AUTHOR=Liu Jun , Wu Tao , Peng Yun , Luo Rongguang 

TITLE=Grade Prediction of Bleeding Volume in Cesarean Section of Patients With Pernicious Placenta Previa Based on Deep Learning

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 8 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2020.00343

DOI=10.3389/fbioe.2020.00343

ISSN=2296-4185

ABSTRACT=In order to predict the amount of bleeding in the cesarean section of the patients with pernicious placenta previa, and to provide the evidence for the formulation of hemostasis plan before the operation, this paper designs an automatic prediction method based on MRI uterus image. Firstly, the method uses the DeepLab-V3+ network to segment the original MRI abdominal image to obtain the uterine region image. Then, the uterine region image and the corresponding blood loss data are trained by VGGNet-16 network, and the classification model of blood loss level is obtained. The date show that the accuracy, sensitivity and specificity of the classification model are 75.61%, 73.75% and 77.46% respectively on 82 sets of positive and 128 sets of negative MRI images. The results show that this method has a potential clinical application in the prediction of the bleeding volume of cesarean section.