Citation :

C. Iyyappan, R. Latha, "Machine Learning-Driven Prediction of Liver Enzyme Abnormalities in Type 2 Diabetes Mellitus," International Journal of Electronics and Communication Engineering, vol. 13, no. 4, pp. 168-182, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I4P113

Abstract

Liver complications are a concern in people living with Type 2 Diabetes Mellitus (T2DM). But they remain unnoticed, and enzyme levels increase beyond normal ranges. Increases in alanine aminotransferase and aspartate aminotransferase are used as early biochemical signals of hepatic stress. Early assessment is important in individuals with T2DM. This study presents a data-based framework to identify variations in serum transaminase levels using regularly collected demographic and metabolic information (age, sex, body mass index, HbA1c, lipid profile measures, blood pressure, liver fat content, and diabetes status). Graph Attention Networks, combined with the minimum Redundancy Maximum Relevance method, were applied for feature selection. It retains influential variables and minimizes redundancy. ALT and AST levels were estimated using Bayesian-optimized XGBoost and a deep neural network trained with the Adam optimizer. The models achieved an accuracy of 95.3% and a precision of 93.5%. The findings indicate that integrating routine clinical data with analytical models assists in early identification of hepatic abnormalities in individuals with T2DM.

Keywords

Type 2 Diabetes Mellitus (T2DM), Liver enzyme prediction, Alanine Aminotransferase (ALT), Aspartate Aminotransferase (AST), Graph Attention Networks (GAT), Machine learning in healthcare, Deep Neural Networks (DNNs).

References

1. M. Ammar Kalas et al., “Abnormal Liver Enzymes: A Review for Clinicians,” World Journal of Hepatology, vol. 13, no. 11, pp. 1-12, 2021.
   [CrossRef] [Google Scholar] [Publisher Link]
2. Elena Raya-Cano et al., “Metabolic Syndrome and Transaminases: Systematic Review and Meta-Analysis,” Diabetology & Metabolic Syndrome, vol. 15, pp. 1-12, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
3. Panpan He et al., “A Healthy Lifestyle, Life's Essential 8 Scores and New-Onset Severe NAFLD: A Prospective Analysis in UK Biobank,” Metabolism, vol. 146, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
4. Hina Wazir et al., “Diagnosis and Treatment of Liver Disease: Current Trends and Future Directions,” Cureus, vol. 15, no. 12, pp. 1-12, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
5. David Nam et al., “Artificial Intelligence in Liver Diseases: Improving Diagnostics, Prognostics and Response Prediction,” JHEP Reports, vol. 4, no. 4, pp. 1-11, 2022.
   [CrossRef] [Google Scholar] [Publisher Link]
6. Aditya Chilay et al., “Liver Function Test and Diabetes Mellitus: Correlation from a Laboratory Perspective,” Indian Journal of Medical Biochemistry, vol. 27, no. 2, pp. 40-44, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
7. Christina Antza et al., “The Links between Sleep Duration, Obesity and Type 2 Diabetes Mellitus,” Journal of Endocrinology, vol. 252, no. 2, pp. 125-141, 2022.
   [CrossRef] [Google Scholar] [Publisher Link]
8. Zobair M. Younossi et al., “Lifestyle Interventions in Nonalcoholic Fatty Liver Disease,” Nature Reviews Gastroenterology & Hepatology, vol. 20, pp. 708-722, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
9. Fredrik Åberg et al., “Alcohol Consumption and Metabolic Syndrome: Clinical and Epidemiological Impact on Liver Disease,” Journal of Hepatology, vol. 78, no. 1, pp. 191-206, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
10. Boonsub Sakboonyarat et al., “Association between Raised Blood Pressure and Elevated Serum Liver Enzymes among Active-Duty Royal Thai Army Personnel in Thailand,” BMC Cardiovascular Disorders, vol. 23, pp. 1-11, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
11. Ruhul Amin et al., “Prediction of Chronic Liver Disease Patients using Integrated Projection-Based Statistical Feature Extraction with Machine Learning Algorithms,” Informatics in Medicine Unlocked, vol. 36, pp. 1-11, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
12. S. Murugesan et al., “Feature Selection and Classification of Clinical Datasets using Bioinspired Algorithms and Super Learner,” Computational and Mathematical Methods in Medicine, vol. 2021, no. 1, pp. 1-18, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
13. Hua Niu, and Yinghua Zhou, “Nonlinear Relationship between AST-to-ALT Ratio and the Incidence of Type 2 Diabetes Mellitus: A Follow-Up Study,” International Journal of General Medicine, vol. 14, pp. 8373-8382, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
14. Changchun Cao et al., “Nonlinear Relationship between Aspartate Aminotransferase to Alanine Aminotransferase Ratio and the Risk of Prediabetes: A Retrospective Study based on Chinese Adults,” Frontiers in Endocrinology, vol. 13, pp. 1-13, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
15. Kebao Zhang et al., “Association between the Aspartate Aminotransferase-to-Alanine Aminotransferase Ratio and the Reversion to Normoglycemia in People with Impaired Fasting Glucose: A 5-year Retrospective Cohort Study,” Frontiers in Endocrinology, vol. 16, pp. 1-16, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
16. R. Kalaiselvi, K. Meena, and V. Vanitha, “Liver Disease Prediction Using Machine Learning Algorithms,” 2021 International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA), Coimbatore, India, pp. 1-6, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
17. Srilatha Tokala et al., “Liver Disease Prediction and Classification using Machine Learning Techniques,” International Journal of Advanced Computer Science and Applications, vol. 14, no. 2, pp. 1-8, 2023.
18. [CrossRef] [Google Scholar] [Publisher Link]
19. Wenquan Cheng et al., “A Dataset on 24-h Electrocardiograph, Sleep and Metabolic Function of Male Type 2 Diabetes Mellitus,” Data in Brief, vol. 49, pp. 1-7, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Katarzyna Nabrdalik et al., “Machine Learning Identifies Metabolic Dysfunction–Associated Steatotic Liver Disease in Patients with Diabetes Mellitus,” The Journal of Clinical Endocrinology & Metabolism, vol. 109, no. 8, pp. 2029-2038, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
21. Mojgan Noroozi Karimabad et al., “Serum Liver Enzymes and Diabetes from the Rafsanjan Cohort Study,” BMC Endocrine Disorders, vol. 22, pp. 1-12, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
22. Manila Makaju et al., “Pattern of Serum Liver Enzymes in the Patients with Type 2 Diabetes Mellitus,” Journal of Research in Applied and Basic Medical Sciences, vol. 9, no. 4, pp. 272-279, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
23. Zhao Li, Wenzhong Wu, and Hyunsik Kang, “Machine Learning-Driven Metabolic Syndrome Prediction: An International Cohort Validation Study,” Healthcare, vol. 12, no. 24, pp. 1-22, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
24. Pima Indians Diabetes Database, Kaggle. [Online]. Available: https://www.kaggle.com/datasets/uciml/pima-indians-diabetes-database
25. Ahlam Rashid, “Diabetes Dataset,” Mendeley Data, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
26. QinpeiZhao et al., “Chinese Diabetes Datasets for Data-Driven Machine Learning,” Scientific Data, vol. 10, pp. 1-8, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
27. Diabetes Health Indicators Dataset, Kaggle. [Online]. Available: https://www.kaggle.com/datasets/alexteboul/diabetes-health-indicators-datase