On the Predictability of non-CGM Diabetes Data for Personalized Recommendation

TL;DR

This study evaluates the potential of machine learning models to predict blood glucose levels using infrequent, non-continuous data for personalized diabetes management, addressing the challenge of limited data availability.

Contribution

It introduces post-prediction methods to improve blood glucose prediction accuracy from sporadic measurements, expanding predictive capabilities beyond continuous glucose monitoring systems.

Findings

Post-prediction methods marginally improve prediction accuracy.

Predictability varies across patients with limited data.

Machine learning models show potential for non-CGM blood glucose prediction.

Abstract

With continuous glucose monitoring (CGM), data-driven models on blood glucose prediction have been shown to be effective in related work. However, such (CGM) systems are not always available, e.g., for a patient at home. In this work, we conduct a study on 9 patients and examine the online predictability of data-driven (aka. machine learning) based models on patient-level blood glucose prediction; with measurements are taken only periodically (i.e., after several hours). To this end, we propose several post-prediction methods to account for the noise nature of these data, that marginally improves the performance of the end system.