Supervised and Dynamic Neuro-Fuzzy Systems to Classify Physiological Responses in Robot-Assisted Neurorehabilitation
Open Access
- 22 May 2015
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 10 (5) , e0127777
Abstract
This paper presents the application of an Adaptive Resonance Theory (ART) based on neural networks combined with Fuzzy Logic systems to classify physiological reactions of subjects performing robot-assisted rehabilitation therapies. First, the theoretical background of a neuro-fuzzy classifier called S-dFasArt is presented. Then, the methodology and experimental protocols to perform a robot-assisted neurorehabilitation task are described. Our results show that the combination of the dynamic nature of S-dFasArt classifier with a supervisory module are very robust and suggest that this methodology could be very useful to take into account emotional states in robot-assisted environments and help to enhance and better understand human-robot interactions.Keywords
This publication has 18 references indexed in Scilit:
- Comprehensive Overview of Nursing and Interdisciplinary Rehabilitation Care of the Stroke PatientStroke, 2010
- Neural systems subserving valence and arousal during the experience of induced emotions.Emotion, 2010
- Distinct brain systems underlie the processing of valence and arousal of affective picturesBrain and Cognition, 2009
- An affective circumplex model of neural systems subserving valence, arousal, and cognitive overlay during the appraisal of emotional facesNeuropsychologia, 2008
- The neurophysiological bases of emotion: An fMRI study of the affective circumplex using emotion‐denoting wordsHuman Brain Mapping, 2008
- On the relationship between emotion and cognitionNature Reviews Neuroscience, 2008
- Brain activity underlying emotional valence and arousal: A response‐related fMRI studyHuman Brain Mapping, 2004
- Dissociated neural representations of intensity and valence in human olfactionNature Neuroscience, 2003
- Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional mapsIEEE Transactions on Neural Networks, 1992
- A circumplex model of affect.Journal of Personality and Social Psychology, 1980