To compare the diagnostic accuracy of diffusion-weighted magnetic resonance imaging (DW-MRI) and fluorine 18 fluorodeoxyglucose positron emission tomography/computed tomography (18 F-FDG PET/CT) in detection of residual or recurrent tumors and their local extension in patients with head and neck squamous cell carcinoma after treatment with (chemo) radiotherapy (CRT). Method: Twenty-five patients (17 men, 8 women, median age 64 years, range 49-79) who underwent surgical salvage for residual or recurrent tumors after CRT were included. The histopathologic analysis after the surgical salvage served as the gold standard. Results: Both DW-MRI and 18 F-FDG PET/CT had a sensitivity of 92 % (23/25) in the detection of residual or recurrent tumors. MRI had a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for detecting perineural spread of 62 % (5/8), 88 % (15/17), 71 % (5/7) and 83 % (15/18), respectively; in comparison, PET/CT did not detect any cases of perineural spread. The sensitivity, specificity, PPV and NPV of MRI in detecting muscle infiltration was 75 % (9/12), 77 % (10/13), 75 % (9/12) and 77 % (10/13) respectively, while the values for 18 F-FDG PET/CT were 58 % (7/12), 69 % (9/13), 64 % (7/11) and 64 % (9/14). Conclusions: DW-MRI-and 18 F-FDG PET/CT-imaging have an identical detection rate of residual or recurrent tumors after (chemo) radiotherapy. MRI has a higher sensitivity in detecting local perineural spread, has a better accuracy in the detection of muscle infiltration and more accurately correlates the lesion size to the histopathologic specimen.