§  Home

§  Research

§  Publications

 

CV

 

 

 

 

 

 

My research is data fusion and statistical machine learning intersecting with systems or domains having complex data structures, such as high dimensionality, heterogeneity, multi-source, multi-level, and multi-task.

Methodological development

·       Transfer learning

·       Sparse learning

·       Graphical models

·       Semi-supervised learning

·       Model fusion (data-driven and mechanistic models)

Application areas

Health and medicine

·       Health care data analytics

Technology advances in diagnostic imaging, sequencing technologies, Electronic Health Records (EHR), and remote and smart sensing have created a data-rich environment in health care. This provides a great opportunity for Precision Medicine (PM), i.e., to provide the right treatment to the right patient at the right time. On the other hand, there are enormous data science challenges to be tackled to best leverage the existing data resources. My research is driven by emerging challenging problems faced by PM and develops statistical models and machine learning algorithms to tackle the data science challenges to provide decision support for diagnosis and treatment.

Example Projects

o   Integration of semi-supervised learning and mechanistic models for radiomics mapping of brain tumor

o   Positive transfer learning for telemonitoring of Parkinson’s Disease

o   Collaborative learning of modality-wise-missing multi-modality imaging data for Alzheimer’s Disease early detection

o   Hierarchically-structured factor models for sub-classification of migraine using multi-modality images

o   Sparse clustering and reproducibility analysis for subtyping of traumatic brain injury using multi-faceted datasets

·       Health system informatics

Health care improvement should integrate a patient view (i.e., PM) with a system view. The latter aims to optimize the operational aspect of the system by improving quality and efficiency and reducing cost, which provides the necessary infrastructure to enable the PM.

Examples:

o   Integration of multiple health information systems for quality improvement of radiologic care

o   Multi-biomarker test sequence optimal frontier for Alzheimer’s Disease

Engineering systems

·       Process data mining for manufacturing quality improvement (semi-conductor, metal forming, etc)

·       Monitoring and anomaly detection in large communication networks