A computer-aided decision system for gait analysis
The term ‘cerebral palsy’ refers to a group of permanent disorders of the development of movement and posture that are attributed to non-progressive injuries of the developing brain. Although the brain lesion itself is static, the motor disorders are often accompanied by secondary musculoskeletal impairments. The primary motor problems of spasticity, reduced motor control and weakness typically lead to bony torsions (rotations) and muscle contractures (shortened muscle-tendon units). The major consequence of these developing deformities is increased difficulty with walking.
Gait analysis measures the functional symptoms to inform diagnosis and improve surgical management. The use of gait analysis data in planning complex orthopaedic surgery for children with cerebral palsy is now well established. Surgical planning without gait analysis leads to different recommendations and surgery based on gait analysis data has better outcomes. In the developed world, gait analysis is seen as an essential diagnostic tool to plan complex surgery.
Historically gait analysis has been expensive, prone to measurement error and the results were difficult to interpret. Over the last decade, however, the real cost of movement analysis equipment has more than halved. Expensive manual processing of data has been replaced by computerised processing in real-time. Recent work has demonstrated that modern systems are capable of high levels of accuracy. The major remaining barrier to the wider use of gait analysis is the difficulty of interpreting the complex data that it provides.
The goal of gait analysis is to diagnose a list of impairments that prevent the individual from walking normally. Impairments are musculoskeletal problems such as muscle contracture, muscle spasticity or bony deformities. Interpretation of gait analysis data has remained essentially unchanged since the pioneering case studies published by Dr. Jim Gage in 1983. The impairments may be detected from abnormal features present in the kinematic and kinetic curves. Features that might be interpreted include the magnitude and waveform of the different curves, the difference between the patient’s curve and those from healthy subjects or the differences between left and right curves.
Interpretation is not straightforward because: (i) one feature in one curve may relate to several impairments, (ii) there may be several features corresponding to several impairments superimposed in one curve and (iii) one impairment may lead to abnormal features in several other curves. Gait analysis generates large quantities of data and interpretation is only possible after detailed and time-consuming study by experts in the field. It currently takes 4 hours to interpret gait analysis data by an expert in our centre. Despite more than 20 years of experience interpreting data, there has been little progress in making this process faster, more objective, more reliable and more accessible to non-specialised clinicians or other key users. The aim of this project is to develop a computer-aided decision system to assist and streamline the interpretation of gait analysis data.
- NH&MRC CP-CRE, Chief Investigators: Professor Dinah Reddihough, Professor H.Kerr Graham, Professor Christine Imms, Professor Nadia Badawi, Associate Professor Michael Coory, Professor Eve Blair, Professor Rob Carter.
- Canadian Institutes of Health Research: Professor Unni Narayanan, Dr Darcy Fehlings, Professor H. Kerr Graham, Dr R Hamdy, Dr Kishore Mulpuri.
- NHMRC project grant 1100376
- Sangeux M, Passmore E, Graham HK, Tirosh O. The gait standard deviation, a single measure of kinematic variability. Gait Posture. 2016;46:194-200.
- Sangeux, M. and S. Armand, Kinematic Deviations in Children with Cerebral Palsy in Orthopedic Management of Children with Cerebral Palsy: A Comprehensive Approach, F. Canavese and J. Deslandes, Editors. 2015, Nova science publishers.
- Sangeux, M., Rodda J. and HK. Graham, Sagittal gait patterns in cerebral palsy: The plantarflexor-knee extension couple index. Gait & Posture. 2015;41(2):586-91.
- Sangeux, M. and J. Polak. A simple method to choose the most representative stride and detect outliers. Gait & Posture. 2015;41(2):726-30.
Faculty Research Themes
School Research Themes
For further information about this research, please contact the research group leader.
Department / Centre
Unit / Centre
MDHS Research library
Explore by researcher, school, project or topic.