Jing Gong1,2, Jiyu Liu3, Wen Hao1, Shengdong Nie4, Bin Zheng5, Shengping Wang6,7, Weijun Peng8,9. 1. Department of Radiology, Fudan University Shanghai Cancer Center, 270 Dongan Road, Shanghai, 200032, China. 2. Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, 200032, China. 3. Department of Radiology, Shanghai Pulmonary Hospital, 507 Zheng Min Road, Shanghai, 200433, China. 4. School of Medical Instrument and Food Engineering, University of Shanghai for Science and Technology, 516 Jun Gong Road, Shanghai, 200093, China. 5. School of Electrical and Computer Engineering, University of Oklahoma, Norman, OK, 73019, USA. 6. Department of Radiology, Fudan University Shanghai Cancer Center, 270 Dongan Road, Shanghai, 200032, China. shengpingwang2007@126.com. 7. Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, 200032, China. shengpingwang2007@126.com. 8. Department of Radiology, Fudan University Shanghai Cancer Center, 270 Dongan Road, Shanghai, 200032, China. cjr.pengweijun@vip.163.com. 9. Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, 200032, China. cjr.pengweijun@vip.163.com.
Abstract
OBJECTIVE: To develop a deep learning-based artificial intelligence (AI) scheme for predicting the likelihood of the ground-glass nodule (GGN) detected on CT images being invasive adenocarcinoma (IA) and also compare the accuracy of this AI scheme with that of two radiologists. METHODS: First, we retrospectively collected 828 histopathologically confirmed GGNs of 644 patients from two centers. Among them, 209 GGNs are confirmed IA and 619 are non-IA, including 409 adenocarcinomas in situ and 210 minimally invasive adenocarcinomas. Second, we applied a series of pre-preprocessing techniques, such as image resampling, rescaling and cropping, and data augmentation, to process original CT images and generate new training and testing images. Third, we built an AI scheme based on a deep convolutional neural network by using a residual learning architecture and batch normalization technique. Finally, we conducted an observer study and compared the prediction performance of the AI scheme with that of two radiologists using an independent dataset with 102 GGNs. RESULTS: The new AI scheme yielded an area under the receiver operating characteristic curve (AUC) of 0.92 ± 0.03 in classifying between IA and non-IA GGNs, which is equivalent to the senior radiologist's performance (AUC 0.92 ± 0.03) and higher than the score of the junior radiologist (AUC 0.90 ± 0.03). The Kappa value of two sets of subjective prediction scores generated by two radiologists is 0.6. CONCLUSIONS: The study result demonstrates using an AI scheme to improve the performance in predicting IA, which can help improve the development of a more effective personalized cancer treatment paradigm. KEY POINTS: • The feasibility of using a deep learning method to predict the likelihood of the ground-glass nodule being invasive adenocarcinoma. • Residual learning-based CNN model improves the performance in classifying between IA and non-IA nodules. • Artificial intelligence (AI) scheme yields higher performance than radiologists in predicting invasive adenocarcinoma.
OBJECTIVE: To develop a deep learning-based artificial intelligence (AI) scheme for predicting the likelihood of the ground-glass nodule (GGN) detected on CT images being invasive adenocarcinoma (IA) and also compare the accuracy of this AI scheme with that of two radiologists. METHODS: First, we retrospectively collected 828 histopathologically confirmed GGNs of 644 patients from two centers. Among them, 209 GGNs are confirmed IA and 619 are non-IA, including 409 adenocarcinomas in situ and 210 minimally invasive adenocarcinomas. Second, we applied a series of pre-preprocessing techniques, such as image resampling, rescaling and cropping, and data augmentation, to process original CT images and generate new training and testing images. Third, we built an AI scheme based on a deep convolutional neural network by using a residual learning architecture and batch normalization technique. Finally, we conducted an observer study and compared the prediction performance of the AI scheme with that of two radiologists using an independent dataset with 102 GGNs. RESULTS: The new AI scheme yielded an area under the receiver operating characteristic curve (AUC) of 0.92 ± 0.03 in classifying between IA and non-IA GGNs, which is equivalent to the senior radiologist's performance (AUC 0.92 ± 0.03) and higher than the score of the junior radiologist (AUC 0.90 ± 0.03). The Kappa value of two sets of subjective prediction scores generated by two radiologists is 0.6. CONCLUSIONS: The study result demonstrates using an AI scheme to improve the performance in predicting IA, which can help improve the development of a more effective personalized cancer treatment paradigm. KEY POINTS: • The feasibility of using a deep learning method to predict the likelihood of the ground-glass nodule being invasive adenocarcinoma. • Residual learning-based CNN model improves the performance in classifying between IA and non-IA nodules. • Artificial intelligence (AI) scheme yields higher performance than radiologists in predicting invasive adenocarcinoma.
Authors: Julien G Cohen; Jin Mo Goo; Roh-Eul Yoo; Chang Min Park; Chang Hyun Lee; Bram van Ginneken; Doo Hyun Chung; Young Tae Kim Journal: Eur Radiol Date: 2016-04-05 Impact factor: 5.315
Authors: Hee-Dong Chae; Chang Min Park; Sang Joon Park; Sang Min Lee; Kwang Gi Kim; Jin Mo Goo Journal: Radiology Date: 2014-08-01 Impact factor: 11.105