Stress usually causes changes in the status of the primary and secondary sympathetic nerves of the autonomic nervous system. This is also reflected in changes in indicators related to heart rate variability. Therefore, stress identification through the characteristics of heart rate variability is very effective. In this paper, Electrocardiographic Stress Detection (ESD) model based on heart rate variability is proposed to accurately identify stress from an objective perspective using physiological signals. Among them, the heart rate variability (HRV) in the ECG signal is the main feature involved in the ESD model, which is related to the response pattern of the human autonomic nervous system, and the changes of the relevant features represent the active state of the main sympathetic and parasympathetic nerves. Based on these patterns, we extracted the relevant HRV features and used a support vector machine (SVM) classifier for classification calculations with an accuracy of 94.8%.
Heart disease has become one of the most significant threats to life and health. Heart Rate Variability (HRV) features extracted from ECG signals are utilized in the detection of various heart diseases. Additionally, HRV is closely associated with emotions. Therefore, leveraging this characteristic, emotional fluctuations can be linked to heart diseases. This paper proposes a Heart Disease Prediction (HDP) perception model based on features fusion, which achieves high-precision recognition of 16-dimensional emotions through HRV features with an accuracy of 91.1%. It establishes a correlation model between emotional fluctuation rule system and heart disease, utilizing potential emotional fluctuation patterns to predict heart diseases without obvious symptoms. The final model demonstrates promising results. The prediction accuracies for Arrhythmia (less common), Atrial Fibrillation, and Myocardial Ischemia are 91.2%, 90.9%, and 95.7% respectively.