Ensembled deep learning model outperforms human experts in diagnosing biliary atresia from sonographic gallbladder images

被引：82

作者：

Zhou, Wenying ^{[1
]}

Yang, Yang ^{[2
]}

Yu, Cheng ^{[3
]}

Liu, Juxian ^{[4
]}

Duan, Xingxing ^{[5
]}

Weng, Zongjie ^{[6
]}

Chen, Dan ^{[7
]}

Liang, Qianhong ^{[8
]}

Fang, Qin ^{[9
]}

Zhou, Jiaojiao ^{[4
]}

Ju, Hao ^{[10
]}

Luo, Zhenhua ^{[11
]}

Guo, Weihao ^{[1
]}

Ma, Xiaoyan ^{[7
]}

Xie, Xiaoyan ^{[1
]}

Wang, Ruixuan ^{[2
]}

Zhou, Luyao ^{[1
]}

机构：

[1] Sun Yat Sen Univ, Affiliated Hosp 1, Inst Diagnost & Intervent Ultrasound, Dept Med Ultrason, Guangzhou, Peoples R China

[2] Sun Yat Sen Univ, Sch Comp Sci & Engn, Guangzhou, Peoples R China

[3] Huazhong Univ Sci & Technol, Union Hosp, Tongji Med Coll, Dept Ultrasound, Wuhan, Peoples R China

[4] Sichuan Univ, West China Hosp, Dept Ultrasound, Chengdu, Peoples R China

[5] Hunan Childrens Hosp, Dept Ultrasound, Changsha, Peoples R China

[6] Fujian Med Univ, Affiliated Hosp, Fujian Prov Matern & Childrens Hosp, Dept Med Ultrason, Fuzhou, Peoples R China

[7] Guangdong Women & Children Hosp, Dept Ultrasound, Guangzhou, Peoples R China

[8] Southern Med Univ, Hexian Mem Affiliated Hosp, Dept Ultrasound, Guangzhou, Peoples R China

[9] First Peoples Hosp Foshan, Dept Ultrasound, Foshan, Peoples R China

[10] China Med Univ, Shengjing Hosp, Dept Ultrasound, Shenyang, Peoples R China

[11] Sun Yat Sen Univ, Affiliated Hosp 1, Inst Precis Med, Guangzhou, Peoples R China

来源：

NATURE COMMUNICATIONS | 2021年 / 12卷 / 01期

基金：

中国国家自然科学基金;

关键词：

TRIANGULAR CORD SIGN; US; EPIDEMIOLOGY; ULTRASOUND; FREQUENCY;

D O I：

10.1038/s41467-021-21466-z

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

It is still challenging to make accurate diagnosis of biliary atresia (BA) with sonographic gallbladder images particularly in rural area without relevant expertise. To help diagnose BA based on sonographic gallbladder images, an ensembled deep learning model is developed. The model yields a patient-level sensitivity 93.1% and specificity 93.9% [with areas under the receiver operating characteristic curve of 0.956 (95% confidence interval: 0.928-0.977)] on the multi-center external validation dataset, superior to that of human experts. With the help of the model, the performances of human experts with various levels are improved. Moreover, the diagnosis based on smartphone photos of sonographic gallbladder images through a smartphone app and based on video sequences by the model still yields expert-level performances. The ensembled deep learning model in this study provides a solution to help radiologists improve the diagnosis of BA in various clinical application scenarios, particularly in rural and undeveloped regions with limited expertise. It is still challenging to make accurate diagnosis of biliary atresia (BA) with sonographic gallbladder images particularly in rural areas without relevant expertise. Here, the authors develop a diagnostic deep learning model which favourable performance in comparison with human experts in multi-center external validation.

引用

页数：14

共 39 条

[1] End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography [J].