Home-based rehabilitation has evolved in recent years as a cost-effective and convenient alternative to traditional clinical rehabilitation. Researchers have developed various types of sensors-based rehabilitation systems that incorporate Virtual Reality games aimed to offer the patient an entertaining and beneficial training experience from the comfort of home. This has consequently created the need to design reliable assessment and adaptation mechanisms that are able to measure and analyze the patient's performance and condition, and to accordingly make proper adjustments that conform to the abilities of the patient during the training. In this dissertation, we introduce our context-based adaptive home-based rehabilitation framework (CAHR) that offers the patients a rehabilitation environment that can adapt based on their physical, physiological, and psychological context, while taking into consideration the environmental conditions that may hinder their progress. CAHR is a generic framework that can be implemented to fit any of the upper or lower extremity rehabilitation. However, in this dissertation, we base our modeling and analysis mainly on the wrist. In CAHR, the physical condition of the patient is assessed by a fuzzy logic-based mechanism that uses the various kinematics captured during the training to provide a quantified value which reflects the Quality of Physical Performance of the patient. The rehabilitation task adaptation is achieved based on a special algorithm that defines how the physical training, psychophysiological responses, and environmental conditions must be manipulated in order to match the desired performance target parameters set by the therapist. The simulation results have shown that the proposed adaptation engine can properly adjust the rehabilitation environment based on different simulated performance behavior that might be produced by a patient. In addition, training with a special game that has been designed based on the developed framework has shown improvement in the physical capabilities of two patients suffering from upper extremity impairments.

Home-based rehabilitation has evolved in recent years as a cost-effective and convenient alternative to traditional clinical rehabilitation. Researchers have developed various types of sensors-based rehabilitation systems that incorporate Virtual Reality games aimed to offer the patient an entertaining and beneficial training experience from the comfort of home. This has consequently created the need to design reliable assessment and adaptation mechanisms that are able to measure and analyze the patient's performance and condition, and to accordingly make proper adjustments that conform to the abilities of the patient during the training. In this dissertation, we introduce our context-based adaptive home-based rehabilitation framework (CAHR) that offers the patients a rehabilitation environment that can adapt based on their physical, physiological, and psychological context, while taking into consideration the environmental conditions that may hinder their progress. CAHR is a generic framework that can be implemented to fit any of the upper or lower extremity rehabilitation. However, in this dissertation, we base our modeling and analysis mainly on the wrist. In CAHR, the physical condition of the patient is assessed by a fuzzy logic-based mechanism that uses the various kinematics captured during the training to provide a quantified value which reflects the Quality of Physical Performance of the patient. The rehabilitation task adaptation is achieved based on a special algorithm that defines how the physical training, psychophysiological responses, and environmental conditions must be manipulated in order to match the desired performance target parameters set by the therapist. The simulation results have shown that the proposed adaptation engine can properly adjust the rehabilitation environment based on different simulated performance behavior that might be produced by a patient. In addition, training with a special game that has been designed based on the developed framework has shown improvement in the physical capabilities of two patients suffering from upper extremity impairments.