The University of Southampton
University of Southampton Institutional Repository

LEARN algorithm: a novel option for predicting non-alcoholic steatohepatitis

LEARN algorithm: a novel option for predicting non-alcoholic steatohepatitis
LEARN algorithm: a novel option for predicting non-alcoholic steatohepatitis

BACKGROUND: There is an unmet need for accurate non-invasive methods to diagnose non-alcoholic steatohepatitis (NASH). Since impedance-based measurements of body composition are simple, repeatable and have a strong association with non-alcoholic fatty liver disease (NAFLD) severity, we aimed to develop a novel and fully automatic machine learning algorithm, consisting of a deep neural network based on impedance-based measurements of body composition to identify NASH [the bioeLectrical impEdance Analysis foR Nash (LEARN) algorithm].

METHODS: A total of 1,259 consecutive subjects with suspected NAFLD were screened from six medical centers across China, of which 766 patients with biopsy-proven NAFLD were included in final analysis. These patients were randomly subdivided into the training and validation groups, in a ratio of 4:1. The LEARN algorithm was developed in the training group to identify NASH, and subsequently, tested in the validation group.

RESULTS: The LEARN algorithm utilizing impedance-based measurements of body composition along with age, sex, pre-existing hypertension and diabetes, was able to predict the likelihood of having NASH. This algorithm showed good discriminatory ability for identifying NASH in both the training and validation groups [area under the receiver operating characteristics (AUROC): 0.81, 95% CI: 0.77-0.84 and AUROC: 0.80, 95% CI: 0.73-0.87, respectively]. This algorithm also performed better than serum cytokeratin-18 neoepitope M30 (CK-18 M30) level or other non-invasive NASH scores (including HAIR, ION, NICE) for identifying NASH (P value <0.001). Additionally, the LEARN algorithm performed well in identifying NASH in different patient subgroups, as well as in subjects with partial missing body composition data.

CONCLUSIONS: The LEARN algorithm, utilizing simple easily obtained measures, provides a fully automated, simple, non-invasive method for identifying NASH.

2304-3881
507-522
Li, Gang
9a8d00e0-e85a-46ff-8833-eea48ca259e7
Zheng, Tian-Lei
337471b8-5643-4e11-8c9a-10390b6f5bb3
Chi, Xiao-Ling
d05cf6ff-9b3a-4948-8359-41c04a76d644
Zhu, Yong-Fen
eaad2b81-6c97-4cbb-adfc-d05691624001
Chen, Jin-Jun
09c5fa8b-f049-414e-8a3e-e73f976a194a
Xu, Liang
d489ded9-eec9-43fe-8fa8-576b62f75026
Shi, Jun-Ping
08a12e82-1887-42aa-836e-fc7077329313
Wang, Xiao-Dong
5ee24e5d-8cd7-42c1-b66a-cb39c7a7d7e4
Zhao, Wei-Guo
bb0216ed-4848-4d50-8089-2e759de01e71
Byrne, Christopher
1370b997-cead-4229-83a7-53301ed2a43c
Targher, Giovanni
11a8b68e-f6f5-4306-a81b-1c932aa3a09a
Rios, Rafael S.
c2287b9c-7293-4785-9e07-dbe326c09ef1
Huang, Ou-Yang
b8cf4537-47a4-45e2-a191-cfe5140a6018
Tang, Liang-Jie
e93cfc5a-9506-40d3-a6fc-03455087fe6d
Zhang, Shi-Jin
355a7ca0-6a81-46c5-9193-e0ccd22812d0
Geng, Shi
564fbd99-651c-4c2b-97ad-b69571b2606a
Xiao, Huan-Ming
0a92a418-1636-4d31-a099-de13273ffd04
Chen, Sui-Dan
3e06e46d-ebcf-4d32-8854-129e16753d10
Zhang, Rui
264a4633-bcfe-45db-81e1-3d9099cd7bcc
Zheng, Ming-Hua
a00a88ab-ccc8-422c-808f-7d2acc91e4e1
Li, Gang
9a8d00e0-e85a-46ff-8833-eea48ca259e7
Zheng, Tian-Lei
337471b8-5643-4e11-8c9a-10390b6f5bb3
Chi, Xiao-Ling
d05cf6ff-9b3a-4948-8359-41c04a76d644
Zhu, Yong-Fen
eaad2b81-6c97-4cbb-adfc-d05691624001
Chen, Jin-Jun
09c5fa8b-f049-414e-8a3e-e73f976a194a
Xu, Liang
d489ded9-eec9-43fe-8fa8-576b62f75026
Shi, Jun-Ping
08a12e82-1887-42aa-836e-fc7077329313
Wang, Xiao-Dong
5ee24e5d-8cd7-42c1-b66a-cb39c7a7d7e4
Zhao, Wei-Guo
bb0216ed-4848-4d50-8089-2e759de01e71
Byrne, Christopher
1370b997-cead-4229-83a7-53301ed2a43c
Targher, Giovanni
11a8b68e-f6f5-4306-a81b-1c932aa3a09a
Rios, Rafael S.
c2287b9c-7293-4785-9e07-dbe326c09ef1
Huang, Ou-Yang
b8cf4537-47a4-45e2-a191-cfe5140a6018
Tang, Liang-Jie
e93cfc5a-9506-40d3-a6fc-03455087fe6d
Zhang, Shi-Jin
355a7ca0-6a81-46c5-9193-e0ccd22812d0
Geng, Shi
564fbd99-651c-4c2b-97ad-b69571b2606a
Xiao, Huan-Ming
0a92a418-1636-4d31-a099-de13273ffd04
Chen, Sui-Dan
3e06e46d-ebcf-4d32-8854-129e16753d10
Zhang, Rui
264a4633-bcfe-45db-81e1-3d9099cd7bcc
Zheng, Ming-Hua
a00a88ab-ccc8-422c-808f-7d2acc91e4e1

Li, Gang, Zheng, Tian-Lei, Chi, Xiao-Ling, Zhu, Yong-Fen, Chen, Jin-Jun, Xu, Liang, Shi, Jun-Ping, Wang, Xiao-Dong, Zhao, Wei-Guo, Byrne, Christopher, Targher, Giovanni, Rios, Rafael S., Huang, Ou-Yang, Tang, Liang-Jie, Zhang, Shi-Jin, Geng, Shi, Xiao, Huan-Ming, Chen, Sui-Dan, Zhang, Rui and Zheng, Ming-Hua (2023) LEARN algorithm: a novel option for predicting non-alcoholic steatohepatitis. Hepatobiliary Surgery and Nutrition, 12 (4), 507-522. (doi:10.21037/hbsn-21-523).

Record type: Article

Abstract

BACKGROUND: There is an unmet need for accurate non-invasive methods to diagnose non-alcoholic steatohepatitis (NASH). Since impedance-based measurements of body composition are simple, repeatable and have a strong association with non-alcoholic fatty liver disease (NAFLD) severity, we aimed to develop a novel and fully automatic machine learning algorithm, consisting of a deep neural network based on impedance-based measurements of body composition to identify NASH [the bioeLectrical impEdance Analysis foR Nash (LEARN) algorithm].

METHODS: A total of 1,259 consecutive subjects with suspected NAFLD were screened from six medical centers across China, of which 766 patients with biopsy-proven NAFLD were included in final analysis. These patients were randomly subdivided into the training and validation groups, in a ratio of 4:1. The LEARN algorithm was developed in the training group to identify NASH, and subsequently, tested in the validation group.

RESULTS: The LEARN algorithm utilizing impedance-based measurements of body composition along with age, sex, pre-existing hypertension and diabetes, was able to predict the likelihood of having NASH. This algorithm showed good discriminatory ability for identifying NASH in both the training and validation groups [area under the receiver operating characteristics (AUROC): 0.81, 95% CI: 0.77-0.84 and AUROC: 0.80, 95% CI: 0.73-0.87, respectively]. This algorithm also performed better than serum cytokeratin-18 neoepitope M30 (CK-18 M30) level or other non-invasive NASH scores (including HAIR, ION, NICE) for identifying NASH (P value <0.001). Additionally, the LEARN algorithm performed well in identifying NASH in different patient subgroups, as well as in subjects with partial missing body composition data.

CONCLUSIONS: The LEARN algorithm, utilizing simple easily obtained measures, provides a fully automated, simple, non-invasive method for identifying NASH.

Text
manuscript_HBSN-21-523-R2 - Accepted Manuscript
Download (568kB)
Text
Supplementary table 1_R1 - Accepted Manuscript
Download (381kB)
Text
Supplementary table 2_R1 - Accepted Manuscript
Download (360kB)
Text
Supplementary table 3_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
TABLE1_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
TABLE2_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
TABLE3_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
TABLE4_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Image
FIG1_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Image
FIG2_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Image
FIG3_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Image
FIG4_R1 - Accepted Manuscript
Restricted to Repository staff only
Request a copy
Text
112234-PB9-6419-R3 - Version of Record
Download (1MB)

Show all 13 downloads.

More information

Accepted/In Press date: 9 May 2022
e-pub ahead of print date: 30 March 2023
Published date: 1 August 2023
Additional Information: 2023 Hepatobiliary Surgery and Nutrition. All rights reserved.

Identifiers

Local EPrints ID: 457355
URI: http://eprints.soton.ac.uk/id/eprint/457355
ISSN: 2304-3881
PURE UUID: c458c6bb-f8b9-4f15-a725-ee38636ece78
ORCID for Christopher Byrne: ORCID iD orcid.org/0000-0001-6322-7753

Catalogue record

Date deposited: 01 Jun 2022 16:45
Last modified: 17 Mar 2024 02:49

Export record

Altmetrics

Contributors

Author: Gang Li
Author: Tian-Lei Zheng
Author: Xiao-Ling Chi
Author: Yong-Fen Zhu
Author: Jin-Jun Chen
Author: Liang Xu
Author: Jun-Ping Shi
Author: Xiao-Dong Wang
Author: Wei-Guo Zhao
Author: Giovanni Targher
Author: Rafael S. Rios
Author: Ou-Yang Huang
Author: Liang-Jie Tang
Author: Shi-Jin Zhang
Author: Shi Geng
Author: Huan-Ming Xiao
Author: Sui-Dan Chen
Author: Rui Zhang
Author: Ming-Hua Zheng

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×