Histogram-based diffusion metrics for differentiation of benign &malignant gynecologic lesions

Abstract

Purpose: To evaluate the ability of ADC based histogram parameters to differentiate between benign and malignant gynecologic lesions. Results: The study was performed on 100 cases having 66 benign lesions &34 malignant lesions which were pathologically proved. There was statistically significant difference between benign and malignant lesions on qualitative diffusion weighted. The range of traditional partial slice mean ADC parameter for benign masses was (1.095 ± 0.315 x 10−3 mm2/s) and for malignant lesions was (0.791 ± 0.244 x 10−3 mm2/s) which was lower than the values of the benign lesions. there was statistically significant difference as regard mean and minimum ADC values (P = 0.01*, ) The cutoff mean &minimum ADC values used to differentiate benign lesions from malignant lesions were 0.905 & 0.680 x 10-3 mm2/s, respectively. We found that The sensitivity, specificity and accuracy of partial slice mean ADC were higher than histogram mean ADC. Conclusions: Mean & minimum ADC can differentiate between benign and malignant gynecologic lesions with less diagnostic performance than traditional partial slice mean ADC.