Detection of Emotional Sensitivity Using fNIRS Based Dynamic Functional Connectivity

In this study, we proposed an analytical framework to identify dynamic task-based functional connectivity (FC) features as new biomarkers of emotional sensitivity in nursing students, by using a combination unsupervised and supervised machine learning techniques. The FC was measured Near-Infrared Spectroscopy (fNIRS), computed sliding window correlation (SWC) analysis. A k-means clustering technique applied derive four recurring states. states were characterized both graph theory semi-metric Occurrence probability state transition extracted network features, Random Forest (RF) classifier implemented detect sensitivity. method trialled on 39 students 19 registered nurses during decision-making, where assumed have developed strategies cope with Emotional stimuli selected from International Affective Digitized Sound System (IADS) database. Experiment results showed that demonstrated single dominant task-relevance, while displayed two had higher level task-irrelevant connectivity. also more susceptive stimuli, the derived provided stronger discriminating power than heart rate variability (accuracy 81.65% vs 71.03%) This work forms first study demonstrate stability fNIRS based biomarker. conclusion, support distribution could help reveal differentiating factors between decision making, it is anticipated might be used indicators when developing professional training related