Beyond radiologist-level liver lesion detection on multi-phase contrast-enhanced CT images by deep learning
Lei Wu,
Haishuai Wang,
Yining Chen,
Xiang Zhang,
Tianyun Zhang,
Ning Shen,
Guangyu Tao,
Zhongquan Sun,
Yuan Ding,
Weilin Wang,
Jiajun Bu
Affiliations
Lei Wu
Zhejiang Provincial Key Laboratory of Service Robot, College of Computer Science, Zhejiang University, Hangzhou, China; Pujian Technology, Hangzhou, Zhejiang, China
Haishuai Wang
Zhejiang Provincial Key Laboratory of Service Robot, College of Computer Science, Zhejiang University, Hangzhou, China; Corresponding author
Yining Chen
Department of Hepatobiliary and Pancreatic Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
Xiang Zhang
Department of Computer Science, University of North Carolina at Charlotte, Charlotte, NC, USA; Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
Tianyun Zhang
Liangzhu Laboratory, Zhejiang University Medical Center, Hangzhou, Zhejiang, China
Ning Shen
Liangzhu Laboratory, Zhejiang University Medical Center, Hangzhou, Zhejiang, China
Guangyu Tao
Department of Radiology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
Zhongquan Sun
Department of Hepatobiliary and Pancreatic Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
Yuan Ding
Department of Hepatobiliary and Pancreatic Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China; Corresponding author
Weilin Wang
Department of Hepatobiliary and Pancreatic Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China; Corresponding author
Jiajun Bu
Zhejiang Provincial Key Laboratory of Service Robot, College of Computer Science, Zhejiang University, Hangzhou, China; Corresponding author
Summary: Accurate detection of liver lesions from multi-phase contrast-enhanced CT (CECT) scans is a fundamental step for precise liver diagnosis and treatment. However, the analysis of multi-phase contexts is heavily challenged by the misalignment caused by respiration coupled with the movement of organs. Here, we proposed an AI system for multi-phase liver lesion segmentation (named MULLET) for precise and fully automatic segmentation of real-patient CECT images. MULLET enables effectively embedding the important ROIs of CECT images and exploring multi-phase contexts by introducing a transformer-based attention mechanism. Evaluated on 1,229 CECT scans from 1,197 patients, MULLET demonstrated significant performance gains in terms of Dice, Recall, and F2 score, which are 5.80%, 6.57%, and 5.87% higher than state of the arts, respectively. MULLET has been successfully deployed in real-world settings. The deployed AI web server provides a powerful system to boost clinical workflows of liver lesion diagnosis and could be straightforwardly extended to general CECT analyses.
