J Korean Diabetes.  2020 Sep;21(3):130-139. 10.4093/jkd.2020.21.3.130.

Machine Learning Application in Diabetes and Endocrine Disorders

• Hong N
• Park H
• Rhee Y

• KMID: 2507044
• DOI: http://doi.org/10.4093/jkd.2020.21.3.130

Abstract

Recently, machine learning (ML) applications have received attention in diabetes and metabolism research. This review briefly provides the basic concepts of ML and specific topics in diabetes research. Exemplary studies are reviewed to provide an overview of the methodology, main findings, limitations, and future research directions for ML-based studies. Well-defined, testable study hypotheses that stem from unmet clinical needs are always the first prerequisite for successful deployment of an MLbased approach to clinical scene. The management of data quality with enough quantity and active collaboration with ML engineers can enhance the ML development process. The interpretable highperformance ML models beyond the black-box nature of some ML principles can be one of the future goals expected by ML and artificial intelligence in the diabetes research and clinical practice settings that is beyond hype. Most importantly, endocrinologists should play a central role as domain experts who have clinical expertise and scientific rigor, for properly generating, refining, analyzing, and interpreting data by successfully integrating ML models into clinical research.

Keyword

Artificial intelligence; Deep learning; Diabetes mellitus; Machine learning; Metabolism

Figure

• 

  Fig. 1. A brief workflow of machine learning-based medical research.

Reference

1. McCarthy J. From here to human-level AI. Artifi Intell. 2007; 171:1174–82.

2. McCarthy J. What is artificial intelligence? Available from: http://www-formal.stanford.edu/jmc/whatisai.html (Accessed on 9th June, 2020).

3. Beaulieu-Jones B, Finlayson SG, Chivers C, Chen I, McDermott M, Kandola J, et al. Trends and focus of machine learning applications for health research. JAMA Netw Open. 2019; 2:e1914051.

4. Artzi NS, Shilo S, Hadar E, Rossman H, Barbash-Hazan S, Ben-Haroush A, et al. Prediction of gestational diabetes based on nationwide electronic health records. Nat Med. 2020; 26:71–6.

5. Oroojeni Mohammad Javad M, Agboola SO, Jethwani K, Zeid A, Kamarthi S. A reinforcement learning-based method for management of type 1 diabetes: exploratory study. JMIR Diabetes. 2019; 4:e12905.

6. Ballinger B, Hsieh J, Singh A, Sohoni N, Wang J, Tison GH, et al. DeepHeart: semi-super vised sequence learning for cardiovascular risk prediction. CoRR. 2018; arXiv:1802.02511.

7. Shomorony I, Cirulli ET, Huang L, Napier LA, Heister RR, Hicks M, et al. An unsupervised learning approach to identify novel signatures of health and disease from multimodal data. Genome Med. 2020; 12:7.

8. Dinga R, Penninx BWJH, Veltman DJ, Schmaal L, Marquand AF. Beyond accuracy: measures for assessing machine learning models, pitfalls and guidelines. bioRxiv. 2019; 743138.

9. Handelman GS, Kok HK, Chandra RV, Razavi AH, Huang S, Brooks M, et al. Peering into the black box of artificial intelligence: evaluation metrics of machine learning methods. AJR Am J Roentgenol. 2019; 212:38–43.

10. Saito T, Rehmsmeier M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS One. 2015; 10:e0118432.

11. Perakakis N, Polyzos SA, Yazdani A, Sala-Vila A, Kountouras J, Anastasilakis AD, et al. Non-invasive diagnosis of non-alcoholic steatohepatitis and fibrosis with the use of omics and supervised learning: a proof of concept study. Metabolism. 2019; 101:154005.

12. Segar MW, Vaduganathan M, Patel KV, McGuire DK, Butler J, Fonarow GC, et al. Machine learning to predict the risk of incident heart failure hospitalization among patients with diabetes: the WATCH-DM risk score. Diabetes Care. 2019; 42:2298–306.

13. Basu S, Raghavan S, Wexler DJ, Berkowitz SA. Characteristics associated with decreased or increased mortality risk from glycemic therapy among patients with type 2 diabetes and high cardiovascular risk: machine learning analysis of the ACCORD trial. Diabetes Care. 2018; 41:604–12.

14. Liu Y, Wang Y, Ni Y, Cheung CKY, Lam KSL, Wang Y, et al. Gut microbiome fermentation determines the efficacy of exercise for diabetes prevention. Cell Metab. 2020; 31:77–91.e5.

15. Katsiki N, Gastaldelli A, Mikhailidis DP. Predictive models with the use of omics and supervised machine learning to diagnose non-alcoholic fatty liver disease: a "non-invasive alternative" to liver biopsy? Metabolism. 2019; 101:154010.

16. Choi BG, Rha SW, Kim SW, Kang JH, Park JY, Noh YK. Machine learning for the prediction of new-onset diabetes mellitus during 5-year follow-up in non-diabetic patients with cardiovascular risks. Yonsei Med J. 2019; 60:191–9.

17. Seo W, Lee YB, Lee S, Jin SM, Park SM. A machine-learning approach to predict postprandial hypoglycemia. BMC Med Inform Decis Mak. 2019; 19:210.

18. Kim HS, Kim DJ, Yoon KH. Medical big data is not yet available: why we need realism rather than exaggeration. Endocrinol Metab (Seoul). 2019; 34:349–54.

19. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019; 25:44–56.