# One certified open-source de-identification pipeline for scans and slides

Source: https://onco.cc/ideas/idea-data-open-deidentification-pipeline/  
OnCo record `idea-data-open-deidentification-pipeline` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Build and certify a single free tool that strips names and identifying marks from cancer scans and pathology slides, so every hospital stops writing its own.

## Summary

Every institution that shares imaging builds or buys its own DICOM de-identification, with inconsistent handling of burned-in text, private tags and slide label images. The Cancer Imaging Archive has curation experience and there are open tools (for example the RSNA anonymizer and CTP), but no certified reference implementation for whole-slide images or radiotherapy objects. The proposal funds a maintained open pipeline with a public test corpus of adversarial cases and an independent certification.

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: A certified reference pipeline adopted by more than 50 centres will cut time-to-share for a new imaging cohort from months to weeks and reduce identified leakage incidents to near zero in audits.
- Rationale: De-identification uncertainty is one of the most common reasons legal teams block imaging release; a certified standard shifts the risk decision from each hospital to a shared, audited tool.
- Proposed test: Run the pipeline and three institutional pipelines against a red-team corpus of 5,000 images with planted identifiers; publish leak rates. Then measure adoption and release times among centres that switch.
- Maturity: early-clinical
- Actor: engineering

## Sources

- The Cancer Imaging Archive: https://www.cancerimagingarchive.net/

## Connected records

- collections: [The Cancer Imaging Archive (TCIA)](https://onco.cc/collections/tcia/)
- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/)
- technologies: [CT (computed tomography)](https://onco.cc/technologies/ct/), [Digital pathology & AI](https://onco.cc/technologies/digital-pathology-ai/), [MRI](https://onco.cc/technologies/mri/)
- bottlenecks: [AI that is built but not validated or deployed](https://onco.cc/bottlenecks/b-ai-validation/), [Data silos](https://onco.cc/bottlenecks/b-data-silos/)

---
JSON: https://onco.cc/api/v1/entities/idea-data-open-deidentification-pipeline.json