curaJOY, a tech-forward nonprofit addressing global behavioral health inequities, recently captivated audiences at the 23rd IEEE/WIC International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT 2024). Presented during the AI for Social Good Workshop, their study, “Using LLMs to Analyze Antecedent, Behavior, and Consequence Narrative Recordings in Behavioral Health Science,” revealed how AI can revolutionize Functional Behavior Assessments (FBAs).

The research, which will be published in the IEEE journal, demonstrated how curaJOY’s cutting-edge AI models outperform traditional methods in automating FBAs, achieving 82% accuracy in identifying antecedents and 72% in labeling consequences—compared to the current clinical average of 50%. According to recent guidance from the U.S. Department of Education, FBAs play a crucial role in creating supportive learning environments by identifying and addressing the environmental and behavioral factors that trigger these behaviors. curaJOY’s AI-driven solutions advance this practice by automating key components of FBAs, significantly improving their efficiency and accuracy. By leveraging data-driven methodologies, curaJOY aims to expand access to behavioral health support for marginalized communities worldwide, addressing systemic inequities and empowering educators and clinicians to better meet students’ needs.

Reflections from the Presentation: A Milestone Moment

curaJOY’s presentation drew a full house of machine learning and behavioral health professionals, sparking a dynamic discussion about the future of AI in healthcare. curaJOY’s approach leverages Large Language Models (LLMs) like GPT-4 to classify antecedents and consequences from ABC Narrative Recordings, a foundational method in behavioral health science. The session highlighted the success of techniques like Chain-of-Thought (CoT) Prompt Engineering, which guided AI models to reason step-by-step, dramatically improving accuracy and reliability.

Jeevan Redd, curaJOY’s Senior Researcher and key presenter, shared his experience: “The diversity of presentations, ranging from Bayesian statistics to LLMs at the WI-IAT 2024 AI for Social Good workshop, was impressive. curaJOY’s presentation stood out because it uniquely combined deep domain expertise with technical innovation, demonstrating practical applications of AI in behavioral health. The audience’s interest was evident, with methodology sparking the highest number of follow-up questions during the session.”

“The audience’s engagement reinforced the urgency of this work,” said Caitlyn Wang, CEO of curaJOY. “Their insights will shape our next steps in scaling these solutions.”

Why It Matters: Addressing a Global Behavioral Health Crisis

Behavioral health expertise remains out of reach for millions worldwide. curaJOY’s research underscores the stark reality that only 4.7% of Board-Certified Behavior Analysts (BCBAs) practice outside North America, leaving entire regions without sufficient resources to address developmental disabilities and youth mental health challenges. Even in the U.S., where funding is limited, 40% of public schools lack adequate staff to conduct FBAs.

“curaJOY’s AI-driven solutions are designed to close these gaps, providing culturally adaptive, scalable tools that empower clinicians and families,” emphasized, Haley Scheer, curaJOY’s Digital Health Lead. “This is not just about automating tasks—it’s about democratizing access to behavioral healthcare worldwide.”

curaJOY invites educators, healthcare providers, and researchers to collaborate to advance equitable behavioral healthcare. Learn more about the nonprofit’s groundbreaking work, explore partnership opportunities, or support their mission by visiting https://curajoy.org.

304 South Jones Boulevard
#2986
Las Vegas
NV
89107
United States

Information contained on this page is provided by an independent third-party content provider. XPRMedia and this Site make no warranties or representations in connection therewith. If you are affiliated with this page and would like it removed please contact [email protected]