Publication date: Jun 23, 2025

Heart rate variability (HRV) is a validated biomarker of autonomic and inflammatory regulation, and has been associated with both acute COVID-19 and long COVID. Although RT-PCR remains the diagnostic gold standard for acute infection, there is a lack of accessible, noninvasive physiological tools to support ongoing monitoring and stage differentiation of COVID-19 and its sequelae. The growing availability of wearable devices capable of real-time HRV data collection opens opportunities for early detection and health status classification using machine learning. This study aimed to identify HRV patterns capable of distinguishing individuals with active COVID-19, long COVID, and healthy controls, using data collected from wearable devices and processed with machine learning models. A secondary objective was to assess the feasibility of a near-real-time health monitoring system based on these patterns using wearable-derived HRV data. HRV indices (SDNN, RMSSD, LF%, HF%) were collected from 61 participants (21 with active COVID-19, 20 with long COVID, and 20 healthy controls) using two standardized datasets. Classification models were developed using supervised machine learning algorithms (Decision Tree, SVM, k-NN, Neural Networks) and evaluated through cross-validation. A contextual clinical variable indicating recent SARS-CoV-2 infection was incorporated into one model configuration to assess its impact on classification performance. In addition, a prototype system for near-real-time monitoring was implemented and tested in a separate group of 4 participants. Participants with active COVID-19 showed significantly lower HRV indices (SDNN, RMSSD, LF%, HF%) compared to both long COVID and healthy controls (P ≤ . 0007), while differences between the long COVID and healthy groups were not statistically significant. Decision tree models trained solely on HRV features achieved 76. 4% accuracy, with high discriminative performance for active COVID-19 (F1 = 88%, AUC = 0. 85), but limited detection of long COVID (F1 = 56%). When a contextual clinical variable indicating recent SARS-CoV-2 infection was included, overall accuracy increased to 87%, and the F1-score for long COVID rose to 92%, with improved AUC metrics across classes. A prototype system tested on four independent participants correctly classified their status, demonstrating feasibility for near-real-time application. HRV patterns collected from wearable devices and analyzed via machine learning successfully distinguished active COVID-19 from healthy individuals with high accuracy using physiological data alone. When a clinical contextual variable indicating recent infection was added, the model also achieved strong performance in identifying long COVID cases. A prototype system demonstrated feasibility for near-real-time application, reinforcing the potential of HRV for individualized health monitoring in line with emerging trends in digital and predictive healthcare.

Semantics

Original Article