TY - JOUR
T1 - Penalty Weighted Glucose Prediction Models Could Lead to Better Clinically Usage
AU - Cichosz, Simon Lebech
AU - Kronborg, Thomas
AU - Jensen, Morten Hasselstrøm
AU - Hejlesen, Ole
PY - 2021/11
Y1 - 2021/11
N2 - Background and objective: Numerous attempts to predict glucose value from continuous glucose monitors (CGM) have been published. However, there is a lack of proper analysis and modeling of penalty for errors in different glycemic ranges. The aim of this study was to investigate the potential for developing glucose prediction models with focus on the clinical aspects. Methods: We developed and compared six different models to test which approach were best suited for predicting glucose levels at different lead times between 10 and 60 min. The models were: last observation carried forward, linear extrapolation, ensemble methods using LSBoost and bagging, neural networks, one without error-weights and one with error-weights. The modeling and test were based on 225 type 1 diabetes patients with 315,000 h of CGM data. Results: Results show that it is possible to predict glucose levels based on CGM with a reasonable accuracy and precision with a 30-min prediction lead time. A comparison of different methods shows that there are improvements on performance gained from using more advanced machine learning algorithms (MARD 10.26–10.79 @ 30-min lead time) compared to a simple modeling (MARD 10.75–12.97 @ 30-min lead time). Moreover, the proposed use of error weights could lead to better clinical performance of these models, which is an important factor for real usage. E.g., the percentages in the C-zone of the consensus error grid without error-weights (0.57-0.68%) vs including error-weights (0.28%). Conclusions: The results point toward that using error weighting in the training of the models could lead to better clinical performance.
AB - Background and objective: Numerous attempts to predict glucose value from continuous glucose monitors (CGM) have been published. However, there is a lack of proper analysis and modeling of penalty for errors in different glycemic ranges. The aim of this study was to investigate the potential for developing glucose prediction models with focus on the clinical aspects. Methods: We developed and compared six different models to test which approach were best suited for predicting glucose levels at different lead times between 10 and 60 min. The models were: last observation carried forward, linear extrapolation, ensemble methods using LSBoost and bagging, neural networks, one without error-weights and one with error-weights. The modeling and test were based on 225 type 1 diabetes patients with 315,000 h of CGM data. Results: Results show that it is possible to predict glucose levels based on CGM with a reasonable accuracy and precision with a 30-min prediction lead time. A comparison of different methods shows that there are improvements on performance gained from using more advanced machine learning algorithms (MARD 10.26–10.79 @ 30-min lead time) compared to a simple modeling (MARD 10.75–12.97 @ 30-min lead time). Moreover, the proposed use of error weights could lead to better clinical performance of these models, which is an important factor for real usage. E.g., the percentages in the C-zone of the consensus error grid without error-weights (0.57-0.68%) vs including error-weights (0.28%). Conclusions: The results point toward that using error weighting in the training of the models could lead to better clinical performance.
KW - Neural network
KW - Ensemble learning
KW - Prediction
KW - CGM
KW - Type 1 diabetes
KW - Glucose
KW - Continuous glucose monitoring
U2 - 10.1016/j.compbiomed.2021.104865
DO - 10.1016/j.compbiomed.2021.104865
M3 - Journal article
SN - 0010-4825
VL - 138
JO - Computers in Biology and Medicine
JF - Computers in Biology and Medicine
M1 - 104865
ER -