Radiograph Ready

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Radiograph Ready

Inspiration

Before an AI assistant interprets a dental radiograph, it should know whether the image is usable at all. A fluent explanation is not the same as a reliable quality decision.

What we built

Radiograph Ready is an interactive research prototype for image-quality triage before diagnosis. It presents pilot context and model-comparison findings, then lets users adjust five transparent criteria: anatomical coverage, positioning, exposure and sharpness, motion and artefacts, and overall interpretability. The illustrative score changes live so students and reviewers can see the rubric rather than trust a black box.

Health impact and rigor

The project supports students, researchers, and dental teams by making calibration visible. It does not diagnose disease, rank treatment, or accept patient files. The contribution handoff prepares an email conversation and explicitly asks contributors to de-identify data until a secure endpoint is approved.

What we learned

A health prototype can be more trustworthy when it shows its boundary conditions. We kept the interaction deterministic, documented the safety limits, and included a one-page description and code overview PDF for reviewers.

Built With

• accessible-html/css
• dental-imaging-research
• next-compatible-static-export
• react-19
• typescript