Towards the Next Generation of Clinical Decision Support: Overcoming the Integration Challenges of Genomic Data and Electronic Health Records

The wide adoption of electronic health records (EHRs), the unprecedented abundance of genomic data, and the rapid advancements in computational methods have paved the way for next generation clinical decision support (NGCDS) systems. NGCDS provides significant opportunities for the prevention, early detection, and the personalized treatment of complex diseases. The integration of genomic and EHR data into the NGCDS workflow is faced with significant challenges due to the high complexity and sheer magnitude of the associated data. This dissertation performs an in depth investigation to address the computational and algorithmic challenges of integrating genomic and EHR data within the NGCDS workflow. In particular, the dissertation (i) defines the major genomic challenges NGCDS faces and discusses possible resolution directions, (ii) proposes an accelerated method for processing raw genomic data, (iii) introduces a data representation and compression method to store the processed genomic outcomes in a database schema, and finally, (iv) investigates the feasibility of using EHR data to produce accurate disease risk assessments. We hope that the proposed solutions will expedite the adoption of NGCDS and help advance the state of healthcare