I have worked on a variety of projects funded by various government and industrial bodies (listed here). My recent focus has been on Mobile Health and how we can improve devices and sensing, as well as their generated data analysis, to aid the diagnostic and illess progession monitoring at scale and effectively. Things like on-device machine learning, challenges of machine learning for sensor/multisensory data, systems for machine learning are key to my current research. Below a sample of most recent projects. For all results of my work please check my publication list.
On Device Audio based Analytics
I work on a variety of techniques to use audio to understand behaviour (e.g., emotions) and medical conditions from user worn devices efficiently and effectively. My ERC Project concentrates on this.
Device Efficient and Effective Mobile Technologies for Diagostics of Neurological Diseases
Through a variety of funding sources, we are exploring how to make better mobile/wearable technologies for early diagnosis and progression of neurological diseases such as Alzheimer's Disease.
Machine Learning for Multisensory Data in Mobile Health.
Data from mobile and wearable device sensors present unique challenges to machine learning: we solve them in the context of mobile health analytics.
Mobile Health: Behaviour Monitoring and Interventions
We have a number of projects which use mobile device software to monitor people's behaviour in a variety of settings, for example of medical adherence, smoking cessation and mental health. .