Ami's principal research interests lie in developing solutions that enable in-natura markerless motion capture for biomechanical modelling in Biomedical and Sports Engineering. Specifically, he is interested in the reconstruction of person-specific human pose, kinematics, and surface geometry to enhance our understanding of the non-linear behaviour of human motion, musculoskeletal injury and disease and enable modelling of soft-tissue dynamics. He specialises in translational applied research to develop innovative, highly accurate and tailored deployable evidence-based decision support tools for optimising sporting performance, diagnosis and treatment, improved neuromotor disorder identification using mobility degeneration classification models, and unimpeded patient monitoring of postural control, ambulatory activities and assisted living.
Biomedical Engineering
- Development of automated diagnosis and monitoring systems for quantification of Dyskinesia in patients with Parkinson's Disease from image sequences
- Estimation of pose and surface geometry in general movement assessment of infants with high risk of Cerebral Palsy using convolutional neural network approach
- Development of motor function classification models for the evaluation of rehabilitation outcomes efficacy in stroke patients
Sports Engineering
- Markerless pose estimation
- Automated event detection
- Surface geometry estimation of cyclists for minimisation of aerodynamic drag
- Human-instrument interface design for kayaking
- Activity recognition and motion tracking using convolutional neural networks
- On-water boat instrumentation for rowing and kayaking