TRIADS announces new round of seed grants
Eight transdisciplinary teams will leverage data science to address pressing societal questions
The Transdisciplinary Institute in Applied Data Sciences (TRIADS) has announced its newest crop of seed grant recipients, with eight teams of researchers receiving funding.
Featuring faculty from four different WashU schools (Arts & Sciences, Brown School, McKelvey School of Engineering, and the School of Medicine), these projects leverage data science to address pressing societal issues. Each research team includes faculty representing multiple academic disciplines, providing the knowledge base to view problems and their potential solutions from many different angles.
“We were extremely impressed with the quality of submissions for this round of seed grants,” said TRIADS co-director Bo Li. “These projects represent WashU faculty asking big, provocative questions and assembling high-quality, transdisciplinary teams to study them.”
Additional funding for the seed grant teams will be provided by WashU Here and Next, the McDonnell Center for the Space Sciences, McKelvey School of Engineering, and the Weidenbaum Center on the Economy, Government, and Public Policy.
A strategic initiative of the Arts & Sciences Strategic Plan, TRIADS provides funding, training, and resources to elevate WashU’s use of data science tools.
The projects receiving funding are:
Confronting the Next Decade of Data-Intensive Astronomy Ushered by LSST
Principal Investigator: Tansu Daylan (Physics)
Team Members: Nathan Jacobs (McKelvey School of Engineering), Soumendra Lahiri (Statistics and Data Science)
Astronomers at the Vera C. Rubin Observatory in Chile are about to launch a massive, decade-long study of the night sky called the Legacy Survey of Space and Time (LSST). By regularly imaging the southern hemisphere's skies, the LSST aims to advance our understanding of dark matter, among other scientific goals.
But LSST faces a serious challenge by the sheer number of artificial objects that orbit our planet, such as satellite constellations and space debris. These reflect sunlight and contaminate images taken by ground-based telescopes in the form of streaks, potentially interfering with the ambitious goals of the LSST. The TRIADS-funded team will simulate LSST images with streaks caused by artificial objects, assessing their expected impact on the imaging dataset and the implications for similar future studies.
In Vitro Neurotoxicity and Socio-Environmental Analysis for Mapping Alzheimer's Disease Risk Due to Particulate Matter Exposure
Principal Investigator: Joseph Puthussery (McKelvey School of Engineering)
Team Members: Rajan Chakrabarty (McKelvey School of Engineering), John Cirrito (School of Medicine)
Can increased exposure to fine particulate matter contribute to the risk of developing Alzheimer’s disease? This team will study this potential link through a multidisciplinary approach that integrates neurotoxicological research, machine learning, socioeconomic data analysis, and more. The team plans to produce an Alzheimer’s Disease Health Risk Map for the United States, highlighting hotspots with elevated levels of particulate matter and associated Alzheimer’s disease risks, emphasizing environmental justice and socioeconomic disparities while advancing understanding of how air pollution impacts neurological health.
Integrating Geographic Information Systems and Electronic Health Records for Scalable Real-World Evidence Generation: A Case Study on Opioid Use Disorder
Principal Investigator: Linying Zhang (School of Medicine)
Team Members: Devin Banks (School of Medicine), Nan Lin (Statistics and Data Science), Min Lian (School of Medicine), Chenyang Lu (McKelvey School of Engineering)
With opioid overdoses on the rise in Missouri, this project seeks to develop a more holistic view of the crisis by combining geospatial data with electronic health records. In particular, this research team wants to consider the demographic and socioeconomic factors that could contribute to disparate outcomes for people suffering from opioid addiction. By considering the bigger societal picture, the project could potentially develop better tools for determining treatment effectiveness, leading to better interventions for all patients.
Investigating Geographic Disparities in Social and Environmental Determinants of Hypertension in the Greater St. Louis Area
Principal Investigator: Lindsay Underhill (School of Medicine)
Team Members: Jenna Ditto (McKelvey School of Engineering), Kenan Li (Saint Louis University)
Hypertension — or high blood pressure — is a major risk factor for cardiovascular disease, and disproportionately affects older, lower-income, and minority communities. This project will build a geospatial model to illustrate the distribution of hypertension and related social determinants of health across St. Louis, melding data from electronic medical records, air quality measurements, and estimated travel times to healthcare facilities with people’s perceptions of their neighborhood conditions and healthcare availability. This comprehensive study could potentially identify St. Louis neighborhoods most at risk for developing hypertension, leading to targeted interventions and better health outcomes for residents.
Machine Learning Using Cardiotocography and Other Intrapartum Data to Predict Birth Outcomes
Principal Investigator: Christopher Ryan King (School of Medicine)
Team Members: Chenyang Lu (McKelvey School of Engineering), Michael Dombrowski (School of Medicine)
Electronic fetal monitoring is a common practice for high-risk pregnancies, allowing doctors to track a baby’s heart rate and other vital signs moment by moment throughout the labor process. But its effectiveness is questionable — it has not been shown to reduce negative outcomes, and its subjective interpretation by physicians can result in interventions like unnecessary cesarean deliveries. This project will use machine learning to make electronic fetal monitoring more objective and effective for physicians, studying the data of about 13,000 deliveries completed between 2019 and 2023.
By analyzing this data with machine learning models, the team aims to improve health outcomes for mothers and babies, particularly those in underserved minority populations.
See the full list of recipients here.
