You are here:
Smart-NanoEHS: Strategic Machine Learning and Artificial Intelligence Use for NanoEHS Discovery
Citation:
Mortensen, H., A. Kumari, R. Bjorkland, K. Dellinger, AND M. Zarate-Bermudez. Smart-NanoEHS: Strategic Machine Learning and Artificial Intelligence Use for NanoEHS Discovery. The International Society of Exposure Science (ISES) and the International Society of Environmental Epidemiology (ISEE), Atlanta, GA, August 17 - 20, 2025.
Impact/Purpose:
The coordinated nanotechnology research and development efforts in the United States have and continue to make substantial investments in the responsible development of nanotechnology. We discuss potential advances, opportunities, challenges, and limitations of the use of artificial intelligence (AI) and machine learning (ML) tools to retrospectively analyze the nanotechnology environmental health and safety (nanoEHS) data generated since the early 2000s. We also explore the potential for organized crowdsourced solutions to use AI to tackle the problem of distinct data types, nanomaterials, and formats of nanoEHS data and to expand the availability of data for robust decision-making. Leveraging existing data to extract novel findings and insights could advance our understanding of the factors that drive uncertainty in nanoEHS measurements, support and improve rigorous data quality assessment and extraction, and boost predictive risk assessment, pattern recognition, and the development of probabilistic models. These efforts may accelerate the development of data-driven minimum characterization standards, upgrade prevention-by-design tools for new materials, and spur greater use of new approach methodologies for risk reduction, dosage determination, process refinement, and replacement of current testing models, when feasible.
Description:
This revised abstract for the upcoming ISES/ISEE Anjali Kumari (Joint School of Nanoscience & Nanoengineering, Greensboro, NC)Rhema Bjorkland (National Nanotechnology Coordination Office, Alexandria, VA) et al:Placeholder Title: Using AI to unlock the power of retrospective nanoEHS data Abstract:The coordinated nanotechnology R&D efforts of the United States have and continue to make substantial investments in the responsible development of nanotechnology. This presentation discusses the opportunities, challenges, and limitations of the use of artificial intelligence (AI) and machine learning (ML) tools to retrospectively analyze the nanomaterial environmental health and safety (nanoEHS) data generated since the early 2000s. This presentation will explore the potential for organized crowdsourced solutions to use AI to tackle the problem of the diversity of data types, materials, and formats of nanoEHS data and to expand the availability of data for robust decision making.Leveraging existing data to extract novel findings and insights could advance our understanding of the factors that drive uncertainty in nanoEHS measurements, boosting predictive risk assessment and the development of probabilistic models. These efforts may possibly accelerate the development of data-driven minimum characterization standards, boost prevention-by-design tools for new materials, and spur greater use of new approach methodologies (NAMs) in the reduction, refinement, or replacement of animals in testing where feasible.