Key result
Deep learning classifier using wearable ECG and HRV detects acute hyperglycemia with ~89% accuracy.
Why the study?
Noninvasive blood glucose monitoring is crucial for detecting early dysglycemia, but continuous monitors remain invasive and costly, while ECG and derived HRV may offer a noninvasive indicator.
Does a deep learning classifier using ECG and HRV features accurately detect acute hyperglycemia in adults?
Observational (n=30)
Does a deep learning classifier using ECG and HRV features accurately detect acute hyperglycemia in adults?
Answers are grounded in this paper and cite their sources.
A deep learning model utilizing ECG and HRV features can accurately detect acute hyperglycemia, demonstrating the potential for noninvasive diabetes risk screening using wearable devices.
Noninvasive blood glucose monitoring is crucial for detecting early dysglycemia, yet continuous glucose monitors remain invasive and costly. Electrocardiogram (ECG) and its derived heart rate variability (HRV) measure may offer a noninvasive indicator of autonomic and cardiac responses associated with acute changes in glucose. In this study, 30 adults underwent a 75 g oral glucose tolerance test with concurrent ECG Holter and interstitial glucose monitoring. From these recordings, HRV and ECG features were extracted. A deep learning classifier with HRV and ECG was then trained to detect hyperglycemia (glucose ≥ 180 mg/dL). Cross-correlation analysis confirmed a significant association between HRV and glucose (Pearson r ~0.65, p < 0.05) when aligning each participant’s data according to individual response delays. The model achieved high classification performance under rigorous temporal validation (accuracy ~89%, area under the receiver operating characteristic curve ~0.89). Saliency analyses revealed that the classifier’s decisions focus on distinct ECG waveform transitions and key HRV features linked to glucose-induced autonomic changes. Overall, acute hyperglycemia elicited discernible changes in HRV and cardiac conduction, supporting the feasibility of this physiologically grounded approach for detecting the acute hyperglycemic phase under controlled conditions. This method holds promise for real-time implementation in wearable devices, enabling early diabetes risk screening.
No takes yet. Share an insight, caveat, or question.
Ha et al. (2026) conducted an observational in Acute hyperglycemia (n=30). Deep learning classifier using HRV and ECG features was evaluated on Detection of hyperglycemia (glucose ≥ 180 mg/dL) (AUC ~0.89). A deep learning classifier using wearable ECG and HRV features detected acute hyperglycemia (glucose ≥ 180 mg/dL) with ~89% accuracy and an AUC of ~0.89.
Synapse has enriched 10 related papers. Consider them for comparative context: