Prediction of 5-year Progression-Free Survival in Advanced Nasopharyngeal Carcinoma with Pretreatment PET/CT using Multi-Modality Deep Learning-based Radiomics

Bingxin Gu and Mingyuan Meng contributed equally to this work. Abstract Objective: Deep Learning-based Radiomics (DLR) has achieved great success in medical image analysis and has been considered as a replacement to conventional radiomics that relies on handcrafted features. In this study, we aimed to explore the capability of DLR for the prediction of 5-year Progression-Free Survival (PFS) in advanced Nasopharyngeal Carcinoma (NPC) using pretreatment PET/CT images. Methods: A total of 257 patients (170/87 patients in internal/external cohorts) with advanced NPC (TNM stage III or IVa) were enrolled. We developed an end-to-end multi-modality DLR model, in which a 3D convolutional neural network was optimized to extract deep features from pretreatment PET/CT images and predict the probability of 5-year PFS. TNM stage, as a high-level clinical feature, could be integrated into our DLR model to further improve the prognostic performance. For a comparison between conventional radiomics and DLR, 1456 handcrafted features were extracted, and optimal conventional radiomics methods were selected from 54 cross-combinations of 6 feature selection methods and 9 classification methods. In addition, risk group stratification was performed with clinical signature, conventional radiomics signature, and DLR signature. Results: Our multi-modality DLR model using both PET and CT achieved higher prognostic performance (AUC = 0.842 0.034 and 0.823 0.012 for the internal and external cohorts) than the optimal conventional radiomics method (AUC = 0.796 0.033 and 0.782 0.012). Furthermore, the multi-modality DLR model outperformed single-modality DLR models using only PET (AUC = 0.818 0.029 and 0.796 0.009) or only CT (AUC = 0.657 0.055 and 0.645 0.021). For risk group stratification, the conventional radiomics signature and DLR signature enabled significant difference between the high-and low-risk patient groups in the both internal and external cohorts (P < 0.001), while the clinical signature failed in the external cohort (P = 0.177).