Privacy
How your data is protected
Dr.Holesto Blood collects your blood test history in one place. Upload a photo or PDF of your lab report, check the recognized values, and get a clear overview: which markers fall outside the lab range, what changed since your last test, and which questions to prepare for your doctor.
The service does not diagnose or prescribe treatment. It helps you keep your data, not lose track of important changes over time, and arrive at your appointment with a clear report in the language you need.
Below are direct answers to the questions people usually have about medical data online.
Where are files physically stored?
In Supabase (Postgres and file storage), data center in the ap-south-1 region. Your uploaded report and the results of its analysis live in the same place as the rest of your test history.
How is data encrypted in transit and at rest?
All traffic between your browser and the server goes over HTTPS. Storage is handled by Supabase, which encrypts data at rest under the platform’s standard settings.
Who has access to your original documents?
Only the application’s server side, through a database service key. Every table has row-level security enabled with no read policy at all — meaning your data cannot be read with the public (anonymous) key under any circumstances, not merely “denied by default.”
Are documents sent to outside OCR and AI providers?
Yes. The image of your report is sent to a PDF conversion service (PDF.co, for PDF files only) and to the recognition and analysis model (OpenAI). Automated analysis is technically impossible without this step — stating that plainly is more honest than hiding it.
Is your data used to train models?
We do not train our own models on user data. Requests to OpenAI go through their API — under that channel’s public policy (unlike the same company’s consumer products), data is not used to train models by default. That is the provider’s stated policy, not a guarantee of our own, so we name it as such rather than claim it as ours.
What is kept in technical logs, and for how long?
Diagnostic request bodies are cleared after 24 hours, raw model responses after 30 days, and the technical records of analysis attempts themselves are deleted after 180 days. After that, only aggregate figures remain (duration, error code) with no report content.
What exactly happens when you delete a test?
The record and the original file in storage are deleted together. If a doctor link already existed for that test and the doctor already downloaded the generated PDF, that file cannot be recalled from their side — you can revoke the link itself and delete our copy.
How are backups deleted?
We are finalizing our backup procedure and will publish it here before the service leaves beta.
Can you use the service without your real name?
Yes. Most profile fields, including your name and ID number, are optional — an empty field simply is not shown, to you or to any doctor you choose to grant access to.
How do you download all of your data?
Your profile has a "Download all your data" button that bundles an archive with every uploaded report, every generated doctor report, and structured data for each test. Each original report and each doctor report also remain available individually from the same place.
Who is the data controller, and how do you reach them?
We will publish the controller’s legal details and a contact address here before the service leaves beta.