Harnessing serum VOCs and machine learning for the early detection of MAFLD

Aug 5, 2026

Metabolic dysfunction-associated fatty liver disease (MAFLD) remains difficult to diagnose at early stages, as ultrasound lacks sufficient sensitivity. This study evaluates serum volatile organic compound (VOC) profiling using GC-IMS and machine learning in a preliminary single-center cohort of MAFLD patients and healthy controls. Seven discriminative VOCs support model performance, with random forest yielding a test AUC of 0.941, enabling non-invasive distinction and potential stage stratification via 2-pentylfuran, pending larger multi-center validation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Methods

Results

Discussion

TYPE Original Research PUBLISHED 18 November 2025 DOI 10.3389/fendo.2025.1691853 OPEN ACCESS EDITED BY Tong Wang, University of Connecticut, United States REVIEWED BY Yuan Gao, Genentech Inc., United States He Gao, University of California, San Diego, United States *CORRESPONDENCE Xuewei Zhuang [zhuangxuewei@sdu.edu.cn](zhuangxuewei@sdu.edu.cn) RECEIVED 24 August 2025 ACCEPTED 29 October 2025 PUBLISHED 18 November 2025 CITATION Li X, Zhao X, Zhang R and Zhuang X (2025) Harnessing serum VOCs and machine learning for the early detection of MAFLD. Front. Endocrinol. 16:1691853 . doi: 10.3389/fendo.2025.1691853 COPYRIGHT © 2025 Li, Zhao, Zhang and Zhuang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Harnessing serum VOCs and machine learning for the early detection of MAFLD Xin Li 1, Xiaoyue Zhao 1, Ruonan Zhang 1 and Xuewei Zhuang 2* 1Shandong University of Traditional Chinese Medicine, Jinan, China, 2Shandong University Afﬁliated Shandong Provincial Third Hospital Department of Clinical Laboratory, Jinan, China Introduction: Metabolic dysfunction-associated fatty liver disease (MAFLD) is a complex metabolic disorder and one of the leading causes of chronic liver disease worldwide. Current diagnostic tools, such as ultrasound, lack sufﬁcient sensitivity for detecting early-stage disease, emphasizing the urgent need for novel and non-invasive diagnostic strategies. Metabolomics, particularly the proﬁling of volatile organic compounds (VOCs) in bioﬂuids, has emerged as a promising approach for biomarker discovery in metabolic diseases. Methods: In this preliminary single-center study, serum samples were collected from 199 participants, including 110 MAFLD patients and 89 healthy controls. Volatile organic compounds were analyzed using gas chromatography–ion mobility spectrometry (GC-IMS) . Machine learning algorithms, including random forest, were applied to construct diagnostic models and identify key discriminatory metabolites. Clinical and biochemical parameters such as age, body mass index, liver function, and lipid proﬁles were also compared between groups. Results: A total of 79 serum VOCs were detected, among which 54 showed signiﬁcant differences between MAFLD patients and controls (29 identiﬁed and 25 unidentiﬁed) . The random forest model exhibited the best diagnostic performance, achieving a test AUC of 0.941, with 86.7% sensitivity and 88.5% speciﬁcity. Seven key VOCs were identiﬁed as important contributors to the model, including two upregulated compounds (2-Butoxyethanol and Cyclopentanone-D) and ﬁve downregulated compounds ((E)-3-hexenoic acid, 2-Ethylbutanal, 2-Propyl acetate, Benzaldehyde-M, and Furaneol) . Notably, 2-pentylfuran displayed signiﬁcant variation across different pathological grades of MAFLD, suggesting potential as a stage-speciﬁc biomarker. Discussion: This study demonstrates that serum VOC proﬁling using GC-IMS combined with machine learning can effectively distinguish MAFLD patients from healthy individuals. The identiﬁed VOC signatures, particularly 2-pentylfuran, may serve as non-invasive biomarkers for MAFLD diagnosis and staging. However, due to the limited sample size and single-center design, these ﬁndings require validation in larger, multi-center, and longitudinal studies to conﬁrm their clinical applicability, especially for early disease detection. KEYWORDS metabolic dysfunction-associated fatty liver disease, volatile organic compounds, gas chromatography–ion mobility spectrometry, machine learning, biomarker, early diagnosis Fronti

FAQs

Why is early detection of MAFLD challenging?

Early-stage MAFLD is often asymptomatic, and ultrasound may not detect disease reliably due to limited sensitivity.

How were serum VOCs measured in this study?

Serum VOCs were analyzed using gas chromatography–ion mobility spectrometry (GC-IMS).

Which model and biomarkers showed the best diagnostic value?

The random forest model achieved a test AUC of 0.941, identifying seven key VOCs; 2-pentylfuran varied across pathological grades and may act as a stage-specific biomarker.

Aug 5, 2026

Upload by Genevieve