News

Researchers harness AI to study family involvement with child protection systems

WashU researchers Eunhye Ahn and Yevgeniy Vorobeychik are teaming up to study the multifaceted dynamics of families involved in child protection systems. The interdisciplinary team, combining expertise in social work and computer science, is supported by a Here and Next seed grant.

Traditionally, child welfare research has been narrow in scope, “primarily focused on individual children without fully considering the broader context of family dynamics,” said Eunhye Ahn, an assistant professor at the Brown School. But what if we could study a more complete picture of children involved in the child welfare system to ensure the best possible outcomes for them and their families?

Ahn’s current research project, titled ‘AI for Examining and Predicting Family Involvement with Child Protection Systems,’ aims to develop simulation models that provide invaluable insights into the multifaceted dynamics of families involved in child protection systems. “AI-driven simulation modeling helps us understand various factors and how they interact,” said Ahn, “Such as older siblings’ involvement with the system and their association with the younger child’s experiences.”

With funding from the Digital Transformation priority area of the Here and Next seed grant program, the interdisciplinary team is co-led by Ahn and Yevgeniy Vorobeychik, a professor of computer science and engineering at WashU’s McKelvey School of Engineering. By combining their expertise, the team aims to eventually develop more comprehensive and effective interventions for families and children.

While AI can help process vast amounts of data, Ahn describes how the human element remains indispensable in daily decision-making. “We hope our research can offer some insight into understanding how the system operates so that we can inform our human decision-makers when they navigate complex systems.”

The project is in its early stages, with the team working to analyze extensive child welfare involvement data from across the country. Using current understandings of how family dynamics unfold over time, they aim to fine-tune their simulations to more accurately reflect the lived experiences of families involved in child welfare cases.

However, this research may not have gotten off the ground without the support of the Here and Next seed grant. “The Here and Next seed grant encouraged us to actually put together a project proposal,” said Ahn. Beyond the critical financial support, the grant’s requirement to include faculty members from two schools sparked the collaboration between Ahn and Vorobeychik, merging expertise in social work and computer science to tackle this interdisciplinary challenge. “The grant was especially meaningful for me as an early career faculty member,” added Ahn, who gained invaluable learning experiences working alongside Vorobeychik.

Building on the momentum from the Here and Next seed grant, the team is now setting its sights on the future with ambitious plans. They aim to incorporate more comprehensive data, such as neighborhood information and family histories, to further improve their simulation models. They are also eager to share their findings with both researchers and practitioners in the field, hoping to positively influence real-world practices. Additionally, the foundational work supported by the Here and Next grant has positioned the team to seek external funding to scale their research further.

Here and Next seed grants support interdisciplinary research activities at WashU across disciplines, departments, and schools in priority areas such as Digital Transformation. WashU researchers interested in applying for one of the four core grants available to faculty members can learn more at the OVCR’s website or contact the DI2 Accelerator at di2accelerator@wustl.edu for proposal writing assistance, researcher matchmaking, or data and software engineering support.