DICOM anonymizer (online)
Strip patient-identifying tags from DICOM files right in your browser before sharing them for research, a second opinion, or upload. The files never leave your device.
Removes name, ID, birth date, address, phone, referring/performing physician, institution, accession number and comments. Processed in your browser — nothing is uploaded.
Opens: One or more DICOM (.dcm) files. · Files are processed in your browser — nothing is uploaded.
Text burned into the image pixels (a name or date printed on the scan) isn’t a tag — this tool can’t remove it. When you share a study, anonymize before sharing to black it out.
What this tool does
Clears the identifying tags, by name
Patient name, patient ID, other patient IDs and names, birth date, birth time, age, address, telephone, ethnic group and comments; referring, performing, recording, reading and requesting physician, plus the operator; institution name and address, accession number, admission ID and patient residence. Twenty-one tags, each one blanked in place.
Tells you what it found in each file
Every processed file is listed with the tags that were actually cleared in it — or a note that no identifying tags were present. You are not asked to trust that something happened.
The pixel data is untouched
Only header bytes are overwritten. The image comes out identical to the one that went in: no re-encoding, no quality loss, no change to the geometry, and the file stays a valid DICOM that any viewer or PACS will open.
Several files at once
Select a batch of .dcm files and each is processed and offered as its own cleaned download, named with an -anon suffix so the originals are never overwritten.
Refuses rather than mangles
It handles implicit VR little-endian, explicit VR little-endian, deflated and the JPEG-family transfer syntaxes. Anything else — big-endian, most often — is skipped with a message instead of being silently rewritten, and a file that is not a Part-10 DICOM is reported as such. A file it cannot parse confidently is never altered.
Nothing is uploaded
The DICOM bytes are parsed and edited by your own browser. Sending files to a server to have their identifiers removed would mean sending identified files to a server, which rather defeats the purpose.
What it does not do
It cannot remove identity burned into the image pixels — a name or date printed onto the picture by the scanner is not a tag; when you share a study through this site, the sharing flow blacks that out for you. It leaves study, series and instance UIDs intact, so the series stay linked and remain re-identifiable by anyone holding the original archive, and it does not touch private vendor tags. This is a practical scrub before sharing a file, not a certified DICOM PS3.15 de-identification profile — review the result before releasing data.
How to use it
- 1.Click “Choose DICOM file(s)” and select the .dcm files to clean.
- 2.Identifying tags (name, ID, birth date, address, physician, etc.) are blanked locally.
- 3.Download each anonymized file — the pixel data is untouched, only identifiers are cleared.
FAQ
- Which tags are removed?
- Patient name, ID, birth date/time, age, address, phone, ethnic group, comments, plus referring/performing physician, institution and accession number. Pixel data is left unchanged.
- Are my files uploaded to anonymize them?
- No. The DICOM bytes are parsed and edited entirely in your browser; nothing is sent to a server.
- Is this a certified de-identification?
- It removes the standard identifying tags but is not a certified, exhaustive de-identification — it cannot remove identity burned into the image pixels or every private tag. Review before sharing.
Want your scan explained in plain language?
The viewer is free — and so is understanding your radiologist’s report. Create an account, upload your scan (de-identified in your browser) and your report, and every finding is mapped onto the images, explained in plain language and scored — no card needed. An independent AI reading of the images is a paid extra. Informational only — not a medical diagnosis.