The National Library of Medicine (NLM) supports innovative research aimed at advancing biomedical informatics and data science. Biomedical informatics applies theories and analytical processes or methods to data to improve decision-making and human health. The NLM strategic plan outlines a platform for biomedical discovery and data-powered health, integrating streams of complex and interconnected data that can be translated into scientific insights, clinical care, public health practices, and personal wellness. NIH defines data science as the interdisciplinary field of inquiry in which quantitative and analytical approaches, processes, and systems are developed and used to extract knowledge and insights from increasingly large and/or complex sets of data.” Research problems that can be addressed with biomedical informatics and data science are broad, but should align with NLM's focus for the acceleration of data-driven discovery by the advancement of human health using the exposome (from the intracellular environment to the built environment), broadening analytics across heterogeneous data sources including natural language processing and deep learning, and increasing computable biomedical knowledge,(e.g., diagnostics), and decision-analytic models.
The NLM strategic plan reinforces the need to accelerate discovery by enhancing health through data-driven research. Applications proposed to NLM should align with the strategic plan. Proposals should emphasize novel methods to foster data driven discovery in biomedical and clinical health sciences that are domain-independent, reusable/reproducible and use FAIR (Finable, Accessible, Interoperable, Reusable) standards for increased harmonization.
NLM supports innovative research projects focused on biomedical data that combine elements of computer science and information technology to optimize the use of information and technology to improve individual and public health and biomedical research. Research areas of interest to NLM include, but are not limited to:
- Development of novel approaches enabling analysis and discovery at scale across biomedical domains and health care sectors, including those leveraging high-performance cloud computing and federated learning
- Development and demonstration of innovative informatics methods and data science techniques for informing biological, clinical, public health, and social science research.
- Computational approaches integrating structured and unstructured data, natural language processing, automated metadata assignment.
- Advanced information retrieval and knowledge discovery from very large and/or heterogeneous data sets
- Multi-level, reusable, data analytic models, simulations, information visualization, and presentation approaches to enhance decisions, learning or understanding of biological and clinical processes
- Approaches to assess and address algorithmic bias and/or fairness and health equity
- Innovative analytic methods to advance decision support that are generalizable within and across underserved populations
- Applying natural language processing to unstructured health-related data, including Electronic Health Record (EHR) data, to increase provider-patient health care understanding
- Informatics approaches that translate basic biomedical research to clinical methods to support patient and provider decision making
- Data science methods and approaches that enhance the quality, security, understandability and utility of data, information, or knowledge related to health and biomedicine
- Informatics methods and approaches to improve public health and population-level health outcomes
- Using biomedical informatics and data science to address health disparities and health equity
Research in biomedical informatics and data science is inherently multidisciplinary, including mathematics, statistics, information science, computer science and engineering, and social/behavioral sciences. Applications that propose team science approaches are encouraged. NLM expects that investigators will employ rigorous, scientifically defensible research techniques leading to sound empirical and reproducible evidence. These techniques may include quantitative and qualitative approaches, in silico experiments, simulation studies, model generation and testing, computer-based analytical techniques supporting clinical and non-clinical decisions through novel uses of computational analytics, text mining and natural language processing, network inference and pathway analyses, ontologies, and other advanced approaches. For NLM support, a research project's innovation should be centered in the development and testing of novel data science or biomedical informatics methods and approaches.
None Available.
Applications submitted to this funding opportunity should focus on a well-defined research problem, a rigorous research design, based on preliminary studies, and advance the field of informatics or data science to improve human health. NLM will consider supporting projects where the primary focus of informatics or data science is applied to a clinical or disease domain when the approach is novel and will benefit findings in the domain, although priority will be placed on funding applications that propose to develop tools and approaches that can be reproduced, generalized, and scaled to ensure maximum benefit is achieved. NLM will not support infrastructure or product development or continued development of existing software tools or knowledge resources as an endpoint of research funded through this FOA.
Potential applicants are strongly encouraged to discuss their proposed project with one of the Scientific/Research Contacts listed in Section VII for advice about the application process and suitability of the project for support by NLM.