My research agenda is to develop practical AI tools for solving the challenges of real-world applications. In essence, it is to model the practical problems using mathematical languages and develop machine learning algorithms for the optimal solution, bridging the gap between signal/data and human-understandable knowledge. During my study in Open Lab as a CDT student, I have gained extensive experience working with time-series data, such as biosignals, which have broad applications in physical behaviour assessment, health, and well-being monitoring, etc. I also developed mechanisms for increasing the transparency/interpretability of complex AI models, which is crucial for many applications such as automated medical diagnosis.

At the late stage of my PhD, I was a research assistant at the School of Computing at Newcastle University, worked on the IDEA-FAST project (€40 million) to identify digital endpoints and biomarkers of sleep disturbance and fatigue. At the same time, I also joined the MRC Epidemiology Unit at the University of Cambridge as a visiting researcher, focusing on applying machine learning to human activity recognition tasks.

Area of expertise: Sleep, Activity Recognition, Automated Health Assessment, Wearable/Ubiquitous Computing, Machine Learning.

Thesis Title

Towards Automated Sleep Stage Assessment Using Ubiquitous Computing Technologies

After the PhD

I joined the Northumbria University as a lecturer in Computer Science to pursue my future career in ML4Health.

Collaboration Organizations:

• Cambridge University
• MIT
• Georgia Tech University