Health-AI Decision Support Tools
Examines how healthcare AI systems often fail to account for the socio-technical complexities present in real clinical workflows.
Research
How people understand, trust, and work with AI in high-stakes decisions — and how interface design reshapes judgment, expertise, and responsibility.
Overview
The NU-PATH Lab studies human-AI interaction in consequential, high-impact settings. Healthcare — where predictive models and AI recommendations are increasingly woven into clinical judgment — is the lab's primary testbed, but the underlying questions apply to high-stakes decision-making more broadly. The lab's work spans three connected themes:
- AI-assisted decision-making — designing how people interact with AI recommendations and predictions in real, high-stakes, time-constrained settings, such as clinicians weighing an ML-driven risk score.
- Trust, expertise, and invisible labor — understanding how AI changes what expertise looks like, what new work it creates for the people using it, and how interfaces can support rather than override human judgment.
- Community-centered AI design — co-designing AI-based tools directly with the people they're meant to serve, from frontline workers to patients and families, so systems reflect real needs and context.
Current projects
Examples of these themes in the lab's current work:
CASMI: Co-designing Patient-Facing Machine Learning for Prenatal Stress Reduction
Includes patients directly in the development and design of machine-learning prediction tools for prenatal stress.
Co-designing Interventions for Burnout in Frontline Gender-Based Violence Work
A community-based participatory project co-designing interventions with frontline workers at nonprofit gender-based-violence survivor service organizations.
EquityWare: Designing Equitable Wearables
Democratizing wearable technologies by developing and co-designing inclusive wearables with and for marginalized groups.
Understanding Self-Reflection Practices
Researching scalable, effective interventions in response to growing mental health concerns.
Digital Mental Health Interventions for Youth Learning CS
Designing strategies and interventions that support youth's mental health within computer science education environments.
Lab advising
Maia advises graduate and undergraduate researchers through the NU-PATH Lab. Current and recent students:
PhD students
- Monisola Jayeoba — Technology and Social Behavior
- Mara Ulloa — Computer Science
- Connie Chau — Technology and Social Behavior
- Hannah Studd — Media, Technology, and Society
- Stefany Cruz — Computer Engineering
Postdocs
- Sooyeon Jeong — Center for Behavioral Intervention Technologies Fellow — now Assistant Professor, Purdue University
Master's & undergraduate students
- Niharika Gopinath Menon — MS Computer Science, Northwestern (Class of 2024)
- Maya Blumovitz — MS Computer Science, Northwestern (Class of 2023)
- Yuanhui Huang — MS Computer Science, Northwestern (Class of 2023)
- Luke Malaisrie — BS Computer Science, Northwestern (Class of 2025)
- Angela Davis — BS Computer Science, Northwestern (Class of 2024)
- Will Finigan — BS Computer Science, Harvard (Class of 2021)
- Jeffrey He — BS Computer Science, Harvard (Class of 2020)
Team science
Beyond lab-led projects, Maia serves as lead HCI researcher and co-investigator on several large, interdisciplinary NIH-funded initiatives, including the National Cancer Institute's ACCERT Patient Navigation program (PIs: Melissa Simon, Betina Yanez), the STELLAR telehealth cancer-care program (PIs: Sofia Garcia, Brian Hitsman, Siobhan Phillips, Bonnie Spring), and NIMH's Technology Enabled Services for Coordinated Care of Depression program (PI: David Mohr).