TY - GEN
T1 - Forewarning Postprandial Hyperglycemia with Interpretations using Machine Learning
AU - Arefeen, Asiful
AU - Fessler, Samantha
AU - Johnston, Carol
AU - Ghasemzadeh, Hassan
N1 - Funding Information:
We recognize that hyperglycemia depends on many other factors such as physical activity choices, poor disease management, non-diabetes medications, or skipping glucose-lowering medication [13]. To the best of our knowledge, however, current publicly available datasets do not contain such a comprehensive set of variables. For this research, our analysis was limited to designing machine learning models based on baseline glucose level, diet, and insulin amount only. Our on-going and future work involves construction of comprehensive dataset that integrates different modalities from a large cohort of individuals as well as the development of interpretable machine learning models for PPHG forecasting and blood glucose management. V. ACKNOWLEDGEMENT This work was supported in part by the National Science Foundation, under grants CNS-2210133, CNS-2227002, IIS-1954372, and IIS-1852163. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding organizations.
Funding Information:
This work was supported in part by the National Science Foundation, under grants CNS-2210133, CNS-2227002, IIS-1954372, and IIS-1852163. Authors are with the College of Health Solutions, Arizona State University, Phoenix, AZ 85004, USA email: aarefeen@asu.edu 1Resources available at: https://github.com/Arefeen06088/DietNudge
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Postprandial hyperglycemia (PPHG) is detrimental to health and increases risk of cardiovascular diseases, reduced eyesight, and life-threatening conditions like cancer. Detecting PPHG events before they occur can potentially help with providing early interventions. Prior research suggests that PPHG events can be predicted based on information about diet. However, such computational approaches (1) are data hungry requiring significant amounts of data for algorithm training; and (2) work as a black-box and lack interpretability, thus limiting the adoption of these technologies for use in clinical interventions. Motivated by these shortcomings, we propose, DietNudge1, a machine learning based framework that integrates multi-modal data about diet, insulin, and blood glucose to predict PPHG events before they occur. Using data from patients with diabetes, we demonstrate that our model can predict PPHG events with up to 90% classification accuracy and an average F1 score of 0.93. The proposed decision-tree-based approach also identifies modifiable factors that contribute to an impending PPHG event while providing personalized thresholds to prevent such events. Our results suggest that we can develop simple, yet effective, computational algorithms that can be used as preventative mechanisms for diabetes and obesity management.
AB - Postprandial hyperglycemia (PPHG) is detrimental to health and increases risk of cardiovascular diseases, reduced eyesight, and life-threatening conditions like cancer. Detecting PPHG events before they occur can potentially help with providing early interventions. Prior research suggests that PPHG events can be predicted based on information about diet. However, such computational approaches (1) are data hungry requiring significant amounts of data for algorithm training; and (2) work as a black-box and lack interpretability, thus limiting the adoption of these technologies for use in clinical interventions. Motivated by these shortcomings, we propose, DietNudge1, a machine learning based framework that integrates multi-modal data about diet, insulin, and blood glucose to predict PPHG events before they occur. Using data from patients with diabetes, we demonstrate that our model can predict PPHG events with up to 90% classification accuracy and an average F1 score of 0.93. The proposed decision-tree-based approach also identifies modifiable factors that contribute to an impending PPHG event while providing personalized thresholds to prevent such events. Our results suggest that we can develop simple, yet effective, computational algorithms that can be used as preventative mechanisms for diabetes and obesity management.
KW - continuous glucose monitor
KW - decision tree
KW - diabetes
KW - Machine learning
KW - postprandial hyperglycemia
UR - http://www.scopus.com/inward/record.url?scp=85142241414&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85142241414&partnerID=8YFLogxK
U2 - 10.1109/BSN56160.2022.9928449
DO - 10.1109/BSN56160.2022.9928449
M3 - Conference contribution
AN - SCOPUS:85142241414
T3 - BHI-BSN 2022 - IEEE-EMBS International Conference on Biomedical and Health Informatics and IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks - Proceedings
BT - BHI-BSN 2022 - IEEE-EMBS International Conference on Biomedical and Health Informatics and IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks, BSN 2022
Y2 - 27 September 2022 through 30 September 2022
ER -