Project summary

We aim to spot learning disabilities and mental health risks in schoolchildren using a 2-minute classroom video. A machine learning model looks at eye contact, hyperactivity and speech patterns to flag high-risk cases early. With just a phone camera and simple teacher prompts.

https://cold-waxflower-9ec.notion.site/Project-Sunshine-1c9575a9db7180b9bc3cfe0774b6db0d

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

Our goal is to triage high-risk cases of learning and behavioural disorders such as SLD, ADHD, ASD. We will achieve this by combining computer vision models with clinical consultation to build and refine algorithms that can be deployed in schools with minimal training or infrastructure.

How will this funding be used?

Pilot in rural/urban schools, train teachers and to refine the model.

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

I’m Anand Jeevanandham, a business intelligence professional with 10+ years of experience leading data and systems projects in startups. linkedin.com/in/anand-j-aj

We working with clinical psychologists and special educators to co-design this solution, with outstanding support from our volunteer team of ML specialists.

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

The main risk is insufficient video diversity during training, which could affect accuracy across regions. If unaddressed, this might lead to false positives/negatives. The worst-case outcome is a return to status quo, where children continuing to be missed by current systems.

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

Around $30,000 has been self-funded. In addition, the project has received over 200 hours of pro bono support from machine learning engineers and mental health specialists.