Background: Heart Rate Variability (HRV) is intimately associated with stress and can serve as a valuable indicator of the individual’s stress level. HRV is the variation in the length of time between heartbeats. Lower HRV is associated with higher stress levels, while higher HRV indicates better stress resilience and adaptability. The HRV parameters can be classified as − time-domain, frequency-domain, and non-linear. Parameters associated with HRV can be employed to assess individuals’ health and observe the effects of interventions such as exercise, stress reduction, and medication. Research in the field of artificial intelligence (AI) is ongoing that attempts to classify stress based on HRV data. HRV, which is associated with stress and physiological health, has received attention as a potential component to incorporate into models that classify and predict stress levels accurately. Monitoring HRV can offer insights into the interplay between stress and mental health, aiding in early detection and holistic approaches to well-being.
Objective: The primary goal of this study is to perform semantic modeling of the vital HRV features in a knowledge graph, followed by, developing an accurate, reliable, explainable, and ethical AI model pipeline that can perform predictive analysis on HRV data.
Methods: In this regard, we have considered the well-known multimodal SWELL knowledge work (SWELL−KW) dataset as a study case that represents the following stress conditions − no stress, time pressure, and interruption. The selected HRV dataset shows a labeled relationship between HRV and stress levels, which is deemed suitable for this study. We have explored different feature selection and dimensionality reduction techniques to extract relevant features from the HRV dataset to enhance classification accuracy with reduced bias. We have used various machine learning (ML) algorithms (e.g., traditional and ensemble) for the predictive analysis of imbalanced and balanced HRV datasets. We have used different data formats (e.g., scaled, normalized, and standardized) and various oversampling techniques (e.g., Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic (ADASYN)) to generate synthetic samples from the minority class. We have used a Tree-Explainer (e.g., Shapley Additive Explanations (SHAP)) to explain the model classifications.
Results: As the HRV features are non-linear, the genetic algorithm-based feature selection followed by, model classification with Random Forest Classifier has produced the highest classification result on both the imbalanced and balanced datasets. The optimized feature set has been beneficial to design and develop a stress management system with a Semantic framework. Therefore, we have introduced the concept of domain ontology to represent the data and obtained knowledge. The consistency of the Ontology model has been evaluated with Hermit reasoners and reasoning time.
Conclusions: Overall, HRV serves as a valuable physiological marker that can provide insights into the relationship between stress and mental health. It’s crucial to recognize that while HRV is a valuable non-invasive indicator of stress levels, its interpretation should be combined with other subjective and objective measures of stress in order to comprehensively understand the individual’s stress response. Therefore, the monitoring of HRV may help individuals assess the effectiveness of stress management techniques and interventions.