Key result
Wearable ECG-SCG features improve heart failure severity classification, boosting AUC by ~12 points over HRV alone.
Why the study?
Current HF assessment primarily relies on hospital-based examinations, which limits continuous monitoring and repeated evaluation of disease severity in daily life.
Does adding electromechanical coupling timing features from wearable synchronized ECG-SCG measurements to conventional heart rate variability features improve the classification of heart failure severity (NYHA Class IV vs. II) in patients with heart failure?
Observational (n=11)
Does adding electromechanical coupling timing features from wearable synchronized ECG-SCG measurements to conventional heart rate variability features improve the classification of heart failure severity (NYHA Class IV vs. II) in patients with heart failure?
Synchronized wearable ECG-SCG measurement quantifying electromechanical coupling significantly improves the automated classification of heart failure severity compared to heart rate variability features alone.
Answers are grounded in this paper and cite their sources.
May support wearable HF monitoring via electromechanical coupling; hypothesis-generating for larger prospective validation before practice change.
Heart failure (HF) is a chronic cardiovascular disease characterized by impaired cardiac pumping function, leading to insufficient tissue perfusion and substantial morbidity. Current HF assessment primarily relies on hospital-based examinations, limiting continuous monitoring and repeated evaluation in daily life. This study presents a wearable system for synchronized electrocardiography (ECG) and seismocardiography (SCG) measurement to assess HF severity by jointly characterizing cardiac electrical and mechanical activity. ECG and SCG were recorded overnight from 11 patients with HF at admission (NYHA Class IV) and before discharge after treatment (NYHA Class II). In addition to conventional heart rate variability (HRV) features, a set of electromechanical coupling (EMC) timing features was derived from the synchronized ECG-SCG signals. Four classifiers (SVM, KNN, L2-LR, and AdaBoost) were evaluated using subject-wise nested leave-one-subject-out cross-validation. Significant differences in both HRV and EMC features were observed between Class II and Class IV. Adding EMC features improved performance across all classifiers, with SVM AUC increasing from 0.83 to 0.95. Further evaluation on a public dataset showed that SVM AUC increased from 0.65 to 0.84 after incorporating EMC features. SHAP analysis indicated EMC features to provide complementary information to HRV. In addition, the proposed framework outperformed two state-of-the-art methods. These results support the feasibility of synchronized wearable ECG-SCG measurement for quantifying cardiac electromechanical coupling and provide additional evidence for the cross-dataset robustness of EMC measurements in HF severity assessment.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) conducted an observational in Heart failure (n=11). Synchronized wearable ECG-SCG measurement with electromechanical coupling (EMC) and HRV features vs. Conventional heart rate variability (HRV) features alone was evaluated on Classification of heart failure severity (NYHA Class II vs. Class IV) (AUC 0.95 vs 0.83). Adding wearable ECG-SCG electromechanical coupling features to HRV improved heart failure severity classification (functional class II vs IV), increasing SVM AUC from 0.83 to 0.95.
Synapse has enriched 10 related papers. Consider them for comparative context: