% and at least 5% more than other methods.
展开▼
