[Affective Stress Rating Method Based on Improved Hidden Markov Model].

Area of Science:

• Affective computing
• Biomedical signal processing
• Psychophysiology

Background:

• Existing affective stress evaluation methods struggle to accurately rate stress or account for the influence of previous stress states.
• A need exists for a more comprehensive model that integrates physiological data with psychological factors.

Purpose of the Study:

• To propose and validate an improved affective stress rating model using electrocardiogram (ECG) data.
• To address the limitations of current methods by incorporating temporal stress influences and environmental factors.

Main Methods:

• Developed a hidden Markov model (HMM) for affective stress rating, integrating affective computing principles.
• Utilized ECG complexity and approximate entropy as key features within the HMM.
• Combined the HMM with investigation questionnaires for a holistic stress assessment.

Main Results:

• The proposed model demonstrated improved accuracy in affective stress rating compared to existing methods.
• Incorporating environmental factors and the influence of previous stress states significantly enhanced rating precision.
• Feature analysis indicated that ECG complexity and approximate entropy are valuable indicators for stress assessment.

Conclusions:

• The developed affective stress rating model, utilizing HMM and ECG features, is an effective approach for stress evaluation.
• Considering the dynamic nature of stress, including prior states and environmental context, is crucial for accurate affective stress assessment.
• This method offers a more robust and accurate way to understand and quantify affective stress.