Selecting Feature Sets and Comparing Classification Methods for Cognitive State Estimation

Acute stress and high workload are part of everyday work at safety critical fields (e.g. health care). Adaptive human computer interaction systems could support and guide a nurse or a doctor in these hectic situations. Seamless interaction between human and computer requires accurate cognitive state estimation of the person. Currently studies are mainly focused on detecting between two cognitive states with full set of physiologically inspired features. This study demonstrates a classification of different types of stress during Maastricht Acute Stress Test by using feature combinations from electro-oculogram (EOG) and electrocardiogram (ECG) signals in general and personalized approaches, comparing three different classifiers. The classification is evaluated for features extracted from both signals separately and together, and the most important features are selected and reported. Results indicate that the best performance is achieved when features from both EOG and ECG signals are used, and approximately twenty features from EOG and ECG signals are enough to distinguish the two/three states. A personalized approach together with feature selection and support vector machine classifier achieves accuracies of 96.9% and 86.3% in classifying between two states (relaxation and stress) and three states (relaxation, psycho-social stress, and physiological stress), respectively, which exceed state-of-the-art performance. Thus cognitive state estimation benefits from combining selected eye and heart parameters which suggests a promising basis for realtime estimation in the future.