文献类型：期刊文献

英文题名：Pulmonary nodule risk classification in adenocarcinoma from CT images using deep CNN with scale transfer module

作者：Zheng, Jie[1];Yang, Dawei[2,3];Zhu, Yu[1];Gu, Wanghuan[1];Zheng, Bingbing[1];Bai, Chunxue[2,3];Zhao, Lin[4];Shi, Hongcheng[5];Hu, Jie[2,3];Lu, Shaohua[6];Shi, Weibing[7];Wang, Ningfang[2]

机构：[1]East China Univ Sci & Technol, Sch Informat Sci & Engn, Shanghai 200237, Peoples R China;[2]Fudan Univ, Zhongshan Hosp, Dept Pulm Med, Shanghai 200032, Peoples R China;[3]Shanghai Resp Res Inst, Shanghai 200032, Peoples R China;[4]Jining Med Univ, Rizhao Peoples Hosp, Dept Resp & Crit Care Med, Rizhao 276800, Shandong, Peoples R China;[5]Fudan Univ, Zhongshan Hosp, Dept Nucl Med, Shanghai 200032, Peoples R China;[6]Fudan Univ, Zhongshan Hosp, Dept Pathol, Shanghai 200032, Peoples R China;[7]Fudan Univ, Zhongshan Hosp, Med Examinat Ctr, Shanghai 200032, Peoples R China

年份：2020

卷号：14

期号：8

起止页码：1481

外文期刊名：IET IMAGE PROCESSING

收录：;EI(收录号：20202408815144);WOS:【SCI-EXPANDED(收录号:WOS:000537949300006)】；

基金：The authors greatly appreciate the financial supported by the Zhongshan Hospital Clinical Research Foundation Nos. 2016ZSLC05 and 2016ZSCX02, the National Key Scientific and Technology Support Program No. 2013BAI09B09, the Natural Science Foundation of Shanghai No. 15ZR1408700, the Shandong Medical and Health Science and Technology Development Plan Project (2017WS717) and Support Project for Young Teachers of Jining Medical University (JY2016KJ053Y). The first two authors contributed equally to this article, and both should be considered first authors.

语种：英文

外文关键词：lung; learning (artificial intelligence); cancer; biological organs; medical image processing; image classification; patient diagnosis; neural nets; feature extraction; computerised tomography; pulmonary nodule risk classification; CT images; deep CNN; scale transfer module; lung cancer; clinical treatment decision; early diagnosis; lung adenocarcinoma; imaging studies; deep convolutional neural network; named STM-Net; different resolution images; computed tomography database; Zhongshan Hospital Fudan University; lung adenocarcinomas risk; minimally invasive adenocarcinoma; authors; risk stage prediction; classification accuracy; pulmonary nodules classification; physicians diagnosis pulmonary nodules risk classification; early-stage

摘要：Pulmonary nodules risk classification in adenocarcinoma is essential for early detection of lung cancer and clinical treatment decision. Improving the level of early diagnosis and the identification of small lung adenocarcinoma has been always an important topic for imaging studies. In this study, the authors propose a deep convolutional neural network (CNN) with scale-transfer module (STM) and incorporate multi-feature fusion operation, named STM-Net. This network can amplify small targets and adapt to different resolution images. The evaluation data were obtained from the computed tomography (CT) database provided by Zhongshan Hospital Fudan University (ZSDB). All data have a pathological label and their lung adenocarcinomas risk are classified into four categories: atypical adenomatous hyperplasia, adenocarcinoma in situ, minimally invasive adenocarcinoma, and invasive adenocarcinoma. The authors' deep learning network STM-Net was trained and tested for the risk stage prediction. The accuracy and the average area under the receiver operating characteristic curve achieved by their method are 95.455% and 0.987 for the ZSDB dataset. The experimental results show that STM-Net largely boosts classification accuracy on the pulmonary nodules classification compared with state-of-the-art approaches. The proposed method will be an effective auxiliary to help physicians diagnosis pulmonary nodules risk classification in adenocarcinoma in early-stage.