OBJECTIVE : The clinical and imaging characteristics of fibromas of the tendon sheath (FTS) closely resemble those of giant cell tumors of the tendon sheath (GCTTS). We aimed to study MRI features of FTS and GCTTS to distinguish the two entities and improve their differential diagnosis. METHODS : We retrospectively analyzed data from 18 patients (9 men, 9 women; age, 17-62 years) and 24 patients (13 men, 11 women; age, 15-67 years) treated between May 2011 and May 2016, with histologically confirmed FTS and GCTTS, respectively. Specific MRI features of the two groups were compared using the independent sample t tests and chi-square tests. RESULTS : FTS exhibited round or oval shapes. Proton-weighted images (PDWI) showed heterogeneous hypointensity that appeared striped or disordered and was located in the lesion center. Enhanced scans demonstrated asymmetrical signal in the foci. GCTTS mostly exhibited a lobulated or casting mold pattern, with a hypointense ring on PDWI. The hypointense components appeared granular/flaky or separated, sometimes behaving as a uniform signal on PDWI. Significant differences in the following features were observed between the two groups: lesion morphology (p < 0.001); imaging features on PDWI, including whether the signal is homogeneous (p < 0.001); the presence of a hypointense ring (p = 0.006); the location and morphology of hypointensity (p < 0.001); bone absorption (p = 0.008); enhancing pattern (p = 0.008); and whether the tumor crossed the joint (p = 0.026). CONCLUSIONS : FTS and GCTTS demonstrate distinctive MRI features, which can be used for differential diagnosis with sensitivities, specificities, and diagnostic accuracies of 83-100%, 29-79%, and 60-89%, respectively. CONCLUSIONS : • Fibromas and giant cell tumors of the tendon sheath have distinct features on MRI, including differences in lesion morphology and intensity patterns, which can be used for differential diagnosis. • Among other signs, GCTTS are more uniform than FTS, and FTS have a striped or disordered pattern. • Tumors were classified with 90% accuracy into either FTS or GCTTS based on a combination of two features: homogenous signal and hypointensity shape on PDWI.