Projects

BreathAI

Breath AI is an AI-powered, non-invasive diagnostic platform that analyzes exhaled breath to screen for multiple diseases in just one sample. By combining machine learning with mass spectrometry, it detects volatile organic compounds (VOCs) in human breath—biomarkers linked to both infectious and chronic diseases. The system delivers fast infectious disease screening (e.g., COVID-19, tuberculosis) in under 3 minutes, and early warning for chronic illnesses such as cancer, cardiovascular disease, and diabetes. Designed for global impact, the technology supports United Nations Sustainable Development Goal 3 (Good Health and Well-Being) by enabling affordable, accessible, and scalable health screening. It also contributes to SDG 10 by being deployable in underserved settings thus reducing healthcare inequalities.

Reported Video: Youtube

Rewards: UN AI for GOOD Asia Invention Geneva Invention Silicon Vally Invention

Virtual NUS

Virtual NUS

Virtual NUS Outdoor Model Animation Virtual NUS Indoor Model Animation
Informative and Refined 3D Model of NUS

The Virtual NUS program aims at building a 3D digital twin of the NUS campus. The 3D digital model is far more sophisticated, informative, and interactive than the previous versions (e.g. Google 3D Map). It is supposed to pave the way for cutting-edge technology like autonomous driving and virtual reality. More information can be found in the video.

Campus3D Dataset


Visualization of Campus3D Dataset

The Campus3D dataset is the first photogrammetry point cloud dataset with hierarchical annotations. It was built to promote the development of large-scene 3D understanding techniques, which might benefit a wide range of 3D applications. More information can be found on the Official Website.