Research Areas

Computational Intelligence in Gait and Human Movement

We are interested in the detection, rehabilitation and monitoring of gait disorders by incorporating nonlinear analysis via computational intelligence techniques. We are investigating the use of signal processing techniques such as wavelet transforms and autoregressive processes (ARMA) to extract important quasi-temporal dependencies in gait parameters. These data are then further interpreted using computational intelligence techniques such as Support Vector Machines (SVM), neural networks, fuzzy logics and evolutionary methods.

Our current research interests cover the following:

1. Early detection and prediction of elderly individuals at risk of suffering tripping falls.
2. Determination of biomechanical factors leading to patellafemoral pain syndrome (PFPS).
3. Predicting disorder progression and post-surgery recovery in knee osteoartritis patients.
4. Design of intuitive classifications for spastic hemiplegic children suffering from cerebral palsy.

Core research areas

1. Biosignals analysis i.e., waveform and time series analysis
2. Intelligent systems design with new machine learning algorithms
3. Gait parameter feature selection algorithms
4. Dynamical systems and nonlinear analysis

Design of wireless sensing devices for monitoring

We are interested in the design of low-powered, cost effective and accurate sensing devices for biomedical applications. We are currently designing devices for the following applications.

1. Monitoring the toe clearance for elderly gait (prototype available)
2. Activity monitor for detecting postural changes (design and test phase)
3. Monitoring progression of osteoartritis from knee kinematics (design and test phase)
4. Personal indoor tracking using pedastrian dead reckoning (design and test phase)
5. Monitoring oxygen saturation - a cheap pulse oximeter (feasibility study)

Core research areas

1. Effective wireless protocols for low power high bitrate transmissions
2. Improving inertial sensors for displacement measurements
3. Sensor optimization and data fusion techniques
4. Localization for continuous personal tracking