Accurate tumor segmentation and treatment outcome prediction with DeepTOP

作者全名："Li, Lanlan; Xu, Bin; Zhuang, Zhuokai; Li, Juan; Hu, Yihuang; Yang, Hui; Wang, Xiaolin; Lin, Jinxin; Zhou, Ruwen; Chen, Weiwei; Ran, Dongzhi; Huang, Meijin; Wang, Dabiao; Luo, Yanxin; Yu, Huichuan"

作者地址："[Li, Lanlan; Xu, Bin; Hu, Yihuang] Fuzhou Univ, Sch Phys & Informat Engn, Fujian 35010, Peoples R China; [Zhuang, Zhuokai; Wang, Xiaolin; Lin, Jinxin; Huang, Meijin; Luo, Yanxin; Yu, Huichuan] Sun Yat Sen Univ, Affiliated Hosp 6, Guangdong Inst Gastroenterol, Guangzhou 510655, Guangdong, Peoples R China; [Zhuang, Zhuokai; Wang, Xiaolin; Lin, Jinxin; Huang, Meijin; Luo, Yanxin; Yu, Huichuan] Sun Yat Sen Univ, Affiliated Hosp 6, Guangdong Prov Key Lab Colorectal & Pelv Floor Di, Guangzhou 510655, Guangdong, Peoples R China; [Zhuang, Zhuokai; Lin, Jinxin; Huang, Meijin; Luo, Yanxin; Yu, Huichuan] Sun Yat Sen Univ, Affiliated Hosp 6, Dept Colorectal Surg, Guangzhou 510655, Guangdong, Peoples R China; [Zhuang, Zhuokai; Lin, Jinxin; Huang, Meijin; Luo, Yanxin; Yu, Huichuan] Sun Yat Sen Univ, Affiliated Hosp 6, Dept Gen Surg, Guangzhou 510655, Guangdong, Peoples R China; [Li, Juan] Sun Yat Sen Univ, Affiliated Hosp 6, Dept Endoscop Surg, Guangzhou 510655, Guangdong, Peoples R China; [Yang, Hui] Sun Yat Sen Univ, Dept Med Imaging Ctr, Collaborat Innovat Ctr Canc Med, State Key Lab Oncol South China,Canc Ctr, Guangzhou 510060, Guangdong, Peoples R China; [Zhou, Ruwen] Columbia Univ, Joseph LMailman Sch Publ Hlth, Dept Biostat, New York, NY 10032 USA; [Chen, Weiwei] Guizhou Med Univ, Dept Clin Med, Guiyang, Peoples R China; [Chen, Weiwei] Guizhou Prov Canc Hosp, Dept Abdominal Oncol, Guiyang, Peoples R China; [Ran, Dongzhi] Univ Arizona, Coll Med, Dept Pharmacol, 1501 North Campbell Dr,POB 245050, Tucson, AZ 85724 USA; [Ran, Dongzhi] Chongqing Med Univ, Dept Pharmacol, Key Lab Biochem & Mol Pharmacol, Chongqing 400016, Peoples R China; [Wang, Dabiao] Fuzhou Univ, Sch Mech Engn & Automat, Fujian 35010, Peoples R China"

通信作者："Luo, YX; Yu, HC (通讯作者)，Sun Yat Sen Univ, Affiliated Hosp 6, Dept Colorectal Surg, Guangzhou 510655, Guangdong, Peoples R China.; Wang, DB (通讯作者)，Fuzhou Univ, Sch Mech Engn & Automat, Fujian 35010, Peoples R China.; Luo, YX (通讯作者)，Sun Yat Sen Univ, Affiliated Hosp 6, Guangdong Inst Gastroenterol, Dept Colorectal Surg,Guangdong Prov Key Lab Color, 26 Yuancun Erheng Rd, Guangzhou 510655, Guangdong, Peoples R China.; Yu, HC (通讯作者)，Sun Yat Sen Univ, Affiliated Hosp 6, Guangdong Inst Gastroenterol, Guangdong Prov Key Lab Colorectal & Pelv Floor Di, 26 Yuancun Erheng Rd, Guangzhou 510655, Guangdong, Peoples R China."

来源：RADIOTHERAPY AND ONCOLOGY

ESI学科分类：CLINICAL MEDICINE

WOS号：WOS:000952377200001

JCR分区：Q1

影响因子：4.9

年份：2023

卷号：183

期号：　

开始页：　

结束页：　

文献类型：Article

关键词：Neural network; Cancer treatment; Magnetic resonance image; Treatment response

摘要："Background: Accurate outcome prediction prior to treatment can facilitate trial design and clinical deci-sion making to achieve better treatment outcome.Method: We developed the DeepTOP tool with deep learning approach for region-of-interest segmenta-tion and clinical outcome prediction using magnetic resonance imaging (MRI). DeepTOP was constructed with an automatic pipeline from tumor segmentation to outcome prediction. In DeepTOP, the segmenta-tion model used U-Net with a codec structure, and the prediction model was built with a three-layer con-volutional neural network. In addition, the weight distribution algorithm was developed and applied in the prediction model to optimize the performance of DeepTOP.Results: A total of 1889 MRI slices from 99 patients in the phase III multicenter randomized clinical trial (NCT01211210) on neoadjuvant treatment for rectal cancer was used to train and validate DeepTOP. We systematically optimized and validated DeepTOP with multiple devised pipelines in the clinical trial, demonstrating a better performance than other competitive algorithms in accurate tumor segmentation (Dice coefficient: 0.79; IoU: 0.75; slice-specific sensitivity: 0.98) and predicting pathological complete response to chemo/radiotherapy (accuracy: 0.789; specificity: 0.725; and sensitivity: 0.812). DeepTOP is a deep learning tool that could avoid manual labeling and feature extraction and realize automatic tumor segmentation and treatment outcome prediction by using the original MRI images.Conclusion: DeepTOP is open to provide a tractable framework for the development of other segmenta-tion and predicting tools in clinical settings. DeepTOP-based tumor assessment can provide a reference for clinical decision making and facilitate imaging marker-driven trial design.(c) 2023 Elsevier B.V. All rights reserved. Radiotherapy and Oncology 183 (2023) 109550"

基金机构："National Natural Science Foundation of China [81972245, 82173067, 81902877]; Natural Science Foundation of Guangdong Province [2022A1515012656, 2021A1515010134]; Natural Science Foundation of Fujian Province [2020J01453]; Sun Yat-sen University Clinical Research 5010 Program [2018026]; ""Five Five"" Talent Team Construction Project of the Sixth Affiliated Hospital of Sun Yat-sen University [P20150227202010251]; Excellent Talent Training Project of the Sixth Affiliated Hospital of Sun Yat-sen University [R2021217202512965]; Sixth Affiliated Hospital of Sun Yat-sen University Clinical Research-'1010' Program; Program of Introducing Talents of Discipline to Universities; National Key Clinical Discipline (2012)"

基金资助正文："This study was supported by the National Natural Science Foundation of China (No.81972245, YL; No.82173067, YL; No.81902877, HY), the Natural Science Foundation of Guangdong Province (No.2022A1515012656, HY; No.2021A1515010134, MH), the Natural Science Foundation of Fujian Province (No.2020J01453, DW), the Sun Yat-sen University Clinical Research 5010 Program (No.2018026, YL), the ""Five Five"" Talent Team Construction Project of the Sixth Affiliated Hospital of Sun Yat-sen University (No. P20150227202010251, YL), the Excellent Talent Training Project of the Sixth Affiliated Hospital of Sun Yat-sen University (No. R2021217202512965, YL), the Sixth Affiliated Hospital of Sun Yat-sen University Clinical Research-'1010' Program (MH), the Program of Introducing Talents of Discipline to Universities, and National Key Clinical Discipline (2012)."