Histogram Analysis of Apparent Diffusion Coefficient on Diffusion Weighted Magnetic Resonance Imaging in Differentiation between Low and High Grade Serous Ovarian Cancer

Background Ovarian cancer is a leading cause of death in gynecological malignancies. Being the most common subtype in OEC, ovarian serous cancer also includes two subtypes: low grade serous ovarian cancer (LGSC) and high grade serous ovarian cancer (HGSC) (1). Purpose The study aims to assess the capability of apparent diffusion coefficient (ADC) histogram analysis and conventional measurements on magnetic resonance imaging (MRI) in differentiating between LGSC and HGSC. Methods We retrospectively recruited 38 patients with pathologically proven ovarian serous epithelial cancer. The mean ADC value was measured by one technician using two methods on post-processed workstation. The ADC value and histogram parameter difference between LGSC and HGSC groups were compared. The correlation between the ADC value and the Ki-67 expression was calculated across both groups. Results The repeatability of ADC measurements across two methods was good; the ROI method (ADC-roi) had better performance repeatability than the area method (ADC-area). The value of ADC-(mean)?ADC-(min)?ADC-(max,) and ADC-(area) significantly differed between both groups (p < 0.001). The value of ADC-(area) correlated inversely with ki-67 expression in the whole group (Pearson coefficient = -0.382, p = 0.02). The 3D computerized-diagnostic model had the best discriminative performance in determining HGSC than 2D and conventional ADC measurements. The 3D model yielded a sensitivity of 100%, a specificity of 95.45%, and an accuracy of 97.73%. Conclusion In the present study, the 3D ADC histogram model help differentiate HGSC from LGSC with a better performance than conventional ADC measurements.