Abstract:
Stress is a silent killer, affecting millions of people worldwide and taking a toll on physical and mental health. But what if we could identify and manage stress before it becomes a problem? In this talk, we will explore how multiple physiological signals can be used to classify stress levels and develop personalized stress management strategies.

Description:
This study is part of a major collaboration program between Unisinos (Brazil) and Friedrich-Alexander Erlangen-Nuremberg (Germany) institutions, discussing the advantages of using multiple physiological signals, respecting the unique characteristics of each signal and its potential for classification.
Next, we will discuss the challenges associated with using multiple physiological signals for classification, including signal processing, feature extraction, and feature fusion. We will introduce various methods for addressing these challenges, such as signal filtering, time-frequency analysis, and machine learning and Deep Learning algorithms.
Takeaways:
Deeper understanding of the potential benefits of using multiple physiological signals for classification
Challenges associated with stress detection
Laboratory results and state of the art