Picai Prep: 3D MRI Preprocessing for Prostate Cancer Detection

DIAGNijmegen released this resource in 2025 to facilitate the preprocessing of 3D medical image archives for prostate cancer detection. It focuses on standardizing MRI data stored in DICOM and MHA formats for clinical machine learning applications. The dataset and its associated tools are maintained by the Diagnostic Image Analysis Group at Radboud University Medical Center.

Use Cases

• Preprocessing DICOM archives for prostate cancer detection models
• Converting MHA image volumes for standardized medical imaging pipelines
• Standardizing 3D MRI data for diagnostic AI training

Strengths

• Apache-2.0 licensed for open research and commercial use
• Supports standard medical imaging formats including DICOM and MHA
• Developed by DIAGNijmegen, a specialized medical imaging research group

Limitations

• Unknown total record count and sample size
• Domain-specific focus on prostate MRI may limit utility for other anatomical regions
• Lack of detailed metadata for individual image parameters in the provided summary

Provenance

Source

DIAGNijmegen (Diagnostic Image Analysis Group)

Freshness

Last updated March 2025.