A feasible method to determine the depth of myometrial invasion of endometrial carcinoma based on improved thinning algorithm

Endometrial carcinoma (EC) is one of the most common gynecological malignant tumors. The five-year survival rate of patients with early EC without metastatic disease is between 74% and 91%. The myometrial invasion depth (MID) is a key factor in determining the stage of EC and subsequent treatment options. Accurate preoperative staging helps to develop a treatment option and provides patients with the opportunity for minor surgery. This paper proposes a method to determine the MID of EC based on improved thinning algorithm. The accuracy, sensitivity, specificity, PPV and NPV of this method in detecting the MID of early EC are 81%, 86%, 71%, 87% and 68%, respectively. The method has the detection accuracy of 86% in 50 magnetic resonance imaging (MRI) images of EC stage IA, and the detection accuracy of 71.4% in 21 MRI images of EC stage IB. We have demonstrated the feasibility of this method for MID detection in MRI images, which helps radiologist to improve the diagnosis accuracy in their clinical work.