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.

Introduction Biliary atresia (BA) is a life-threatening disease with persistent neonatal cholestasis and progressive liver fibrosis. Timely non-invasive diagnosis of BA can result in early hepatic portoenterostomy (HPE) and better prognosis. Quantitative elastography enables the non-invasive measurement of liver stiffness. However, the studies on elastography methods in the diagnosis of BA and the prediction of post-HPE outcomes vary in their results and have small sample sizes. Thus, we propose this systematic review and meta-analysis to obtain comprehensive evidence on the value of elastography in BA. Methods and analysis We will search the PubMed, Embase and the Cochrane Central Register of Controlled Trials databases for studies evaluating the diagnostic performance of elastography in patients with BA and the prognostic value of postoperative elastography, from inception to 31 December 2020. We plan to use the Quality Assessment of Diagnostic Accuracy Studies-2 list and the Quality In Prognosis Studies tool to assess the risk of bias in the included studies and the study quality. We will evaluate the diagnostic performance of elastography by synthesising the pooled sensitivity, pooled specificity, pooled positive likelihood ratio, pooled negative likelihood ratio, pooled diagnostic OR and summary receiver operating characteristic curve using Meta-Disc V.1.4. We will evaluate the predictive value of elastography after HPE by synthesising the pooled correlation coefficient and pooled OR of prognostic outcomes using STATA V.14. The funnel plot and Egger’s test will be used to evaluate the potential publication bias. Sensitivity analysis will be conducted by examining the estimated effects of individual studies. Ethics and dissemination As this study is a meta-analysis based on previously published literature, ethical approval is not necessary according to the ethics committee of West China Hospital, Sichuan University. The results of this study will be published in a peer-reviewed journal. PROSPERO registration number CRD42020162055.

Original Article A novel prediction tool based on shear wave elastography, gallbladder ultrasound, and serum biomarkers for the early diagnosis of biliary atresia in infants younger than 60 days old Hualin Yan, Juxian Liu, Shuguang Jin, Lanxin Du, Qi Wang, Yan Luo