作者
Yuting Gao,Shaoqun Zeng,Xinzhe Xiong,Ganwei Cai,Z. Wang,Xu Xie,Chi Jiang,Xue Jiao,J. Liu,Ran Li,Shenzhen Yao,X. Li,Kun Song,Jiao Tang,Hui Xing,Zichao Yu,Shaoqing Zeng,Qi Zhang,Chengqi Yi,Beihua Kong,Xuejun Xie,Donglin Ma,Xin Li,Qingguo Gao
摘要
Objective: Since the advent of transvaginal ultrasound, the detection rate of accessory masses has risen sharply, while the low accuracy of conventional ultrasound often leads to overtreatment. Existing methods for the detection of ovarian cancer are limited in diagnosing early-stage ovarian cancer, which could substantially improve survival. We aimed to develop and validate a deep convolutional neural network (DCNN)-enabled model to improve the diagnostic accuracy of ovarian cancer-based pelvic ultrasound images.