Automated Detection of Gait Disorders


Rezaul Begg;
Post Doctoral Research Fellow: Daniel T. H. Lai, Ahsan Khandoker, Pazit Levinger.

Student: Simon Taylor, Kate Lynch, Chandan Karmaker.

Cathy Said, Austin Health;
Keith Hill, NARI;
Richard Baker, Royal Children Hospital;
Adam Miller, Marybed Hospital, USA.

Introduction: There are several projects currently in progress in this area. The major aim here is to develop intelligent automated diagnostic systems that are capable of detecting gait pattern changes caused by diseases or post-operative and other interventions which can affect the locomotor system.
Significance: We are investigating various gait pathologies such as elderly with balance impairments leading to falls, post-operative recovery in knee osteoarthritis and patellofemoral pain syndrome, and multi-classification of hemiplegics in cerebral palsy patients.
Applications: Potential application areas include: screening fallers for minimizing falls incidence, diagnosis of gait pathology, evaluation of treatments and therapies, and rehabilitation monitoring.
Challenges: In this endeavour, various computational intelligence techniques such as support vector machines and artificial neural networks are being employed to learn relationships between gait types (e.g., falling behaviour) and the related gait parameters (e.g., temporal-spatial, lower limb motion, and foot-ground reaction force-time data).
top of page