Analyzing Heart Rate Variability for COVID-19 ICU Mortality Prediction Using Continuous Signal Processing Techniques

Abstract

Background/Objectives: Heart rate variability (HRV) has been widely investigated as a predictor of disease and mortality across diverse patient populations; however, there remains no consensus on the optimal set or combination of time and frequency domain nor on nonlinear features for reliable prediction across clinical contexts. Given the relevance of the COVID-19 pandemic and the unique clinical profiles of these patients, this retrospective observational study explored the potential of HRV analysis for early prediction of in-hospital mortality using ECG signals recorded during the initial moments of ICU admission in COVID-19 patients. Methods: HRV indices were extracted from four ECG leads (I, II, III, and aVF) using sliding windows of 2, 5, and 7 min across observation intervals of 15, 30, and 60 min. The raw data posed significant challenges in terms of structure, synchronization, and signal quality; thus, from an original set of 381 records from 321 patients, after data pre-processing steps, a final dataset of 82 patients was selected for analysis. To manage data complexity and evaluate predictive performance, two feature selection methods, four feature reduction techniques, and five classification models were applied to identify the optimal approach. Results: Among the feature aggregation methods, compiling feature means across patient windows (Method D) yielded the best results, particularly for longer observation intervals (e.g., using LDA, the best AUC of 0.82±0.13 was obtained with Method D versus 0.63±0.09 with Method C using 5 min windows). Linear Discriminant Analysis (LDA) was the most consistent classification algorithm, demonstrating robust performance across various time windows and further improvement with dimensionality reduction. Although Gradient Boosting and Random Forest also achieved high AUCs and F1-scores, their performance outcomes varied across time intervals. Conclusions: These findings support the feasibility and clinical relevance of using short-term HRV as a noninvasive, data-driven tool for early risk stratification in critical care, potentially guiding timely therapeutic decisions in high-risk ICU patients and thereby reducing in-hospital mortality.

Keywords: COVID-19; HRV; ICU; mortality.

Figure 1

Histogram of absolute frequencies, illustrating distribution of RR intervals in a 2 min space, using the data from our study.

Figure 2

HRV analysis methodology: from raw physiological data to automated interpretation. The top panel illustrates RR interval detection from the ECG signal, which forms the basis for feature extraction. The bottom panel outlines the subsequent steps of the HRV analysis pipeline.

Figure 3

Hierarchy of the HDF5 files used in this study. Each file contains signals, storing the biomedical waveforms along with their sampling frequencies, and TS, containing the corresponding timestamp data. The biomedic data and time data entries reflect the actual recorded signals and their temporal alignment, respectively. FS refers to the sampling frequency associated with each signal.

Figure 4

Detection of the different records in each file.

Figure 5

Example for the 15 min signal analysis.

Figure 6

Method C—majority voting approach.

Figure 7

Method D—mean of each feature across all windows.

Figure 8

Beeswarm plot illustrating the contribution of each HRV feature to the model’s output based on SHAP values. Each dot represents a data instance, with color indicating the normalized feature value (blue = low, red = high), and the position on the x axis representing the SHAP value (impact on model prediction). Positive SHAP values indicate a contribution toward the predicted class, while negative values indicate the opposite.