Characterization of malignancy of adnexal lesions using ADC entropy: comparison with mean ADC and qualitative DWI assessment.

PURPOSE To establish the utility of apparent diffusion coefficient (ADC) entropy in discrimination of benign and malignant adnexal lesions, using histopathology as the reference standard, via comparison of the diagnostic performance of ADC entropy with mean ADC and with visual assessments of adnexal lesions on conventional and diffusion-weighted sequences. MATERIALS AND METHODS In all, 37 adult female patients with an ovarian mass that was resected between June 2006 and January 2011 were included. Volume-of-interest was drawn to incorporate all lesion voxels on every slice that included the mass on the ADC map, from which whole-lesion mean ADC and ADC entropy were calculated. Two independent radiologists also rated each lesion as benign or malignant based on visual assessment of all sequences. The Mann-Whitney test and logistic regression for correlated data were used to compare performance of mean ADC, ADC entropy, and the visual assessments. RESULTS No statistically significant difference was observed in mean ADC between benign and malignant adnexal lesions (P = 0.768). ADC entropy was significantly higher in malignant than in benign lesions (P = 0.009). Accuracy was significantly greater for ADC entropy than for mean ADC (0.018). ADC entropy and visual assessment by the less-experienced reader showed similar accuracy (P ≥ 0.204). The more experienced reader's accuracy was significantly greater than that of all other assessments (P ≤ 0.039). CONCLUSION ADC entropy showed significantly greater accuracy than the more traditional metric of mean ADC for distinguishing benign and malignant adnexal lesions. Although whole-lesion ADC entropy provides a straightforward and objective measurement, its potential benefit decreases with greater reader experience.