Applying multiparametric magnetic resonance imaging in differentiating benign and malignant ovarian tumors

Abstract

OBJECTIVE: To investigate the application value of multiparametric MRI containing plain scan, diffusion-weighted imaging and contrast-enhanced to differentiate diagnosis of benign and malignant ovarian tumors. METHODS: Totally 93 patients with 110 ovarian tumors confirmed by operation and histopathology were collected, including the benign group (n=62) and malignant group (n=48). MRI features, including location, shape, margin, component, ascites, DWI signal intensity, enhanced degree were analyzed and compared. Additionally, diagnostic performance is assessed and compared. RESULTS: Results the benign group and malignant group of shape(P ＜ 0.001), margin(P ＜ 0.001), component(P＜0.001), ascites(P＜0.001), DWI signal intensity(P＜0.001), enhanced degree(P＜0.001) were statistically significant differences. While, there were no statistically significant differences in the tumor location(P=0.757). In addition, the AUC, sensitivity and specificity of multiparametric MRI for differentiating benign and malignant ovarian tumors were 0.835, 0.835 and 0.750. CONCLUSION: Multiparametric MRI is an effective method in discrimination between benign and malignant ovarian masses. Therefore, MRI could preoperative primarily diagnosis ovarian tumor to provide the clinical evidence for diagnosis and treatment.