Year: 2025
Type: article
Abstract: OBJECTIVES: Stress is a physiological state that is essential for the survival of living organisms. Heart rate variability (HRV) and cortisol hormone are indicators of the stress system. According to research, it has been demonstrated that the activation of the stress system is not consciously controlled by the individual, but rather occurs subconsciously. It is a novel concept to employ HRV indexes to assess the level of cortisol concentration as a more reliable indicator of stress system activation, as opposed to relying solely on the individual's emotional state. METHODS: In order to understand the relationship between stress and cortisol secretion and its effect on electrophysiological biomarkers like HRV, the algorithms were designed using machine learning algorithms such as SVM, XGB, and MLP in the 634 adult healthy men (20-50 years old). Trait social stress test was utilized to make wide range of cortisol concentration from no to moderate stress. RESULTS: These algorithms classified cortisol level between 9:00 AM and 2:00 PM in the optimal (5-15 ng/mL), non (less than 5 and more than 15 ng/mL) range, using HRV indexes (12 features). The XBG algorithm could achieve best classification with an accuracy rate of 99 % and an F1 rate of 99 %. They also indicated the state of an individual's stress system by indicating the concentration level of cortisol, which is its fundamental indicator. CONCLUSIONS: In addition to classifying stress, the HRV can also classify salivary cortisol in adult health men.
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid:
OpenAlex ID: W4409408087