Endoscopic Segmentation of Kidney Stone based on Transfer Learning
作者单位：The Institute of Robotics and Automatic Information System Tianjin Key Laboratory of Intelligent Robotics College of Artificial Intelligence NanKai University Department of Urology Civil Aviation General Hospital Department of Urology Forth Medical Center of Chinese PLA General Hospital Department of Urology Changhai Hospital
会议名称：《第40届中国控制会议》
会议日期：2021年
学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 1002[医学-临床医学] 100210[医学-外科学(含：普外、骨外、泌尿外、胸心外、神外、整形、烧伤、野战外)] 081104[工学-模式识别与智能系统] 08[工学] 080203[工学-机械设计及理论] 0835[工学-软件工程] 0802[工学-机械工程] 0811[工学-控制科学与工程] 10[医学] 0812[工学-计算机科学与技术（可授工学、理学学位）]
关 键 词：Endoscopic image Kidney stone Image segmentation Semantic segmentation
摘 要：Endoscopic stone segmentation is of great significance in the comprehensive diagnosis and surgical planning of ureteroscopic lithotripsy. Due to the quality problems of endoscopic imaging, such as artifact, highlight, reflection, contrast imbalance and blur, it is a great challenge to segment kidney stone fragments of different shapes. In this study, the improved U-Net model was used to extract the kidney stone fragment area at the pixel level, and its contour information could be accurately obtained. The VGG16 network with strong portability is used as the encoder of U-Net model to extract the semantic information of multiple feature layers, and the up-sampling is realized by using transposal convolution to gradually restore the segmentation *** the experiment, DCE Loss combining Dice Loss and Cross Entropy Loss is adopted as the Loss function of the *** experimental results show that the improved U-Net model has higher accuracy for stone segmentation from endoscopic images, with the MPA, MIoU and F1 score of 96.44%, 97.62% and 97.03% respectively. The F1 score of the modified model is 2.25% higher than that of Deeplabv3 + model, and 14.24% higher than that of standard U-Net model.