Accurate Tumor Segmentation via Octave Convolution Neural Network

Frontiers in Medicine

Lihua An¹  Nana Luo¹  Haixia Feng¹  Jingyang Ai²  Bo Yang³  Jingyi Yang⁴  Zheng You⁵  Bo Wang⁶ 

[1] Affiliated Hospital of Jining Medical University, Jining, China;Beijing Jingzhen Medical Technology Ltd., Beijing, China;China Institute of Marine Technology & Economy, Beijing, China;School of Artificial Intelligence, Xidian University, Xi'an, China;The State Key Laboratory of Precision Measurement Technology and Instruments, Department of Precision Instrument, Tsinghua University, Beijing, China;Innovation Center for Future Chips, Tsinghua University, Beijing, China;The State Key Laboratory of Precision Measurement Technology and Instruments, Department of Precision Instrument, Tsinghua University, Beijing, China;Innovation Center for Future Chips, Tsinghua University, Beijing, China;Beijing Jingzhen Medical Technology Ltd., Beijing, China;

关键词: liver;    liver tumor;    deep learning;    octave convolution;    segmentation;   

DOI  :  10.3389/fmed.2021.653913

来源: Frontiers

【 摘 要 】

Three-dimensional (3D) liver tumor segmentation from Computed Tomography (CT) images is a prerequisite for computer-aided diagnosis, treatment planning, and monitoring of liver cancer. Despite many years of research, 3D liver tumor segmentation remains a challenging task. In this paper, we propose an effective and efficient method for tumor segmentation in liver CT images using encoder-decoder based octave convolution networks. Compared with other convolution networks utilizing standard convolution for feature extraction, the proposed method utilizes octave convolutions for learning multiple-spatial-frequency features, thus can better capture tumors with varying sizes and shapes. The proposed network takes advantage of a fully convolutional architecture which performs efficient end-to-end learning and inference. More importantly, we introduce a deep supervision mechanism during the learning process to combat potential optimization difficulties, and thus the model can acquire a much faster convergence rate and more powerful discrimination capability. Finally, we integrate octave convolutions into the encoder-decoder architecture of UNet, which can generate high resolution tumor segmentation in one single forward feeding without post-processing steps. Both architectures are trained on a subset of the LiTS (Liver Tumor Segmentation) Challenge. The proposed approach is shown to significantly outperform other networks in terms of various accuracy measures and processing speed.

【 授权许可】

CC BY