Machine learning-based FDG PET-CT radiomics for outcome prediction in larynx and hypopharynx squamous cell carcinoma

•Radiomic features predict disease recurrence in larynx/hypopharyx cancer.•Machine-learning methods selected the radiomic/clinical features for the 4 models.•The 4 models were PET only, CT only, clinical features only and combined PET and CT.•The combined PET/CT radiomics model had the highest AUC value (0.94).