Chenyi Xie1,2, Yihuai Hu3,4, Varut Vardhanabhuti5, Hong Yang6, Lujun Han7, Jianhua Fu3. 1. Department of Radiology, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China. 2. Department of Diagnostic Radiology, Li Ka Shing Faculty of Medicine, University of Hong Kong, Pok Fu Lam, Hong Kong SAR, China. 3. Department of Thoracic Surgery, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Esophageal Cancer Institute, Sun Yat-sen University Cancer Center, Guangzhou, China. 4. Department of Thoracic Surgery, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China. 5. Department of Diagnostic Radiology, Li Ka Shing Faculty of Medicine, University of Hong Kong, Pok Fu Lam, Hong Kong SAR, China. varv@hku.hk. 6. Department of Thoracic Surgery, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Esophageal Cancer Institute, Sun Yat-sen University Cancer Center, Guangzhou, China. yanghong@sysucc.org.cn. 7. Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou, China.
Abstract
BACKGROUND: Lymph node status is vital for prognosis and treatment decisions for esophageal squamous cell carcinoma (ESCC). This study aimed to construct and evaluate an optimal radiomics-based method for a more accurate evaluation of individual regional lymph node status in ESCC and to compare it with traditional size-based measurements. METHODS: The study consecutively collected 3225 regional lymph nodes from 530 ESCC patients receiving upfront surgery from January 2011 to October 2015. Computed tomography (CT) scans for individual lymph nodes were analyzed. The study evaluated the predictive performance of machine-learning models trained on features extracted from two-dimensional (2D) and three-dimensional (3D) radiomics by different contouring methods. Robust and important radiomics features were selected, and classification models were further established and validated. RESULTS: The lymph node metastasis rate was 13.2% (427/3225). The average short-axis diameter was 6.4 mm for benign lymph nodes and 7.9 mm for metastatic lymph nodes. The division of lymph node stations into five regions according to anatomic lymph node drainage (cervical, upper mediastinal, middle mediastinal, lower mediastinal, and abdominal regions) improved the predictive performance. The 2D radiomics method showed optimal diagnostic results, with more efficient segmentation of nodal lesions. In the test set, this optimal model achieved an area under the receiver operating characteristic curve of 0.841-0.891, an accuracy of 84.2-94.7%, a sensitivity of 65.7-83.3%, and a specificity of 84.4-96.7%. CONCLUSIONS: The 2D radiomics-based models noninvasively predicted the metastatic status of an individual lymph node in ESCC and outperformed the conventional size-based measurement. The 2D radiomics-based model could be incorporated into the current clinical workflow to enable better decision-making for treatment strategies.
BACKGROUND: Lymph node status is vital for prognosis and treatment decisions for esophageal squamous cell carcinoma (ESCC). This study aimed to construct and evaluate an optimal radiomics-based method for a more accurate evaluation of individual regional lymph node status in ESCC and to compare it with traditional size-based measurements. METHODS: The study consecutively collected 3225 regional lymph nodes from 530 ESCC patients receiving upfront surgery from January 2011 to October 2015. Computed tomography (CT) scans for individual lymph nodes were analyzed. The study evaluated the predictive performance of machine-learning models trained on features extracted from two-dimensional (2D) and three-dimensional (3D) radiomics by different contouring methods. Robust and important radiomics features were selected, and classification models were further established and validated. RESULTS: The lymph node metastasis rate was 13.2% (427/3225). The average short-axis diameter was 6.4 mm for benign lymph nodes and 7.9 mm for metastatic lymph nodes. The division of lymph node stations into five regions according to anatomic lymph node drainage (cervical, upper mediastinal, middle mediastinal, lower mediastinal, and abdominal regions) improved the predictive performance. The 2D radiomics method showed optimal diagnostic results, with more efficient segmentation of nodal lesions. In the test set, this optimal model achieved an area under the receiver operating characteristic curve of 0.841-0.891, an accuracy of 84.2-94.7%, a sensitivity of 65.7-83.3%, and a specificity of 84.4-96.7%. CONCLUSIONS: The 2D radiomics-based models noninvasively predicted the metastatic status of an individual lymph node in ESCC and outperformed the conventional size-based measurement. The 2D radiomics-based model could be incorporated into the current clinical workflow to enable better decision-making for treatment strategies.
Authors: Asad Kutup; Michael F Nentwich; Elfriede Bollschweiler; Dean Bogoevski; Jakob R Izbicki; Arnulf H Hölscher Journal: Ann Surg Date: 2014-12 Impact factor: 12.969
Authors: Thomas W Rice; Hemant Ishwaran; Wayne L Hofstetter; Paul H Schipper; Kenneth A Kesler; Simon Law; E M R Lerut; Chadrick E Denlinger; Jarmo A Salo; Walter J Scott; Thomas J Watson; Mark S Allen; Long-Qi Chen; Valerie W Rusch; Robert J Cerfolio; James D Luketich; Andre Duranceau; Gail E Darling; Manuel Pera; Carolyn Apperson-Hansen; Eugene H Blackstone Journal: Ann Surg Date: 2017-01 Impact factor: 12.969
Authors: Emmanuel Gabriel; Kristopher Attwood; William Du; Rebecca Tuttle; Raed M Alnaji; Steven Nurkin; Usha Malhotra; Steven N Hochwald; Moshim Kukar Journal: JAMA Surg Date: 2016-03 Impact factor: 14.766
Authors: Nabil P Rizk; Hemant Ishwaran; Thomas W Rice; Long-Qi Chen; Paul H Schipper; Kenneth A Kesler; Simon Law; Toni E M R Lerut; Carolyn E Reed; Jarmo A Salo; Walter J Scott; Wayne L Hofstetter; Thomas J Watson; Mark S Allen; Valerie W Rusch; Eugene H Blackstone Journal: Ann Surg Date: 2010-01 Impact factor: 12.969