Development of a defacing algorithm to protect the privacy of head and neck cancer patients in publicly‐accessible radiotherapy datasets
Bibliographic record
Abstract
BACKGROUND: The increase in public medical imaging datasets has raised concerns about potential patient reidentification from head CT scans. However, existing defacing algorithms, which help protect patient confidentiality, fail to preserve critical radiotherapy structures, including organs at risk (OARs) and planning target volumes (PTVs) in head and neck cancer (HNC) patients. Furthermore, current algorithms do not address the defacing of DICOM-RT structure set and dose data, which also contain information for facial surface rendering. PURPOSE: To develop and validate a novel automated defacing algorithm that preserves OARs and PTVs while removing identifiable features from HNC CTs and DICOM-RT data. METHODS: Eye contours were used as landmarks to automate the removal of CT pixels above the inferior-most slice of the eye and anterior to the midpoint of the eye. Pixels within PTVs were retained if they intersected with the removed region. The body contour and dose map were then reshaped to reflect the defaced image. We validated our approach on 829 HNC CT-simulation scans from 622 patients. To evaluate privacy protection, we applied the FaceNet512 facial recognition algorithm before and after defacing on 3D-rendered CT scan pairs from 70 patients at two time points. To assess research utility, we examined the impact of defacing on auto-contouring performance using LimbusAI and analyzed the locations of PTVs relative to the defaced regions. RESULTS: Before defacing, the facial recognition algorithm matched 97% of patients' CT scans. After defacing, this rate dropped to just 4%. LimbusAI effectively auto-contoured organs in the defaced CTs, with perfect Dice scores of 1 for OARs below the defaced region, and mean Dice scores exceeding 0.95 for OARs on the same slices as the defaced region. PTV analysis revealed that 86% of PTVs were entirely below the cropped region, 9.1% were on the same slice as the crop without overlap, and only 4.9% extended into the cropped area. All overlapping PTVs were preserved through our algorithm's design. CONCLUSIONS: We developed a novel defacing algorithm that anonymizes HNC CT scans and related DICOM-RT data. Our algorithm balances patient privacy while preserving essential structures for radiotherapy research, facilitating the sharing of HNC imaging datasets for Big Data and AI.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".