Deep Convolutional Neural Network for Nasopharyngeal Carcinoma Discrimination on MRI by Comparison of Hierarchical and Simple Layered Convolutional Neural Networks
Li Ji,
Rongzhi Mao,
Jian Wu,
Cheng Ge,
Feng Xiao,
Xiaojun Xu,
Liangxu Xie,
Xiaofeng Gu
Affiliations
Li Ji
Department of Otorhinolaryngology, The Second People’s Hospital of Changzhou Affiliated to Nanjing Medical University, Changzhou 213003, China
Rongzhi Mao
Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213001, China
Jian Wu
Department of Otorhinolaryngology, The Second People’s Hospital of Changzhou Affiliated to Nanjing Medical University, Changzhou 213003, China
Cheng Ge
Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213001, China
Feng Xiao
Department of Otorhinolaryngology, The Second People’s Hospital of Changzhou Affiliated to Nanjing Medical University, Changzhou 213003, China
Xiaojun Xu
Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213001, China
Liangxu Xie
Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213001, China
Xiaofeng Gu
Department of Otorhinolaryngology, The Second People’s Hospital of Changzhou Affiliated to Nanjing Medical University, Changzhou 213003, China
Nasopharyngeal carcinoma (NPC) is one of the most common head and neck cancers. Early diagnosis plays a critical role in the treatment of NPC. To aid diagnosis, deep learning methods can provide interpretable clues for identifying NPC from magnetic resonance images (MRI). To identify the optimal models, we compared the discrimination performance of hierarchical and simple layered convolutional neural networks (CNN). Retrospectively, we collected the MRI images of patients and manually built the tailored NPC image dataset. We examined the performance of the representative CNN models including shallow CNN, ResNet50, ResNet101, and EfficientNet-B7. By fine-tuning, shallow CNN, ResNet50, ResNet101, and EfficientNet-B7 achieved the precision of 72.2%, 94.4%, 92.6%, and 88.4%, displaying the superiority of deep hierarchical neural networks. Among the examined models, ResNet50 with pre-trained weights demonstrated the best classification performance over other types of CNN with accuracy, precision, and an F1-score of 0.93, 0.94, and 0.93, respectively. The fine-tuned ResNet50 achieved the highest prediction performance and can be used as a potential tool for aiding the diagnosis of NPC tumors.
