Intelligent real-time postural control monitoring system for self-rehabilitation systems

The essence of all human motion is maintaining postural alignment while standing upright. Reliable sensory information and its proper integration for the Central Nervous System (CNS) is necessary for human body postural control. Loss of sensory information resulting from injuries (previous or current) or aging is seen to be the leading cause of falls. Poor postural control is a strong indicator of degrading or a loss of sensory information to the CNS. Loss and degradation of certain sensory information, such as the proprioception, can be retrained through rehabilitation by providing augmented or compensatory signals to the CNS. However, conventional rehabilitation, referred to as face-to-face therapy, is known to incur high costs. Therefore, this thesis looks to design and develop a real-time approach for monitoring and correcting postural control, with the view of enabling and promoting self-rehabilitation routines. Self-rehabilitation routines allow for certain rehabilitation procedures to be conducted independent of professionals and therapists, without compromising the final outcome of rehabilitation. Rehabilitation in this work is considered with regard to monitoring occurrences of postural instability and the steps leading towards correction and restoration of posture. A Force Sensing Platform (FSP) which is sensitive to changes of the postural control was designed. The platform was capable of measuring effects of postural control in static and dynamic conditions. The FSP functioned as a qualitative tool for initial on-the-spot postural control assessment and quantitative measurement tool for post-acquisition assessment of balancing abilities. The design and implementation of the FSP presents the potential of incorporating commonly available Force Sensing Resistor (FSR) as a sensing element for gauging and quantifying postural control. The FSP functions to test and gauge postural control of individuals by monitoring pressure distribution at the feet, while providing real-time quantitative information to end-user. Real-time qualitative assessment assisted end users in visually identifying areas of the feet with high force concentration. Soft-computing techniques were applied to the measured data, for post-acquisition analysis which provided quantitative measure of postural control. Successful implementation of FSRs in the FSP improved mobility of the designed platform, allowing for outside laboratory data acquisition which is a desirable feature for self-rehabilitation. An intelligent bio-feedback system was also designed, using vibrotactile feedback. Vibration actuators (vibrotactors) have a simplistic manner of providing augmented or complementary feedback to the CNS using pulses of vibration at varying levels. The designed approach presents a real-time biofeedback system, which uses artificial intelligence for decision making. Artificial intelligence was used to decide on the severity of measured postural control, which then determined the level of feedback provided in real-time. The feedback provided acted as a forewarning mechanism to the CNS. The system relied on miniature kinematic sensors for measures relating to postural control, ensuring a compact and lightweight design. The designed systems demonstrated competence in clinical and biomedical applications, with potential of being expanded into self-rehabilitation. Components of the systems presented simple attachments with minimal calibration, and were capable of providing reliable and consistent measurements. The results in this work demonstrated the viability of the prototypes designed for monitoring an improving postural control for self-rehabilitation purposes. This work was supported in part by Moves Fitness International, USA and the Ministry of Science, Technology and Innovation Malaysia (MOSTI).