Research Output
Peer-reviewed research at the intersection of machine learning, wearable sensing and human health.
Combines an instrumented clinical mobility test with machine learning to assess locomotive syndrome, enabling objective and scalable screening of mobility decline in older adults.
A multimodal deep learning architecture that fuses visual data with physiological bio-signals to recognise both the intensity and category of human emotion.
Demonstrates that fall risk in older adults can be assessed from a single inertial measurement unit — a low-cost, unobtrusive alternative to lab-based gait analysis.
Validates a single-IMU approach for assessing locomotive syndrome, moving clinical mobility screening out of the lab and into everyday settings.
A comprehensive performance evaluation of a novel balance assessment mat prototype benchmarked against inertial sensing, supporting affordable balance screening technology.
Systematically reviews and meta-analyses the evidence on dual-task, sensor-based motion analysis as a digital biomarker for dementia — mapping what movement patterns reveal about cognitive decline.
I'm open to research collaborations in ML for health, wearable sensing and computer vision.