The contrast-enhanced MRI can be substituted by unenhanced MRI in identifying and automatically segmenting primary nasopharyngeal carcinoma with the aid of deep learning models: An exploratory study in large-scale population of endemic area. (April 2022)

• Record Type: Journal Article
• Authors: Deng, Yishu; Li, Chaofeng; Lv, Xing; Xia, Weixiong; Shen, Lujun; Jing, Bingzhong; Li, Bin; Guo, Xiang; Sun, Ying; Xie, Chuanmiao; Ke, Liangru
• Is Part Of: Computer methods and programs in biomedicine. Volume 217(2022)
• Publication Date: 2022-04
• Subjects: Nasopharyngeal carcinoma -- Deep learning -- Differential diagnosis -- Segmentation -- Enhanced examination; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2022.106702
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: The deep learning models developed with enhanced or unenhanced MRI exhibited similar satisfactory performance in discriminating any T stage of NPC from benign hyperplasia. The merged output of T1WI and T2WI models exhibited equivalent segmentation performance for NPC to T1WIC model with limit influence of the T stage. Unenhanced MRI could substitute enhanced MRI in detecting and automatic segmentation for primary NPC with the aid of AI models. Abstract: Background and objectives: Administration of contrast is not desirable for all cases in clinical setting, and no consensus in sequence selection for deep learning model development has been achieved, thus we aim to explore whether contrast-enhanced magnetic resonance imaging (ceMRI) can be substituted in the identification and segmentation of nasopharyngeal carcinoma (NPC) with the aid of deep learning models in a large-scale cohort. Methods: A total of 4478 eligible individuals were randomly split into training, validation and test sets, and self-constrained 3D DenseNet and V-Net models were developed using axial T1-weighted imaging (T1WI), T2WI or enhanced T1WI (T1WIC) images separately. The differential diagnostic performance between NPC and benign hyperplasia were compared among models using chi-square test. Segmentation evaluation metrics, including dice similarity coefficient (DSC) and average surface distance (ASD), were compared using paired student's t -test between T1WIC and T1WI or T2WI models or M_T1/T2, a merged output of malignant region derived from T1WI and T2WI models. Results: All models exhibited similar satisfactory diagnostic performance in discriminating NPC from benign hyperplasia, all attaining overall accuracy over 99.00% in all T stages of NPC. And T1WIC model exhibited similar average DSC and ASD with those of M_T1/T2 (DSC, 0.768±0.070 vs 0.764±0.070; ASD, 1.573±10.954 mm vs 1.626±10.975 mm 1.626±0.975 mm vs 1.573±0.954 mm, all p > 0.0167) in primary NPC using DenseNet, but yielded a significantly higher DSC and lower ASD than either T1WI model or T2WI model (DSC, 0.759±0.065 or 0.755±0.071; ASD, 1.661±0.898 mm or 1.722±1.133 mm, respectively, all p < 0.01) in the entire test set of NPC cohort. Moreover, the average DSCs and ASDs were not statistically significant between T1WIC model and M_T1/T2 in both.