Artificial Intelligence (AI) based machine learning models predict glucose variability and hypoglycaemia risk in patients with type 2 diabetes on a multiple drug regimen who fast during ramadan (The PROFAST – IT Ramadan study). (November 2020)

• Record Type: Journal Article
• Authors: Elhadd, Tarik; Mall, Raghvendra; Bashir, Mohammed; Palotti, Joao; Fernandez-Luque, Luis; Farooq, Faisal; Mohanadi, Dabia Al; Dabbous, Zainab; Malik, Rayaz A.; Abou-Samra, Abdul Badi
• Is Part Of: Diabetes research and clinical practice. Volume 169(2020)
• Publication Date: 2020-11
• Subjects: Diabetes mellitus -- Ramadan -- Type-2 diabetes -- Artificial intelligence -- Flash glucose monitoring system -- Hypoglycaemia -- Hyperglycaemia; Diabetes -- Periodicals; Diabetes Mellitus -- Periodicals; 616.462
• DOI: 10.1016/j.diabres.2020.108388
• Languages: English
• ISSNs: 0168-8227

Abstract

Abstract: Objective: To develop a machine-based algorithm from clinical and demographic data, physical activity and glucose variability to predict hyperglycaemic and hypoglycaemic excursions in patients with type 2 diabetes on multiple glucose lowering therapies who fast during Ramadan. Patients and methods: Thirteen patients (10 males and three females) with type 2 diabetes on 3 or more anti-diabetic medications were studied with a Fitbit-2 pedometer device and Freestyle Libre (Abbott Diagnostics) 2 weeks before and 2 weeks during Ramadan. Several machine learning techniques were trained to predict blood glucose levels in a regression framework utilising physical activity and contemporaneous blood glucose levels, comparing Ramadan to non-Ramadan days. Results: The median age of participants was 51 years (IQR 49–52); median BMI was 33.2 kg/m 2 (IQR 33.0–35.9) and median HbA1c was 7.3% (IQR 6.7–7.8). The optimal model using physical activity achieved an R 2 of 0.548 and a mean absolute error (MAE) of 30.30. The addition of electronic health record (ehr) information increased R 2 to 0.636 and reduced MAE to 26.89 and the time of the day feature further increased R 2 to 0.768 and reduced MAE to 20.55. Combining all the features together resulted in an optimal XGBoost model with an R 2 of 0.836 and MAE of 17.47. This model accurately estimated normal glucose levels in 2584/2715 (95.2%) readings and hyperglycaemic events in 852/1031 (82.6%) readings, but fewer hypoglycaemic events (48/172 (27.9%)). The optimal XGBoost model prioritized age, gender, BMI and HbA1c followed by glucose levels and physical activity. Interestingly, the blood glucose level prediction by our model was influenced by use of SGLT2i. Conclusion: XGBoost, a machine learning AI algorithm achieves high predictive performance for normal and hyperglycaemic excursions, but has limited predictive value for hypoglycaemia in patients on multiple therapies who fast during Ramadan.