AI in chronic disease is the application of data monitoring, prediction, and Machine Learning to long-term health problems. The goal in this vault is practical prevention: use data to detect patterns early enough to reduce risk or improve intervention timing.
Motivation
The long-term career goal is to use technical expertise to reduce the death rate or burden of chronic diseases. This is not only a modeling problem. It requires useful data, clinically meaningful targets, and deployment paths where predictions can actually change behavior or care.
Learning Path
The current preparation path has three parts:
- strengthen basic theory and implementation instead of rushing for results
- practice leetcode enough to support coding interviews and implementation fluency
- build Pytorch skill through repeated coding, especially tensor manipulation and model implementation
The emphasis on implementation is deliberate: health AI needs implementation ability, not only conceptual understanding.
Research And People To Track
Singapore-related starting points:
- Our Team - Group Mornin
- Jiaxing Xu (徐家兴) | National University of Singapore
- Han Leong Goh | Publications | National University of Singapore
- AI Competition
- AI Competition People
- Phillips Wong
- Ching Chi Keong
- National Heart Center Singapore
Open Questions
- What chronic disease area should be the first concrete target?
- What data streams are realistic: clinical records, wearable data, imaging, lab results, or patient-reported data?
- Which task is most actionable: risk prediction, early warning, disease progression modeling, treatment recommendation, or patient stratification?
- What evaluation metric would reflect patient benefit rather than only model accuracy?