NSF Award
2408401
Artificial Intelligence (AI) is making huge strides across a range of areas, including natural language processing, computer vision, reinforcement learning, and other machine learning areas. Researchers at the University of Illinois Chicago (UIC) are uniquely positioned to lead research at the nexus of AI and healthcare using the vast amount of data available from the UI Health system. The Graphics Processing Unit (GPU) Cluster for Accelerating Health Insurance Portability and Accountability Act (HIPAA)-Compliant Data-Driven Research enables the development and use of computationally-intensive artificial intelligence (AI) models for a variety of health-related research projects and applications at the University of Illinois Chicago. The project’s novelties are its research capabilities (HIPAA-compliant data analysis and modeling) in applying AI to sensitive healthcare data within the field of computer science. The project's broader significance and importance is refining health treatments and improving health outcomes for individuals from groups that are not as well-represented in other university-associated health systems. This instrument enables leading-edge, cross-disciplinary research on AI and health/medicine, and is a key catalyst for accelerating AI’s progress in providing health benefits to the diverse population served by University of Illinois Health.<br/><br/>This project uses six mini-supercomputers and just under two petabytes of storage, creating a catalyst for AI and Machine Learning research, designed for the handling of sensitive data, and purpose-built for simulation, data analytics, and AI. The software-defined storage is designed to meet the storage requirements of HIPAA data, offering features like encryption, data deduplication, and data-at-rest protection. The instrument is the foundation for transformative research, including: knowledge extraction from electronic health records, language-guided health monitoring and assessment, opioid risk detection, ophthalmic disease detection, prognostication of age-related macular degeneration, and cancer detection using graph neural networks.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
