SMARTER Updates - October 2025
Greetings from the SMARTER team! Our project, funded by NIEHS and initiated in 2024, is developing community-driven, shared, generalizable metadata and data management tools that support reproducible exposure health research. Sensor metadata is a major focus.
Project Updates
Initiated user-centered design and development of the sensor metadata library interface. At this point in time, we are seeking exposure health researchers and staff to help us refine prototypes of the sensor metadata library through brief testing sessions.
Released an updated version of the Sensor Common Metadata Specification (v3.0).
Completed preliminary research in support of automated metadata extraction using LLMs.
Currently accelerating software development of the sensor metadata repository.
If you are able to contribute by participating in remote, user testing sessions where you would briefly test software designs or features, please let us know!
New Logical Model: Call for Comments
Sensor Common Metadata Specifications (v3.0) is now available in GitHub for public comment.
In this version we’ve incorporated multiple points of feedback from previous public comments and expert panel review.
We invite exposure health researchers and sensor developers to provide feedback on the updated version of the logical model and specifications. Your expertise is invaluable in ensuring the model and specifications are accurate, effective, and aligned with the latest advancements in the field. We encourage you to visit our public comment page for more detailed information and instructions on how to submit your comments.
Who might want to offer feedback? We would like to hear from anybody developing sensors or sensor informatics methods, using sensors in exposure health efforts, and/or developing informatics pipelines for exposure health. Your input will help shape the future of exposure health research, and we look forward to hearing from you!
How We Are Listening: Our User Requirements Journey
This graphic shows how we are incorporating user perspectives into the development of the SMARTER Sensor Metadata Library. Through stakeholder engagement, qualitative research, expert review, and user-centered design cycles, we are working to ensure the platform directly addresses the documented needs and challenges of environmental health researchers.
SMARTER Research at ISES-ISEE 2025
Over the summer, we headed to the ISES-IEEE 2025 meeting in Atlanta, GA to exchange new findings and innovations with the exposure health research community. Several members of our team, including Drs. Ram Gouripeddi, Mollie Cummins, & Fatemah Shah presented work from SMARTER. At the heart of all three presentations is a commitment to scalable, interoperable, and community-driven research tools that make exposure science more efficient and impactful. Here’s what our poster presentations at ISES-IEEE 2025 looked like!
-
In exposure health studies, detailed sensor metadata is vital—but often trapped in scattered and non-standardized sources like PDFs, websites, and technical catalogs. Manual metadata curation is time-consuming and difficult to scale. This presentation unveils an innovative Large Language Model (LLM)-based framework that automates sensor metadata extraction and harmonization.
Presentation highlights:
Metadata Diversity: Sensor specs differ across manufacturers and sources, lacking consistency or computability.
AI Solution: The team leveraged LLMs to identify and extract key sensor attributes—like deployment characteristics and operational settings—from unstructured documents.
Outcomes:
Improved accuracy and depth of extracted metadata.
Reduced manual effort in populating metadata repositories.
Next Steps:
Conducting detailed error analysis to identify gaps in extraction accuracy and assess the potential need for additional learning strategies
Expanding to more data types (webpages, catalogs) and refining accuracy through expert-curated comparisons
Cover additional entities defined in the team’s comprehensive sensor metadata model to enable broader and more scalable adoption in exposure health research.
-
As exposure health studies grow in complexity, researchers face mounting logistical challenges integrating multiple data streams—from ambient environmental conditions to participant physiology. This talk presents the user-centered design behind the Exposure Health Informatics Ecosystem (EHIE).
Presentation highlights:
Complex Research Contexts: Studies often involve sensors, clinical data, and social context—all requiring synchronization.
EHIE Vision: A generalizable, cloud-based platform that integrates environmental, biological, and behavioral data at scale.
Features:
Metadata- and event-driven architecture for real-time data operations.
Support for observational, interventional, and hybrid studies.
Big data principles applied to ensure reliability, simplicity, and scalability.
Goal: Build a reproducible informatics foundation for exposure research across institutions.
-
DescriptionTo truly enable FAIR (Findable, Accessible, Interoperable, Reusable) data practices in exposure research, the metadata challenge must be tackled not just technically—but collaboratively. This presentation emphasizes community engagement in building shared metadata infrastructure.
Presentation highlights:
The Challenge: Lack of metadata standards and tooling hinders reuse of sensor data in new studies.
The SMARTER Solution:
Community-advised logical models for diverse sensor types.
Designing a user-facing metadata repository with searchable libraries and submission portals.
Panel-Based Approach: Experts in exposure science, sensor engineering, and informatics guide the process.
Call to Action: This session invites feedback from the ISES-ISEE community to shape future development of interoperable metadata tools. text goes here
SMARTER’s Dr. Ram Gouripeddi discusses findings with Dr. Rima Habre of USC at a joint meeting of the International Society of Exposure Science/ International Society of Environmental Epidemiology in Atlanta, Ga, August 2025.
Welcome to new UX/UI Designer Urvi Varma!
We are so pleased to welcome Ms. Urvi Varma as a full-time member of the SMARTER team. With a background in architecture and a Master's in Human-Computer Interaction from the University of Maryland, she brings a unique perspective to designing data management tools that serve research communities. Urvi is driven by the challenge of transforming ambiguous problem spaces into elegant, user-centered solutions that empower scientists to do their best work.
SMARTER UX/UI Designer Ms. Urvi Varma.
Stay Informed - Join our Mailing List!
Stay updated as we issue future updates and opportunities for public comment. Join our mailing list here.
The University of Utah SMARTER Project is funded by the National Institutes of Health, National Institute for Environmental Health Sciences (“Community-Driven Sensor Metadata Ecosystem for Exposure Health“, 1R24ES036134-01, Multi-Principal Investigators Ramkiran Gouripeddi & Mollie R. Cummins). The content of this web site is solely the responsibility of the authors and does not necessarily represent the official views of the University of Utah or the National Institutes of Health.
