Purpose: In our study, we aimed to show the efficiency of diffusion-weighted images at different b-values and apparent diffusion coefficient (ADC) values in the differentiation of recurrent tumours from post-treatment tissue changes. Material and methods: The conventional and diffusion magnetic resonance images (MRIs) of 42 patients operated for soft tissue sarcomas between June 2012 and March 2015 followed up with MRIs that were evaluated by 2 radiologists retrospectively. Diffusion MRIs were acquired at 4 different b-values (50, 400, 800, 1000 s/mm2). The lesions were classified according to conventional MRI findings as post-treatment changes and recurrent tumours. Results: When the patient group with recurrent tumours was compared with the patient group with postoperative changes the ADC calculations were statistically significantly lower for the recurrent tumours at all b-levels (p < 0.001 for all b-levels). The sensitivity of b-50 values lower than 3.01 × 103 mm²/s in showing recurrent tumours was 100% and the specificity was 77.78%. The sensitivity of b-400 values lower than 2.1 × 103 mm²/s in showing recurrent tumours was 80% and the specificity was 96.3%. The sensitivity of b-800 values lower than 2.26 × 103 mm²/s in showing recurrent tumours was 100% and the specificity was 88.89%. The sensitivity of b-1000 values lower than 2 × 103 mm²/s in showing recurrent tumours was 93.3% and the specificity was 92.5%. Conclusions: The ADC values obtained from diffusion-weighted images have high sensitivity and specificity in differentiating recurring soft tissue sarcomas during monitoring after treatment from postoperative changes.