Feature-Based Machine Learning Model for Real-Time Hypoglycemia Prediction

Abstract

Background: Hypoglycemia is a serious health concern in youth with type 1 diabetes (T1D). Real-time data from continuous glucose monitoring (CGM) can be used to predict hypoglycemic risk, allowing patients to take timely intervention measures. Methods: A machine learning model is developed for probabilistic prediction of hypoglycemia (<70 mg/dL) in 30- and 60-minute time horizons based on CGM datasets obtained from 112 patients over a range of 90 days consisting of over 1.6 million CGM values under normal living conditions. A comprehensive set of features relevant for hypoglycemia are developed and a parsimonious subset with most influence on predicting hypoglycemic risk is identified. Model performance is evaluated both with and without contextual information on insulin and carbohydrate intake. Results: The model predicted hypoglycemia with >91% sensitivity for 30- and 60-minute prediction horizons while maintaining specificity >90%. Inclusion of insulin and carbohydrate data yielded performance improvement for 60-minute but not for 30-minute predictions. Model performance was highest for nocturnal hypoglycemia (~95% sensitivity). Shortterm (less than one hour) and medium-term (one to four hours) features for good prediction performance are identified. Conclusions: Innovative feature identification facilitated high performance for hypoglycemia risk prediction in pediatric youth with T1D. Timely alerts of impending hypoglycemia may enable proactive measures to avoid severe hypoglycemia and achieve optimal glycemic control. The model will be deployed on a patient-facing smartphone application in an upcoming pilot study.

Machine Learning Techniques for Hypoglycemia Prediction: Trends and Challenges

“…Other works such as Dave et al [ 52 ], Shifrin et al [ 56 ], and Cappon et al [ 66 ] used BG along with insulin and CHO for prediction purposes. Aiello et al [ 67 ] and Oviedo et al [ 53 ] both aimed at postprandial hypoglycemia prediction by utilizing BG data combined with insulin and CHO data.…”

“…This approach fixes the over-fitting problem of decision trees. Seo et al [ 43 ], Güemes et al [ 60 ], Vahedi et al [ 33 ], G Noaro et al [ 72 ], Vu et al [ 47 ], Reddy et al [ 40 ], Chen et al [ 30 ], Dave et al [ 52 ], Calhoun et al [ 45 ], Amar et al [ 75 ], Hidalgo et al [ 77 ], and Rodriguez et al [ 79 ] have all used RF for predicting/detecting hypoglycemia. Ruan et al [ 31 ] and Cappon et al [ 66 ] used the XGboost algorithm.…”

“…The various types of regression used by studies in this review are LR, GE, and MLR. Studies proposed by Chen et al [ 30 ], Dave et al [ 52 ], and Seo et al [ 43 ] use LR. LR fits a logistic function to data and outputs the probability of one or more classes.…”

Ability of Current Machine Learning Algorithms to Predict and Detect Hypoglycemia in Patients With Diabetes Mellitus: Meta-analysis

“…Table 1 shows the summary of study characteristics. Of the 33 studies, 19 studies (58%) [26][27][28][29][30][31]33,35,36,[38][39][40][41][42][44][45][46][47]54] predicted hypoglycemia, and the remaining 14 studies (42%) detected hypoglycemia [15,20,25,32,34,37,43,[48][49][50][51][52][53]55]. As much as 25 of the 33 included studies (76%) [15,20,[25][26][27]29,30,32,35,36,38,39,[41][42][43][44]…”

“…Regarding the time of day when hypoglycemic events occurred, nocturnal hypoglycemia was the most frequently reported (14 studies of the 33 included studies; 42%) [15,20,26,30,32,35,36,41,44,[49][50][51][52][53]). As to the place of the supposed hypoglycemic episode, 16 of the 19 studies that predicted hypoglycemia (84%) [26][27][28][29][30]35,36,[38][39][40][41][42][44][45][46][47] supposed the event took place in an out-of-hospital setting. The remaining 3 studies (16%) [31,33,54] supposed hypoglycemia occurring in an in-hospital setting.…”