Mission
Our mission is to utilize and integrate data-driven and knowledge-based models to define and solve various meaningful problems in medicine and biology. We deal with various types of data, including time series. To achieve this, we develop and utilize diverse methodologies, including statistics, control theory, optimization theory, as well as machine learning and artificial intelligence. From the science of discovering knowledge to the engineering of proposing new applications, we aim to address not only the unmet demands in clinical medicine and biology but also to identify and solve undiscovered needs. We explore in the areas of methodology research and practical field applications, ranging from prediction and inference to decision-making and implementation in the biomedical sciences and clinical fields.
Projects (On-going)
- Infant’s movement analysis by using of deep learning (including GNN, spatio-temporal analysis) to predict neurological outcome
- Development of digital twins to simulate glucose dynamics of diabetic patients
- Deep Learning-based survival analysis with multimodal clinical data
- Explainable medical image prediction models with prior knowledge
- Uncovering clinical interventions/behaviors via reinforcement learning