A machine learning based approach to factors predicting correlation between Hb1Ac and diabetes in patients of malnourished communities
Keywords:
Type 1 Diabetes Mellitus, Diabetic Ketoacidosis (DKA), Machine Learning Models, Glycemic Control (HbA1c), Predictive Analytics in HealthcareAbstract
Type 1 diabetes mellitus (T1DM) is associated with acute complications such as diabetic ketoacidosis (DKA) and long-term glycemic dysregulation. This study aimed to develop and validate machine learning models to predict DKA episodes and glycemic control, defined as HbA1c >7%, using a large multi-center, bi-national database from the Diabetes Data Network (DDN). Nine machine learning algorithms, including Deep Learning (DL), Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR), were trained and validated on clinical and demographic features collected longitudinally from 10,868 individuals aged 2–21 years between January 2012 and May 2022. The DL model demonstrated the highest predictive performance for DKA, achieving an area under the curve (AUC) of 0.887, while SVM was most effective in predicting HbA1c >7% with an AUC of 0.884. Key predictors for DKA included age at diagnosis, diabetes duration, prior DKA events, BMI z-score, HbA1c, CGM use, insulin regimen, and center. For HbA1c prediction, baseline HbA1c and BMI emerged as dominant features. The results suggest that integrating machine learning models into clinical care could enable early identification of high-risk individuals, facilitating timely interventions and potentially reducing hospitalization rates and healthcare costs associated with T1DM complications. These models underscore the importance of personalized management strategies and highlight the feasibility of real-world application in diabetes clinics.
Downloads
Metrics
References
American Diabetes Association Professional Practice Committee, “14. Children and adolescents: standards of care in diabetes—2024,” Diabetes Care, vol. 47, pp. S258–S281, 2024.
G. E. Umpierrez et al., “Hyperglycemic crises in adults with diabetes: a consensus report,” Diabetes Care, vol. 47, pp. 1257–1275, 2024.
D. Tomic, M. E. Craig, D. J. Magliano, and J. E. Shaw, “Reasons for hospitalisation in youth with type 1 diabetes, 2010–2019,” Diabetic Medicine, vol. 40, no. 1, p. e15218, 2023.
I. Contreras and J. Vehi, “Artificial intelligence for diabetes management and decision support: literature review,” Journal of Medical Internet Research, vol. 20, p. e10775, 2018.
S. B. Nagaraj, G. Sidorenkov, J. F. M. van Boven, and P. Denig, “Predicting short- and long-term glycated haemoglobin response after insulin initiation in patients with type 2 diabetes mellitus using machine-learning algorithms,” Diabetes, Obesity and Metabolism, vol. 21, no. 12, pp. 2704–2711, 2019.
Y. Fan et al., “Development of machine learning models for mortality risk prediction after cardiac surgery,” Cardiovascular Diagnosis and Therapy, vol. 12, no. 1, pp. 12–23, 2022.
J. I. Wolfsdorf et al., “ISPAD clinical practice consensus guidelines 2018: diabetic ketoacidosis and the hyperglycemic hyperosmolar state,” Pediatric Diabetes, vol. 19, Suppl. 27, pp. 155–177, 2018.
C. L. Ogden et al., “Centers for disease control and prevention 2000 growth charts for the United States: improvements to the 1977 national center for health statistics version,” Pediatrics, vol. 109, no. 1, pp. 45–60, 2002.
I. Mierswa, Data Science, Machine Learning, Predictive Analytics: RapidMiner Studio, Version 10.1, 2019. [Online]. Available: https://rapidminer.com
R. E. Fan et al., “LIBLINEAR: a library for large linear classification,” Journal of Machine Learning Research, vol. 9, pp. 1871–1874, 2008.
A. Earnest, G. A. Tesema, and R. G. Stirling, “Machine learning techniques to predict timeliness of care among lung cancer patients,” Healthcare (Basel), vol. 11, no. 2, 2023.
A. Garavand et al., “Efficient model for coronary artery disease diagnosis: a comparative study of several machine learning algorithms,” Journal of Healthcare Engineering, vol. 2022, Article ID 5359540, 2022.
E. M. El Malahi et al., “Relationship between time in range, glycemic variability, HbA1c and complications in adults with type 1 diabetes mellitus,” Journal of Clinical Endocrinology and Metabolism, vol. 107, no. 2, 2021.
S. L. C. Lebech and C. Bender, “Development of machine learning models for the identification of elevated ketone bodies during hyperglycemia in patients with type 1 diabetes,” Diabetes Technology & Therapeutics, vol. 26, no. 6, pp. 403–410, 2024.
L. Li et al., “Performance assessment of different machine learning approaches in predicting diabetic ketoacidosis in adults with type 1 diabetes using electronic health records data,” Pharmacoepidemiology and Drug Safety, vol. 30, no. 5, pp. 610–618, 2021.
D. D. Williams et al., “An 'All-Data-on-Hand' deep learning model to predict hospitalization for diabetic ketoacidosis in youth with type 1 diabetes: development and validation study,” JMIR Diabetes, vol. 8, p. e47592, 2023.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
