3D-RAD: A 3D Radiology Visual Question Answering Dataset with Multi-Temporal Analysis

3D-RAD is a medical visual question answering dataset focused on 3D radiology images. The dataset was created by Tang-xiaoxiao and its associated paper was accepted to the NeurIPS 2025 conference. The repository was last updated on October 29, 2025.

Use Cases

• Train medical VQA models based on 3D radiology images mentioned in the description
• Develop models for multi-temporal analysis of medical scans as indicated in the dataset title
• Benchmark AI performance on diverse diagnostic tasks referenced in the description

Strengths

• Dataset is associated with a paper accepted to the NeurIPS 2025 conference
• Focuses on 3D radiology data, which is less common than 2D in public datasets

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count and data scale are unknown, which may limit suitability assessment

Provenance

Source

huggingface

Collection Method

Method of data gathering is not specified in the provided input.

Time Range

Temporal coverage of the data is not specified in the provided input.

Freshness

Last updated 2025-10-29 03:01:34

Geography

Spatial coverage of the data is not specified in the provided input.