Personal Health Information Masking Preview Dataset

EPII Personal Health Information (PHI) Masking Preview Dataset provides 400 samples demonstrating the structure and label taxonomy for identifying and masking sensitive health information in text. This preview represents the quality of the full commercial dataset containing over 100,000 entries, created by ai4privacy and last updated in April 2026.

Use Cases

• Train a named entity recognition model to identify PHI labels like patient names and medical record numbers within text samples.
• Benchmark the performance of a text masking or redaction algorithm against the dataset's annotated sensitive information fields.
• Fine-tune a language model for de-identification tasks using the provided taxonomy of health information categories.
• Develop synthetic data generation techniques by learning patterns from the annotated PHI structures.

Strengths

• Preview contains 400 samples demonstrating data structure and quality.
• Full dataset referenced contains over 100,000 entries for model training.

Limitations

• Preview size of 400 rows is small for training robust machine learning models.
• Specific column names, data distributions, and license terms are not provided in the preview.

Provenance

Source

ai4privacy

Collection Method

null

Time Range

null

Freshness

null

Geography

null