Observational Study
Copyright ©The Author(s) 2022. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Jun 14, 2022; 28(22): 2494-2508
Published online Jun 14, 2022. doi: 10.3748/wjg.v28.i22.2494
Accurate and generalizable quantitative scoring of liver steatosis from ultrasound images via scalable deep learning
Bowen Li, Dar-In Tai, Ke Yan, Yi-Cheng Chen, Cheng-Jen Chen, Shiu-Feng Huang, Tse-Hwa Hsu, Wan-Ting Yu, Jing Xiao, Lu Le, Adam P Harrison
Bowen Li, Ke Yan, Lu Le, Adam P Harrison, Research and Development, PAII Inc., Bethesda, MD 20817, United States
Dar-In Tai, Yi-Cheng Chen, Cheng-Jen Chen, Tse-Hwa Hsu, Wan-Ting Yu, Department of Gastroenterology and Hepatology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan 33305, Taiwan
Shiu-Feng Huang, Division of Molecular and Genomic Medicine, National Health Research Institute, Taoyuan 33305, Taiwan
Jing Xiao, Research and Development, Ping An Insurance Group, Shenzhen 518001, Guangdong, China
Author contributions: Li B contributed to software, visualization, and writing original draft and investigation; Yan K contributed to formal analysis, visualization and writing review and editing; Chen YC, Chen CJ, Huang SF, Hsu TH, and Yu WT contributed to data curation; Huang SF contributed to resources; Xiao J contributed to project administration and funding acquisition, and writing review and editing; Lu L contributed to project administration, supervision, writing review and editing, and resources; Harrison AP contributed to formal analysis, supervision, software, writing review and editing, investigation, and methodology; Tai DI contributed to supervision, data curation, conceptualization, writing review and editing, validation, project administration, investigation, resources, and methodology.
Supported by the Maintenance Project of the Center for Artificial Intelligence, No. CLRPG3H0012 and No. SMRPG3I0011.
Institutional review board statement: The study was reviewed and approved by the Institutional Review Board of the Chang Gung Medical Foundation (CGMH IRB No. 201801283B0).
Informed consent statement: Patients were not required to give informed consent to the study.
Conflict-of-interest statement: All authors declare no conflict of interest.
Data sharing statement: No additional data are available.
STROBE statement: The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Dar-In Tai, MD, PhD, Professor, Department of Gastroenterology and Hepatology, Chang Gung Memorial Hospital, Linkou Medical Center, No. 5 Fuxing Street, Guishan Dist, Taoyuan 33305, Taiwan. tai48978@cgmh.org.tw
Received: December 18, 2021
Peer-review started: December 18, 2021
First decision: January 23, 2022
Revised: February 3, 2022
Accepted: April 22, 2022
Article in press: April 22, 2022
Published online: June 14, 2022
Processing time: 174 Days and 2.5 Hours
ARTICLE HIGHLIGHTS
Research background

Two-dimensional (2D) ultrasound has been used for screening of liver steatosis for more than 5 decades. It is a cheap and non-invasive study.

Research motivation

Two-dimensional ultrasound is a subjective diagnosis that is not suitable for quantitative study.

Research objectives

To produce an objective steatosis diagnostic algorithm by deep learning from big data of 2D ultrasound images.

Research methods

Using multi-view ultrasound big data from a retrospective cohort of patients, we trained a deep learning algorithm to diagnose steatosis stages from clinical ultrasound diagnoses. Performance was validated on two multi-scanner unblinded and blinded histology-proven cohorts with histopathology diagnoses and a subset with FibroScan diagnoses. We also quantified reliability across scanners and viewpoints. Results were evaluated using Bland-Altman and receiver operating characteristic analysis.

Research results

The deep learning algorithm demonstrated repeatable measurements with a moderate number of images and high agreement across three premium ultrasound scanners. High diagnostic performance was observed across all viewpoints. Areas under the curve of the receiver operating characteristic to classify mild, moderate, and severe steatosis grades were 0.85, 0.91, and 0.93, respectively. This algorithm outperformed or performed at least comparably to FibroScan control attenuation parameter on the unblinded or blinded histology-proven cohort.

Research conclusions

This algorithm could give an objective diagnosis of steatosis from prospectively or retrospectively collected 2D ultrasound images.

Research perspectives

The cutoff values for different grades of steatosis would need future studies in different scanners and fibrosis status.