Dictate findings. Get a structured report draft in seconds.
Speak or type your findings in English, Urdu, or a mix of both. RadReport AI turns them into a structured, template-based radiology report you review, edit, and export as PDF or Word.
Welcome
Recent Reports
About RadReport AI
RadReport AI — Voice-to-Report Radiology Assistant · Version 5.0
Dictate in English, Urdu or Roman Urdu and get structured, RadGenie-style radiology reports with automatic study detection across CT, MRI, Ultrasound, X-Ray, Doppler, Fluoroscopy, Mammography and Nuclear Medicine.
Technology: Google Gemini AI · Hosted on Cloudflare Pages · Reports and settings stay on this device unless you connect Supabase sync.
AI-generated drafts only — every report must be verified by the reporting radiologist. Not a medical device.
Patient Demographics
Dictation Studio
Report Type
Normal Macros
Compare with a previous report
Live Transcript
Clinical Report
Draft only — verify every line before clinical use.
History
Custom Templates
Built-in Templates
AI Settings — Google Gemini (your own key)
Get a free key from Google AI Studio (aistudio.google.com/apikey). The key is stored only in this browser on this device and is sent only to Google's Gemini API.
If a model is busy or unavailable, the app automatically falls back to the next model. Indicative cost at ~600 cases/day: about $4/day on gemini-3.8-flash at current paid rates (roughly $120/month); gemini-3.5-flash costs about 2.5x more.
Privacy note: on Google's free Gemini tier, content may be used to improve Google products. Dictate without patient identifiers, or enable billing on your Google project for paid-tier data terms.
Settings now save automatically as you type — the Save button is a backup.
Doctor Profile
App & Developer Info
App: RadReport AI · Version 5.0 · This line saves automatically as you type.
Report Letterhead
Account & Backend (optional)
Mode: Local (this device only)
To enable cloud sync, copy config.example.js to config.js, add your Supabase URL and anon key, and run the SQL in supabase/schema.sql. See README.md.