Project summary

Project Sunshine is an AI-enabled platform that helps teachers identify early signs of learning and mental health conditions in children ages 5–10. Using short teacher–child video interactions, the system analyzes eye movement, posture, and voice patterns to flag potential risks like ADHD, ASD, and anxiety.

What are this project's goals? How will you achieve them?

We will achieve this through a systematic three-phase approach designed to ensure clinical validity and scalability.

1. We will expand our partnerships with schools across India to establish a robust testing cohort (~200 children, Pilot #2), providing access to diverse child populations. Simultaneously, we will benchmark our automated assessments against gold-standard clinical therapist evaluations to establish reliability and clinical concordance.

2. Our current model captures nine key behavioral biomarkers: eye contact percentage, blink rate, fixation rate, eye movement variance, fidgeting frequency, posture stability, head movement, and torso sway. Through comprehensive statistical analysis and error assessment, we will identify which biomarkers demonstrate the strongest predictive power for learning disabilities and mental health conditions. This analysis will drive targeted algorithm refinements to maximize diagnostic accuracy.

3. Following initial validation, we will conduct an adequately powered pilot study (~2000 children, Pilot #3) to confirm both scalability and clinical relevance across broader populations. This final phase will demonstrate real-world applicability and establish the foundation for clinical deployment.

How will this funding be used?

https://docs.google.com/spreadsheets/d/1nn7TYbgIqNAa9sxOI5Gw7d-NeE94AmnIqnPsGO_DBAM/edit?usp=sharing

Who is on your team? What's your track record on similar projects?

Gina Hafez, LPC, NCC, GCDF, is a licensed professional counselor with extensive experience in mental health and with a research background in long term jail studies. Gina helped co-create MentNav for EAs, is actively involved in co-launching Effective Mental Health (EMH). EMH runs community-wide events, a website, newsletter, communication channels, and a global fellowship now in development. Gina helps manage effective teams related to AI-Therapy efficacy, risk-management and advancement, LMIC interventions and improving the therapist-coaches network in EA. She also works on collaborative writing on key existential risk issues in mental health, event planning for building partnerships, and funding and founder matching.

Anand Jeevanandham brings 12+ years of experience across startups, enterprises, and non-profits, with expertise in AI, data analytics, process design, and team leadership. At Sunshine Labs, he founded an AI initiative for early detection of learning disabilities and built a team of AI researchers and mental health experts. At 91Springboard, he led BI and ERP teams, delivering scalable analytics, cost reductions, and process improvements. His skills in AI, workflow automation, data-driven decision making, and child mental health equip him to execute this project effectively.

What are the most likely causes and outcomes if this project fails?

Potential Cause for failure

1. An over estimation of current Computer vision capabilities to detect bio-markers.

2. Critical baseline data for measurement accuracy.

3. Limitations in cross cultural application due to homogeneity in training data.

Outcomes of failure

1. Increased awareness on the global gap in research and deployment for mental health diagnosis and interventions.

2. A road map to future research on multi modal analysis for diagnosing mental health conditions.

How much money have you raised in the last 12 months, and from where?

$3000, self funded