Fully automatic tumor segmentation of breast ultrasound images with deep learning

• Type of Material: Article
• Authors: Shuai Zhang, Nanjing University; Mei Liao, Sun Yat‐Sen University; Jing Wang, Emory University; Yongyi Zhu, Sun Yat‐sen University; Yanling Zhang, Sun Yat‐sen University; Jian Zhang, Nanjing University; Rongqin Zheng, Sun Yat‐sen University; Linyang Lv, Precis Care technol; Dejiang Zhu, Precis Care technol; Hao Chen, Precis Care technol; Wei Wang, Nanjing University
• Language: English
• Date: 2022-12-09
• Publisher: WILEY
• Publication Version: Final Published Version
• Copyright Statement: © 2022 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, LLC on behalf of The American Association of Physicists in Medicine.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1002/acm2.13863
• Title of Journal or Parent Work: JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS
• Volume: 24
• Issue: 1
• Start Page: e13863
• End Page: e13863
• Grant/Funding Information: This work was funded by National Natural Science Foundation of China (Grant No. 11774158 to J.Z., and 11934008 to W.W.).
• Keywords: CANCER DETECTION; automatic segmentation; breast ultrasound; deep learning; Life Sciences & Biomedicine; Science & Technology; NETWORK; breast cancer; Radiology, Nuclear Medicine & Medical Imaging
• Research Categories: Health Sciences, Oncology; Physics, General

Abstract

Background: Breast ultrasound (BUS) imaging is one of the most prevalent approaches for the detection of breast cancers. Tumor segmentation of BUS images can facilitate doctors in localizing tumors and is a necessary step for computer-aided diagnosis systems. While the majority of clinical BUS scans are normal ones without tumors, segmentation approaches such as U-Net often predict mass regions for these images. Such false-positive problem becomes serious if a fully automatic artificial intelligence system is used for routine screening. Methods: In this study, we proposed a novel model which is more suitable for routine BUS screening. The model contains a classification branch that determines whether the image is normal or with tumors, and a segmentation branch that outlines tumors. Two branches share the same encoder network. We also built a new dataset that contains 1600 BUS images from 625 patients for training and a testing dataset with 130 images from 120 patients for testing. The dataset is the largest one with pixel-wise masks manually segmented by experienced radiologists. Our code is available at https://github.com/szhangNJU/BUS_segmentation. Results: The area under the receiver operating characteristic curve (AUC) for classifying images into normal/abnormal categories was 0.991. The dice similarity coefficient (DSC) for segmentation of mass regions was 0.898, better than the state-of-the-art models. Testing on an external dataset gave a similar performance, demonstrating a good transferability of our model. Moreover, we simulated the use of the model in actual clinic practice by processing videos recorded during BUS scans; the model gave very low false-positive predictions on normal images without sacrificing sensitivities for images with tumors. Conclusions: Our model achieved better segmentation performance than the state-of-the-art models and showed a good transferability on an external test set. The proposed deep learning architecture holds potential for use in fully automatic BUS health screening.