Brain CT Scans for Pathology Detection, Over 70,000 Studies

Over 70,000 CT scan studies, including 20,000+ with medical professional protocols, are available for computer vision research. The dataset, created by UniDataPro, focuses on detecting and analyzing brain pathologies such as tumors, hemorrhages, and cancers. It was last updated on April 4, 2025.

Use Cases

• Training image classification models to detect brain tumors based on CT scan imagery.
• Developing segmentation models to identify regions of brain hemorrhage from medical scans.
• Benchmarking model performance on a large-scale medical imaging dataset with over 70,000 studies.
• Conducting research on the efficacy of protocol-driven versus non-protocol medical imaging studies.

Strengths

• Contains over 70,000 CT scan studies, providing a substantial volume for model training.
• Includes 20,000+ studies with protocols developed by medical professionals, suggesting a degree of clinical validation.
• Explicitly designed for detecting and analyzing 5 distinct brain pathologies, offering clear task definition.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count per study and total image count are unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality and annotation detail require manual inspection after download.

Provenance

Source

UniDataPro on Hugging Face

Collection Method

Likely aggregated from medical imaging sources; specific collection method is not detailed.

Freshness

Last updated 2025-04-04 12:30:04; freshness should be verified.