DICOM Anonymizer Online — Remove Patient Data in Your Browser

Strip patient and owner data from DICOM tags. Processed locally — no uploads to our servers, ever.

Drop .dcm files or a DICOM folder

Tags are stripped in-memory — original files are not modified

What data is removed during DICOM de-identification

Every DICOM dataset produced by hospital scanners contains metadata headers detailing patient identification and clinical facility parameters. MedSlice enables precise sanitization of key protected health information (PHI) fields:

Direct Patient Identifiers
  • PatientName (0010,0010) — Full legal name
  • PatientID (0010,0020) — Hospital Medical Record Number
  • PatientBirthDate (0010,0030) — Exact date of birth
  • PatientSex & Weight — Demographic attributes
Institutional & Clinical Context
  • InstitutionName (0008,0080) — Hospital clinic name
  • ReferringPhysician (0008,0090) — Doctor name
  • AccessionNumber (0008,0050) — Order tracking number
  • StudyDate & Time — Procedure timestamps

Pseudo-anonymization options and strategies

Depending on your scientific protocol or regulatory review board requirements, MedSlice supports multiple de-identification strategies:

Empty String

Replaces sensitive tag values with blank characters while preserving DICOM conformance dictionary keys.

Remove Entirely

Completely strips the tag byte sequences from the header payload for maximum data minimization.

Hash Pseudonymization

Calculates cryptographic subject hashes so multi-session studies correlate without revealing identities.

Designed for clinical research, teaching, and legal expert reports

De-identifying medical imaging before sharing is standard practice across academic and professional environments. MedSlice provides a streamlined workflow for:

  • Medical Publications & Journals: De-identify case study scans before submission to journal peer review.
  • University Teaching & Grand Rounds: Prepare real-world CT, MRI, and X-ray teaching files for residency lectures without exposing patient identities.
  • Legal & Expert Witness Opinions: Prepare sanitized dispute records while preserving precise radiometric dimensions for forensic measurement.
  • AI & Machine Learning Training: Batch prepare clean, de-identified datasets for computer vision training pipelines.

Need to inspect the images or verify measurements first? Open them in our online DICOM viewer or export them via DICOM to JPG.

Designed to support privacy-by-design workflows

Conventional web tools often upload medical files to third-party cloud infrastructure. MedSlice takes a technically demonstrable, local approach:

Local Browser Processing: MedSlice processes DICOM files locally in your browser; it does not upload your files to our servers. This architecture is designed to support privacy-by-design workflows by keeping all file decoding and tag sanitization inside your device's browser memory.

Explore MedSlice Pro Open DICOM Viewer

Frequently Asked Questions

How does in-browser processing support privacy and data protection?

MedSlice processes DICOM files locally in your browser; it does not upload your files to our servers. By sanitizing Protected Health Information (PHI) directly inside client-side WebAssembly memory on your device before sharing, it helps align with privacy-by-design workflows.

How does this tool help with de-identification workflows?

MedSlice provides customizable tag removal and replacement tools to help users de-identify key clinical metadata fields (such as patient names, IDs, dates of birth, institutions, and accession numbers) according to DICOM PS 3.15 standards without altering diagnostic pixel data.

What is the difference between clearing tags, removing tags, and pseudonymization?

Clearing tags sets values to an empty string while maintaining tag existence. Removing tags deletes the attribute key entirely. Pseudonymization replaces real identifiers with a consistent cryptographic hash (e.g. "STUDY_9A4F"), allowing multi-scan correlation without revealing patient identity.

Does this tool remove "burned-in" text visible inside the image pixels?

No. Tag-based anonymization modifies the binary DICOM header metadata (0010, 0008, etc.). If ultrasound or endoscopy images have patient names visually stamped directly onto pixel pixels, you should verify that secondary burned-in annotations are not present.

Can I de-identify veterinary imaging records (DICOM Supplement 83)?

Yes. MedSlice includes dedicated veterinary de-identification modes that strip owner names (0010,2297), keeper addresses, and microchip IDs while preserving species and breed tags for veterinary clinical research.

What are the Free vs Pro limits for anonymization?

The Free tier allows de-identifying single files and small batches up to 5 files per session. MedSlice Pro provides unlimited batch anonymization for thousands of files, custom study prefix mapping, and fast ZIP bulk downloads.