Detección de Patrones de Fatiga Cognitiva Mediante Aprendizaje Automático

Abstract

This research presents a comprehensive investigation into the current state of cognitive fatigue detection through the analysis of physiological signals, specifically electrocardiogram (ECG) and electrodermal activity (EDA), using machine learning techniques. The study is based on a critical review of recent scientific literature, focusing on the effectiveness of various machine learning approaches such as Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN) in identifying cognitive fatigue patterns. The investigation explores the efficacy of feature selection techniques for ECG and EDA signals, aiming to identify the most relevant indicators for cognitive fatigue detection as reported in current literature. A key aspect of the research is the examination of individual and cross-population variability, analyzing transfer learning strategies and cross-validation methods proposed by other researchers. This study's scope is purely exploratory, synthesizing existing knowledge without developing new systems or algorithms. It provides a comprehensive overview of the state-of-the-art in non-invasive cognitive fatigue detection using physiological signals and machine learning. The findings and conclusions derived from this analysis offer valuable insights into current methodologies, challenges, and future research directions in this field. By consolidating and critically examining the latest advancements, this research aims to establish a solid foundation for future studies in cognitive fatigue detection, contributing to the ongoing efforts to enhance safety and efficiency in high-stakes environments where sustained cognitive performance is crucial. Keywords: Cognitive fatigue, ECG, EDA, Machine learning, Physiological signals, Transfer learning