High-Performance Computing and Artificial Intelligence Drive Clinical Applications of Precision Medicine at Scale

high-performance-computing-and-artificial-intelligence-drive-clinical-applications-of-precision-medicine-at-scale

The volume and growth of biomedical data is set to exceed other data-intensive sectors, so the analysis of large data sets and development of decision support tools to counter the challenge of cognitive overload will increasingly rely on advances in data science and automated algorithms generated by deep learning and artificial intelligence (AI). Although the science and technology underlying precision medicine has yet to become mainstream in clinical practice, health systems in the United States have started to develop information architectures to manage data flows of the magnitude and diversity involved with precision medicine. The following cases reported in The Journal of Precision Medicine of early adopter healthcare providers highlight the challenges, models, and strategies that have been used by a representative sample of healthcare provider organizations to support the integration of precision medicine in clinics:

  • Mission Hospital for Advanced Medicine (Asheville, North Carolina) is a $400 million investment that includes the integration of genomic knowledge into precision medicine as a standard of care through embedding warnings, triggers, and results into the electronic health record system. The precision medicine initiative is part of larger system transformation in which emerging technologies – genomic, digital, robotic, AI – will come together to provide a greater capacity for prediction and prevention. As one of the few community health systems in the U.S. to develop a personalized medicine program, the initiative aims to be “best in class” for cutting-edge technology in a community hospital system and to serve as a role model for other systems.
  • Partners Healthcare Personalized Medicine (Cambridge, Massachusetts) encompasses full genome and exome sequencing and analytical capabilities and the information technology (IT) applications necessary to facilitate the dissemination of related insights into clinical practice. The investigative and clinical work at Partners is backed by an information technology infrastructure that includes GeneInsight and GIGPAD. The GeneInsight Clinic platform, which can be deployed as a standalone genetic report management system or integrated into an electronic health record, provides the infrastructure that enables analysis, interpretation, and reporting to clinicians of genetic test results, and updates as the state of clinical knowledge changes over time.
  • The Mayo Clinic Center for Individualized Medicine (Rochester, Minnesota) supports the translation of genomics discoveries into clinical practice. The Center’s infrastructure spans next-generation sequencing and the identification of genomic variants and applications for validating and optimizing new laboratory tests, to electronic medical record capabilities to order and review genomic-based tests and results, as well as to access new and improved rules-based decision-support tools for pharmacogenomic drug and gene-variant combinations that may influence treatment and dosing. The Center has joined Oracle’s Strategic Development Partnership to collaborate on expanding Oracle’s Translational Research Center product to centrally manage trillions of unique genetic variants for Mayo Clinic patients and to provide a single infrastructure that can support individualized medicine activities spanning research and clinical practice, complete with privacy and security controls.
  • The Moffitt Cancer Center’s DeBartolo Family Personalized Medicine Institute (Tampa, Florida) provides the organizational hub for personalized care services and research at Moffitt. Initiatives include an effort to develop innovative and integrated tools and services that apply knowledge of personalized medicine to the clinic. Research-related initiatives include the Total Cancer Care partnership between 120,000 patients, doctors, and researchers to improve all aspects of cancer prevention and care. The initiative uses patient biosamples to generate information on cancer and how care can be improved, as well as the Clinical Pathways model for integrating evidence-based medicine with available technology to standardize, benchmark, measure, and improve cancer care in 51 disease-specific areas.
  • Inova Center for Personalized Health (Falls Church, Virginia) is a new initiative that integrates genomic research for patient care, prevention, and wellness. One of its three initiatives, the Inova Translational Medicine Institute (ITMI), is a multidisciplinary research and development organization pursuing the application of genomic and clinical information from individuals to innovative methods for personalized healthcare. The ITMI has partnered with Cloudera Enterprise to apply high-performance SQL analytics and machine learning to clinical and genomic data at unprecedented speeds and scale. The Cloudera platform has reduced the time for end-to-end genomic data analysis from a couple of months to one week, with the goal of reducing that in the future to just hours.
  • The Geisinger Genomic Medicine Institute (Danville, Pennsylvania) supports the research, development, and clinical application of innovative approaches in genomic medicine. Research initiatives include MyCode Community Health, which is sequencing the genomes of more than 160,000 Geisinger patients and combining that information with each participant’s medical record. Geisinger is also a member of the Electronic Medical Records and Genomics (eMERGE) Network that uses biobanks in combination with electronic medical records to advance knowledge discovery and applications to clinical care. The Department of Biomedical and Translational Informatics’ high-performance computing (HPC) core architecture specializes in genomic analytics, machine learning, data visualization, predictive analytics, image analytics, and advanced decision support algorithms to apply enterprise data analytics frameworks to advanced analytical applications.
Comments are closed.