Skip to content

RetinaMind

advancedNational

Edward Kang · Hackensack, New Jersey
Second Place ($175,000) · Regeneron Science Talent Search · 2026

Abstract

An AI screening tool that reads retinal images for patterns associated with autism and ADHD, paired with cellular models built to examine the genetic factors that might explain the association.

Why it worked
ResearchForge's reading of the public record — not the students' words, and not the judges' reasoning.

The cellular work is what separates this from a classifier that finds a correlation nobody can explain. Building a mechanistic arm alongside the model is the answer to the "why would that even work?" question a judge will ask.

Key methods
  • Retinal image analysis
  • Deep-learning classification
  • Cellular disease modelling
  • Genetic factor investigation
What to take from this
ResearchForge's reading of the public record — not the students' words, and not the judges' reasoning. Borrow the habits and decisions, never the project itself.
  • If your model finds a surprising signal, build the arm of the project that explains why the signal exists
  • Screening claims need care about who the training population was — that limitation is worth stating first

Archive entries are summaries compiled for study. Read them for how a project was structured, argued, and defended — not as a template to reproduce. Copying someone else's project is the one thing that will end yours.

Compiled by ResearchForge from the Society for Science public award announcement linked on this record. Names, hometowns, project titles and reported results are as published. The "why it worked" and "what to take from this" notes are ResearchForge's editorial reading of that public record — they are not statements by the students and not the judges' rationale. All rights to the original projects remain with their authors; no project materials are rehosted here.