AUGUST 13, 2024 – Chemical and biological (CB) data is a strategic asset that can be used to inform decision making within the Department of Defense’s (DoD) Chemical and Biological Defense Program (CBDP). However, reliably applying CB datasets to rapidly develop data-informed decision-support capabilities that allow the Joint Force to understand and mitigate CB threats remains a challenge.
Data science as a strategic asset has gained increased traction across the government with the release of policy, including Executive Order 14110 on Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence (October 30, 2023) and the DoD Data, Analytics, and Artificial Intelligence Adoption Strategy (June 27, 2023). Therefore, the Defense Threat Reduction Agency’s (DTRA) Chemical and Biological Technologies Department in its role as the Joint Science and Technology Office (JSTO) for Chemical and Biological Defense, an integral component of the CBDP, has invested in leveraging artificial intelligence and machine learning (AI/ML) to explore and characterize CB threat data. Among a myriad of other use cases, some of the data will be used to more accurately predict new biomarkers of CB exposure for early detection and diagnostics and to optimize developing new medical countermeasures.
While these initial investments in AI/ML have enabled better understanding of the opportunity areas, they have also illuminated some of the fundamental challenges associated with applying AI/ML technologies to CB data as well as larger fundamental data science challenges across technology areas. Although there is no one-size-fits-all approach, creating a strong data science framework for CB data within DTRA JSTO is a key component of a successful CBDP-wide data science enterprise.
To better understand the current data science landscape, DTRA JSTO held a Data Science Workshop that included subject matter experts in both computational and CB science fields from across the government, national labs, federally funded research and development centers, and academia. The overall objectives of this workshop were to:
- Explore new, state-of-the-art technology areas within the field of data science
- Highlight gaps and challenges remaining in this evolving, rapidly changing data science field
- Build an informed path for FY25 DTRA JSTO basic research topic in data science
The workshop included presentations by invited speakers and group discussions. Four speaker presentations covered data standardization, provenance, harmonization, integration, data management and security, as well as community sharing of data. The workshop focused on discussing the regulatory requirements for the quality of data and the challenges faced with the compliance guidelines for rapidly developing medical countermeasures.
There was a shared understanding among participants that the complexity and variability of CB data types is extensive and the current frameworks for data formatting are too varied to harness it for maximum benefit and value across DTRA JSTO. There is an urgent need for DTRA JSTO, along with the broader data science community, to establish a standardized framework to ensure that reliable, reproducible, and re-usable data is available to develop capabilities to treat and protect the Joint Force. Ideally, standardizing specific data and datasets would allow it to be more efficiently validated, analyzed, and re-used.
Other recommendations from the workshop included employing incentive structures and acquisition vehicles to facilitate re-using and sharing data among the CB community and promoting data standards to build trustworthy partnerships between the DoD and industry. Developing a standardized framework for DTRA JSTO’s current efforts will help build a roadmap to inform future research and development investments in data science to produce technology that protects the Joint Force, nation, and our allies.
POC: Sweta Batni, Ph.D., sweta.r.batni.civ@mail.mil
Courtesy Story
Defense Threat Reduction Agency’s Chemical and Biological Technologies Department