Transfer Learning Method for Prenatal Ultrasound Diagnosis of Biliary Atresia

Overview

Journal NPJ Digit Med

Specialty Medical Informatics

Date 2025 Feb 28

PMID 40021764

Authors

Fujiao He

Gang Li

Zhichao Zhang

Chaoran Yang

Zeyu Yang

Hao Ding

Dan Zhao

Wei Sun

Yu Wang

Kaihui Zeng

Xian Li

Mingming Shao

Jiao Yin

Jia Yao

Boxuan Hong

Zhibo Zhang

Zhengwei Yuan

Zongjie Weng

Luyao Zhou

Mo Zhang

Lizhu Chen

Affiliations

Soon will be listed here.

Abstract

Biliary atresia (BA) is a rare and severe congenital disorder with a significant challenge for prenatal diagnosis. This study, registered at the Chinese Clinical Trial Registry (ChiCTR2200059705), aimed to develop an intelligent model to aid in the prenatal diagnosis of BA. To develop and evaluate this model, fetuses from 20 hospitals across China and infants sourced from public database were collected. The transfer-learning model (TLM) demonstrated superior diagnostic performance compared to the basic deep-learning model, with higher area under the curves of 0.906 (95%CI: 0.872-0.940) vs 0.793 (0.743-0.843), 0.914 (0.875-0.953) vs 0.790 (0.727-0.853), and 0.907 (0.869-0.945) vs 0.880 (0.838-0.922) for the three independent test cohorts. Furthermore, when aided by the TLM, diagnostic accuracy surpassed that of individual sonologists alone. The TLM achieved satisfactory performance in predicting fetal BA, providing a low-cost, easily accessible, and accurate diagnostic tool for this condition, making it an effective aid in clinical practice.

References

Di Pasquo E, Kuleva M, Rousseau A, Vitucci A, Sonigo P, Chardot C . Outcome of non-visualization of fetal gallbladder on second-trimester ultrasound: cohort study and systematic review of literature. Ultrasound Obstet Gynecol. 2019; 54(5):582-588. DOI: 10.1002/uog.20252. View

Xie H, Wang N, He M, Zhang L, Cai H, Xian J . Using deep-learning algorithms to classify fetal brain ultrasound images as normal or abnormal. Ultrasound Obstet Gynecol. 2020; 56(4):579-587. DOI: 10.1002/uog.21967. View

Zeng K, Yang Z, Chen L, Sun W, Wang Y, Chen C . Prediction of Fetal Biliary Atresia Based on Second and Third-Trimester Ultrasound Characteristics. Ultraschall Med. 2022; 44(3):307-317. DOI: 10.1055/a-1562-1615. View

Zhou W, Zhou L . Ultrasound for the Diagnosis of Biliary Atresia: From Conventional Ultrasound to Artificial Intelligence. Diagnostics (Basel). 2022; 12(1). PMC: 8775261. DOI: 10.3390/diagnostics12010051. View

Salomon L, Alfirevic Z, Berghella V, Bilardo C, Hernandez-Andrade E, Johnsen S . Practice guidelines for performance of the routine mid-trimester fetal ultrasound scan. Ultrasound Obstet Gynecol. 2010; 37(1):116-26. DOI: 10.1002/uog.8831. View

Harpavat S, Garcia-Prats J, Anaya C, Brandt M, Lupo P, Finegold M . Diagnostic Yield of Newborn Screening for Biliary Atresia Using Direct or Conjugated Bilirubin Measurements. JAMA. 2020; 323(12):1141-1150. PMC: 7093763. DOI: 10.1001/jama.2020.0837. View

Napolitano M, Franchi-Abella S, Damasio M, Augdal T, Avni F, Bruno C . Practical approach to imaging diagnosis of biliary atresia, Part 1: prenatal ultrasound and magnetic resonance imaging, and postnatal ultrasound. Pediatr Radiol. 2020; 51(2):314-331. DOI: 10.1007/s00247-020-04840-9. View

He F, Feng S, Xiu Y, Zhang Y, Wang Y, Zhang Z . Dysmorphic Gallbladder Found on Prenatal Ultrasound as a Hint for Biliary Atresia. J Ultrasound Med. 2022; 42(6):1345-1351. DOI: 10.1002/jum.16145. View

Moon M, Cho J, Kim J, Lee Y, Jung S, Lee M . In utero development of the fetal gall bladder in the Korean population. Korean J Radiol. 2008; 9(1):54-8. PMC: 2627174. DOI: 10.3348/kjr.2008.9.1.54. View

10.

Sendra-Balcells C, Campello V, Martin-Isla C, Vilades D, Descalzo M, Guala A . Domain generalization in deep learning for contrast-enhanced imaging. Comput Biol Med. 2022; 149:106052. DOI: 10.1016/j.compbiomed.2022.106052. View

11.

Salomon L, Alfirevic Z, Berghella V, Bilardo C, Chalouhi G, da Silva Costa F . ISUOG Practice Guidelines (updated): performance of the routine mid-trimester fetal ultrasound scan. Ultrasound Obstet Gynecol. 2022; 59(6):840-856. DOI: 10.1002/uog.24888. View

12.

Arnaout R, Curran L, Zhao Y, Levine J, Chinn E, Moon-Grady A . An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease. Nat Med. 2021; 27(5):882-891. PMC: 8380434. DOI: 10.1038/s41591-021-01342-5. View

13.

Lin M, He X, Guo H, He M, Zhang L, Xian J . Use of real-time artificial intelligence in detection of abnormal image patterns in standard sonographic reference planes in screening for fetal intracranial malformations. Ultrasound Obstet Gynecol. 2021; 59(3):304-316. DOI: 10.1002/uog.24843. View

14.

Cai C, Wang S, Xu Y, Zhang W, Tang K, Ouyang Q . Transfer Learning for Drug Discovery. J Med Chem. 2020; 63(16):8683-8694. DOI: 10.1021/acs.jmedchem.9b02147. View

15.

Zhou W, Yang Y, Yu C, Liu J, Duan X, Weng Z . Ensembled deep learning model outperforms human experts in diagnosing biliary atresia from sonographic gallbladder images. Nat Commun. 2021; 12(1):1259. PMC: 7904842. DOI: 10.1038/s41467-021-21466-z. View

16.

Zhang Y, Hong D, McClement D, Oladosu O, Pridham G, Slaney G . Grad-CAM helps interpret the deep learning models trained to classify multiple sclerosis types using clinical brain magnetic resonance imaging. J Neurosci Methods. 2021; 353:109098. DOI: 10.1016/j.jneumeth.2021.109098. View

17.

Harpavat S, Garcia-Prats J, Shneider B . Newborn Bilirubin Screening for Biliary Atresia. N Engl J Med. 2016; 375(6):605-6. DOI: 10.1056/NEJMc1601230. View

18.

Chen L, He F, Zeng K, Wang B, Li J, Zhao D . Differentiation of cystic biliary atresia and choledochal cysts using prenatal ultrasonography. Ultrasonography. 2021; 41(1):140-149. PMC: 8696141. DOI: 10.14366/usg.21028. View

19.

Li J, Li Y, Tan J, Liu C . It takes two: Dual Branch Augmentation Module for domain generalization. Neural Netw. 2024; 172:106094. DOI: 10.1016/j.neunet.2023.106094. View

20.

Arooj S, Atta-Ur-Rahman , Zubair M, Khan M, Alissa K, Khan M . Breast Cancer Detection and Classification Empowered With Transfer Learning. Front Public Health. 2022; 10:924432. PMC: 9289190. DOI: 10.3389/fpubh.2022.924432. View