RAISE-AI
Resilience Across Intergenerational Stresses and Emotional Health through AI-Enabled Discovery
Lauren White · Children's Hospital of Philadelphia · United States
What they're working onRAISE-AI (Resilience Across Intergenerational Stresses and Emotional Health through AI-Enabled Discovery) aims to improve how we understand and identify early pathways of risk and resilience for childhood anxiety. We will collect and integrate rich multimodal data, including parent–child speech, facial expressions and behavior, cardiac and respiratory physiology, clinical assessments, and family contextual factors, from approximately 300 racially and socioeconomically diverse parent–child dyads with children ages 4–9. Using advanced AI and computational methods, we will examine how patterns of behavior and physiology unfold within parent–child interactions and identify dynamic processes associated with children’s emotional health. A key component will be evaluating and benchmarking computer-vision, automatic speech-recognition, and acoustic feature-extraction tools in pediatric populations, including their performance across skin tones, racial and demographic groups, and developmental characteristics. Caregivers, clinicians, and community partners will help ensure that our methods, interpretations, and outputs are meaningful, culturally responsive, and clinically useful. Ultimately, RAISE-AI seeks to develop scalable and equitable approaches for detecting early risk, identifying modifiable intervention targets, and supporting more personalized strategies to promote resilience in children and families.
AI approach
Multimodal AI · computer vision · speech analysis · physiological data
Mental health focus
Childhood anxiety · early risk and resilience
Links