|
|
|
|
|
 |
|
 |
|
|
|
| |
|
| |
P-2
Advancing Medical Innovation and Health Care
via Biomedical AI Research
Yang C. Fann, Ph.D.
National Institutes of Health, USA
|
Recently, big data combined with machine learning (ML) and artificial Intelligence (AI) are dominating the field of biomedical research. Many data scientists from industries including pharma, startups, as well as academics and research institutions around the world are racing to explore the potential of utilizing AI to drive medical innovations and advance health care. It is believed that future data-driven biomedical discoveries and medical breakthroughs for better health outcomes can be substantially accelerated through ML/AI technology, for example, precision medicine. Although the technology of AI is fast evolving, there are many challenges in applying to biomedical disciplines.
In 2018, NIH has established a strategic plan for data science with commitments to the creation and stewardship of large biomedical datasets (e.g. repositories) to enable ML/AI research to accelerate the transitional applications in biomedicine. In addition, NIH has committed to FAIR principles, that is, making data Findable, Accessible, Interoperable, and Reusable to facilitate data sharing and collaboration while protecting data security and privacy. In this presentation, we will discuss and share the strategy and technical approaches in building large biomedical datasets to enable ML/AI research including examples from several NIH funded data repositories. The current limitations, including data availability, quality, ethics, standards, interoperability, and sharing, as well as lack of transparency and interpretability/explainability in ML/AI tool development will also be discussed. |
|
|
|
|
| |
|
|
 |
|
