MedicDecode is run by Jamal Hussein, an individual, not a company or a healthcare provider. This policy says what the site does with information, in plain terms. A step-by-step walkthrough is on How we handle your documents.
In short
- Your original file never leaves your device.
- Personal identifiers are removed in your browser, and you review the result before sending.
- Only the redacted text is sent, and only when you click send.
- MedicDecode stores no documents and no explanations.
- There are no accounts and no cookies. The page that handles your document has no adverts and loads nothing from outside MedicDecode.
Information that stays on your device
The file you choose (PDF or image) and the full, unredacted text read from it are processed only inside your browser. Text extraction and optical character recognition run locally, using code served from this website. This information is not transmitted to MedicDecode or to anyone else, and it is not saved in browser storage; it is gone when you close or reload the page.
Information sent when you click “Send this redacted text”
- The redacted text, as shown to you on the review screen.
- The document type and reading level you selected.
- Like any website request, your IP address and basic browser information reach the server as part of the connection.
The redaction step looks for names, dates of birth, addresses, phone numbers, email addresses and ID, medical record, insurance and account numbers. It is automatic and can miss things. You are responsible for checking the preview before you send it.
Medicine look-ups
When you look up a medicine on the medicine pages, the name you type is used to fetch that medicine’s official label from the US Food and Drug Administration’s public database (openFDA), and the label text is sent to the AI provider to be rewritten in plain language. No document and no personal details are involved. The resulting explanation is saved on MedicDecode’s server so the next person who looks up the same medicine gets it straight away.
Who receives the redacted text
The MedicDecode server forwards the redacted text to an AI model in order to produce the explanation. No other third party receives it.
The AI model is reached at (not configured), an OpenAI-compatible endpoint configured by the site operator. Personal identifiers are still removed on your device before anything is sent. At the moment this copy of the site is running in demo mode with no provider key, so nothing is sent to a model and explanations are labelled mock output.
MedicDecode has no control over how the AI provider handles data once received. Please read the provider’s own policy if this matters to you, and do not send a document if you are not comfortable with it.
What the MedicDecode server keeps
- Document text and explanations: not stored. They are held in memory only for the seconds it takes to answer your request.
- Usage log: one line per explanation recording token counts, estimated cost, duration, document type and reading level. It contains no document text and no IP address.
- Visit counts: for each page load, the server records the page address, the website that linked to it (for example Google or Pinterest), any campaign tags in the link, the device type (phone, tablet or computer), your browser’s preferred language, your approximate location (country and city, looked up from the IP address in an offline database on MedicDecode’s own server), and an anonymous daily visitor code. The code is made from your IP address and browser with a secret that changes every day and is then deleted, so visits cannot be linked across days or traced back to you. The IP address itself is not saved. This is counted on MedicDecode’s own server: there is no tracking script and no third party, and it never includes document text or explanations.
- Rate limiting: a count of requests per IP address, held in memory, used to enforce a per-minute limit and a daily cap. It is not written to disk by the application.
- Web server logs: the hosting server may keep standard access logs (IP address, time, page requested) for security and troubleshooting. These do not contain document text.
Cookies, advertising and analytics
The explainer page never carries adverts. The page where you add your document, review the redacted text and read your explanation loads nothing from outside MedicDecode: no advertising scripts, no analytics scripts and no cookies. Your document and your explanation are never present on a page that shows adverts.
At present there are no adverts, analytics scripts or cookies anywhere on the site. The only measurement is the first-party visit count described above. MedicDecode plans to fund itself with adverts on its information pages (the home page, the explainer guides and other reading pages). When that starts, this section will be updated first to name the ad network and explain the cookies it uses.
Children
The site is intended for adults. If you are under 18, please use it with a parent or guardian.
Not a medical service
MedicDecode explains documents. It does not provide medical advice, diagnosis or treatment, and using it does not create a doctor and patient relationship.
Your choices
Because MedicDecode keeps no documents, explanations or accounts, there is nothing held about you that could be looked up, exported or deleted afterwards. If you have a question or a concern about privacy, use the contact page.
Changes
If the way the site handles information changes, for example a different AI provider, this page will be updated and the date at the top will change.