An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes
Ein interpretierbares Machine-Learning-Modell zur Vorhersage der metabolischen Dysfunktion-assoziierten steatotischen Lebererkrankung bei Patient:innen mit Typ-2-Diabetes
Diabetes, Obesity and Metabolism · 2026-01 · 28 (1) · 122-136
DOI: 10.1111/dom.70168 ↗Auf einen Blick
Bei 3.836 Patient:innen mit Typ-2-Diabetes sagte ein XGB-Modell das MASLD-Risiko gut voraus. Die beste Leistung zeigte das Modell mit AUROC 0,873 und AUPRC 0,904.
Fragestellung
Kann ein Machine-Learning-Modell das Risiko für MASLD bei Patient:innen mit Typ-2-Diabetes zuverlässig vorhersagen?
Methode
Klinische Daten von stationären T2DM-Patient:innen aus Nanjing Drum Tower Hospital; Entwicklung von 8 Machine-Learning-Methoden; Bewertung mit AUROC, AUPRC, Accuracy, Recall, NPV, PPV und F1-Score; Interpretation mit SHAP.
Ergebnisse
3.836 T2DM-Patient:innen; MASLD-Inzidenz 55,9 %. 13 Merkmale wurden ausgewählt. Das XGB-Modell erzielte die beste Gesamtleistung mit AUROC 0,873 und AUPRC 0,904.
Schlussfolgerung der Autor:innen
Das entwickelte Modell ermöglicht eine zuverlässige und interpretierbare MASLD-Risikovorhersage bei T2DM-Patient:innen auf Basis häufig verfügbarer klinischer Daten.
Kernaussagen
- 3.836 T2DM-Patient:innen mit 55,9 % MASLD
- XGB-Modell: AUROC 0,873; AUPRC 0,904
- SHAP und Clustering zeigten interpretierbare Risikogruppen
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Diabetes, Obesity and Metabolism, online erschienen am 29.09.2025. Verfasst von Zhuolin Zhou und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmacy, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China. Hauptthemen laut MEDLINE: Diabetes Mellitus, Type 2, Machine Learning, Fatty Liver. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Affiliated Drum Tower Hospital, Medical School of Nanjing University.
Themen: Künstliche Intelligenz in der Pflege, Diabetes mellitus
Abstract (Originalsprache)
Aim
Patients with type 2 diabetes mellitus (T2DM) exhibit an elevated prevalence of metabolic dysfunction‐associated steatotic liver disease (MASLD) and are at greater risk of liver‐related adverse events. Existing non‐invasive tools show limited diagnostic performance in this population. This study aims to develop a predictive model that accurately identifies the risk of MASLD among T2DM patients. Materials and methods Clinical data were collected from T2DM patients hospitalised at Nanjing Drum Tower Hospital between January 2018 and May 2025. Eight machine learning methods were developed to predict the risk of MASLD in T2DM patients. The discriminatory ability of the models was evaluated using area under the receiver operating characteristic curve (AUROC), area under the precision‐recall curve (AUPRC), accuracy, recall, negative predictive value, positive predictive value, and F1 score. Calibration curves and decision analysis curves were employed to evaluate the calibration and clinical utility. The models were interpreted using the Shapley additive explanations method, and unsupervised clustering was performed to identify potential high‐risk subgroups.
Results
A total of 3836 T2DM patients constituted the complete dataset, with a MASLD incidence rate of 55.9%. Thirteen feature variables were selected for model construction, and the XGB model achieved optimal overall performance, with an AUROC of 0.873 and an AUPRC of 0.904. Unsupervised clustering identified several high‐risk subgroups with distinct metabolic characteristics.
Conclusion
The model developed enables reliable and interpretable MASLD risk prediction in T2DM patients based on selected commonly available clinical data, providing a practical tool for routine identification and stratified management.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Pharmacy, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
- School of Basic Medicine and Clinical Pharmacy, China Pharmaceutical University, Nanjing, China.
- Department of Pharmacy, Nanjing Drum Tower Hospital, Dalian Medical University, Nanjing, China.
- Department of Endocrinology, Endocrine and Metabolic Disease Medical Centre, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
- Branch of National Clinical Research Centre for Metabolic Diseases, Nanjing, China.
ORCID-Profile: Zhuolin Zhou, Nan Gao, Jiaojiao Liu, Zhijuan Ge, Cheng Ji
Förderung
- Affiliated Drum Tower Hospital, Medical School of Nanjing University
Zitieren
APA
Zhou, Z., Gao, N., Liu, J., Ma, X., Ge, Z., & Ji, C. (2026). An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes. Diabetes, Obesity and Metabolism, 28(1), 122-136. https://doi.org/10.1111/dom.70168
Vancouver
Zhou Z, Gao N, Liu J, Ma X, Ge Z, Ji C. An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes. Diabetes, Obesity and Metabolism. 2026;28(1):122-136. doi:10.1111/dom.70168
BibTeX
@article{Zhou2026,
author = {Zhou, Zhuolin and Gao, Nan and Liu, Jiaojiao and Ma, Xuerong and Ge, Zhijuan and Ji, Cheng},
title = {An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes},
journal = {Diabetes, Obesity and Metabolism},
year = {2026},
volume = {28},
number = {1},
pages = {122--136},
doi = {10.1111/dom.70168}
}
EUVIMED