Codédex VS Read Your Scan

Read Your Scan's answer:

Most tools that "explain your scan" only read the text of the radiology report. Read Your Scan reads the images.

You drop in the DICOM folder from your hospital CD (or a NIfTI file) and get two things at once: a real 3D viewer of your own scan in the browser, and a plain‑language report where every finding is pinned to the exact series, slice and coordinate where it was seen. Click a finding and the crosshair jumps there. If the marker is off, you can correct it and it re‑saves.

Read Your Scan's answer:

The alternatives fall into two camps, and neither closes the loop.

General AI chatbots will happily explain your report, but they can't open your scan. They work from a paragraph of text, so they can never show you where — and pasting your report into a general chatbot is a data decision most people haven't really made.

Professional viewers and PACS tools open the images beautifully, but they were built for radiologists. They explain nothing, and you need a license.

Read Your Scan's answer:

People who just had an MRI, CT or X‑ray and are holding two things they can't use: a CD full of DICOM files no consumer computer will open, and a report written for another doctor.

The sharpest segment is spine patients — cervical and lumbar — because their reports are dense with level‑by‑level findings ("C5‑C6 disc protrusion with mild canal stenosis") that mean everything to their prognosis and nothing to them. Around them: family members and caregivers reading a relative's scan, people preparing for a second‑opinion consultation who want to arrive with questions, and patients in the gap between the scan and the follow‑up appointment — often two or three weeks of silence.

There's a meaningful secondary audience of students and non‑radiologist clinicians who use the free viewer as a fast way to open a study.

Geographically it's global rather than US‑centric — the product ships in 9 languages because a large share of the demand isn't English‑speaking.

Read Your Scan's answer:

It started as a free 3D DICOM viewer — a small tool to solve an absurd problem: hospitals hand you a CD of your own body and no way to look at it.

What turned it into a product was watching what patients already do. In spinal‑stenosis support groups, people post photographs of their MRI to hundreds of strangers with the caption "is this bad?" They're not being reckless — they've been handed a document about their own body that they can't read, and they'll take an answer from anyone. The cost is that their scan, sometimes with their name burned into the corner of the image, is now public forever.

Read Your Scan is the private, instant, better‑informed version of that impulse: an AI that actually reads the pixels, points at the finding on the scan, explains it in your language, and — if you do want to show someone — lets you anonymize it first and send a revocable link instead of a public post.

Swap the first paragraph for your own origin moment if there was one (your own scan, a family member's). SaaSHub story answers read much better with a specific person in them, and that's the one detail I can't write for you.

Read Your Scan's answer:

Frontend / app: Next.js 15 (App Router), React 19, TypeScript, Tailwind CSS v4.

Imaging: NiiVue for WebGL volume rendering, @niivue/dcm2niix compiled to WebAssembly for DICOM→NIfTI conversion entirely in the browser, nifti-reader-js, and Tesseract.js (self‑hosted, no CDN) for OCR‑based redaction of burned‑in text.

AI: Anthropic Claude, Google Gemini / MedGemma, and additional models via OpenRouter, used for vision analysis, report generation, grounded Q&A and translation.

Backend / infra: Supabase (Postgres with row‑level security, Auth, Storage), Stripe for credit purchases, Resend for transactional email, and Node.js on a VPS under PM2 behind Cloudflare.