Research Areas

MedAI-Hub focuses on developing novel AI-driven predictive models to facilitate personalized diagnostics and treatment planning, and disease trajectory prediction. Our research leverages on advanced techniques in natural language processing, computer vision, medical image analysis and other AI methodologies. By integrating diverse data sources such as multimodal imaging, electronic health records, omics and clinical data, these technologies will enable more accurate and efficient clinical decision-making, and optimize treatment interventions.

At MedAI-Hub, we develop new interpretable machine learning methods to integrate multi-omics data - including genomics, transcriptomics, proteomics and metabolomics - to uncover novel biomarkers and improve disease prediction. Our research also focuses on leveraging advanced deep generative models for in silico drug discovery, enabling computational generation and evaluation of potential drug-like molecules, reducing reliance on expensive and time-consuming laboratory experiments. These technologies will also enable discovery of biomarkers for predicting therapeutic responses, and novel targets for emerging treatments.

MedAI-Hub is committed to developing AI solutions that are fair, reliable, and interpretable for real-world healthcare applications. We incorporate privacy-preserving AI techniques such as differential privacy and homomorphic encryption to ensure privacy and security of individual medical data. Our research also embeds principles of fairness and expandability into AI algorithms by designing models that prioritize fairness metrics and provide reliable explanations for AI-driven decision marking in healthcare and medical research.