Treat Me Well: Affective and Physiological Feedback for Wheelchair Users Octavian Postolache Instituto de Telecomunicações ISCTE-IUL, Lisbon, Portugal opostolache@lx.it.pt Pedro Silva Girão, Mário Ribeiro Instituto de Telecomunicações Lisbon, Portugal psgirao@ist.utl.pt Helder Carvalho, André Catarino, Universidade do Minho, Braga, Portugal helder@det.uminho.pt Gabriela Postolache Universidade Atlântica, Oeiras, Portugal gabrielap@uatlantica.pt Abstract — This work reports a electrocardiograph and skin conductivity hardware architecture, based on E-textile electrodes, attached to a wheelchair for affective and physiological computing. Appropriate conditioning circuits and a microcontroller platform that performs acquisition, primary processing, and communication using Bluetooth were designed and implemented. To increase the accuracy and repeatability of the skin conductivity measuring channel, force measurement sensors were attached to the system certifying measuring contact force on the electrode level. Advanced processing including R- wave peak detector, adaptive filtering and autonomic nervous system analysis based on wavelets transform was designed and implemented on a server. A central design of affective recognition and biofeedback system is described. Keywords - affective computing, physiological computing, embedded system, E-textile, heart rate variability I. INTRODUCTION Mrs. Alda (fictitious name) is an old age woman with elegant body posture and clothes. Age traces in her clear pink skin are faded by brilliance of her blue tiny eyes remembering the many grannies who with goodness and wisdom reflected in the hands and body movements, gathered around the warm cakes and fascinating stories, the grandchildren, the family, the community. She is an 88 years old lady invited to a Nursing Workshop to speak about her experience on retirement residence where she has been living for the past 5 years. She says that she feels well mainly when she can receive her family visits and when she can talk with the nurse. When asked to indicate what she desires to demand from any nurse from a retirement residence, she says shortly, with a smile that seems guilty, that what she wants is to be “well treated”. Nobody ask her to give examples of what she means by “well treated”, perhaps because before her presence on the workshop, was analyzed a text from a letter of a American nurse, Mrs. Rosemary Wills, who in order to prepare her future in a nursing home, wrote on the needs of aged people and on the attitude that nurses should have in order to respect old people’s dignity. It is outlined in this letter that – “Perhaps it will seem demanding, exaggerated in what I want, but what I want is just to receive affection, to be well treated, and a friendly, lovely person to take charge of me”. Our team has been involved in the development of electronic health systems that allow non-invasive or unobtrusive health and healthcare monitoring. Currently, we investigate developed or in course of implementation technologies that should allow affective communication, a perception of well treatment, and might turn the caregivers more friendly and lovely. In our work we search for solution to a range of issues related with affective and physiological computing for wheelchair users: What emotions or moods should be outlined, when, and how? Are deep models of emotion necessary? The programed software should respond to all situations or can respond only to some events and selected requirements? How the affective expressions maintain subject engagement? Which expressive modalities should be used (e.g. face expressions, gestures, behavior selection)? What role do the subject´s emotions play in the developed affective and physiological sensing? Does the personality of the involved people must be assessed in affective computing? What information about the person affective reactions should enable adaptations to the daily life challenges? The aim of this work is to explore the rich possibilities of affective computing and physiological computing in order to set out new research for future healthcare solutions. In Sections 2 and 3 of the paper a general introduction on affective and physiological computing is presented. In Section 4 we introduce our work on unobtrusive psychophysiological interactive computer system implementation. Finally, we conclude with future directions of research. II. AFFECTIVE COMPUTING Affective Computing is a “branch of study and development of artificial intelligence that deals with the design of systems and devices that can recognize, interpret, and process human emotions” [1]. The pioneer of this type of research, Rosalind Picard, defines Affective Computing as “computing that relates to, arises from, or deliberately influences emotion” [2]. 978-1-4673-0882-3/12/$31.00 ©2012 IEEE