Habitat-based radiomics analysis utilizing Synthetic MRI for the prediction of lymph node metastasis in oral cancer

Motivation: Current methods for assessing lymph node metastasis preoperatively are insufficient in providing precise and time-efficient evaluations. Goal(s): To develop a habitat-based radiomics model for predicting lymph node metastasis in oral cancer using Synthetic MRI (SyMRI). Approach: 101 oral cancer patients were included. Radiomics models for the entire tumor, intratumor, and peritumor regions were developed using synthetic quantitative images. A nomogram was constructed by integrating the radiomics score and MRI reported status. Results: The habitat-based radiomics signature, which integrated both intra- and peritumoral features based on quantitative T1 and T2 maps, showed enhanced predictive capability for preoperative lymph node metastasis. Impact: This habitat-based radiomics model improves accuracy in predicting lymph node metastasis in oral cancer, aiding clinical decision-making and patient management.