A collection of preprocessed medical images from the Radiological Society of North America (RSNA) Intracranial Hemorrhage (ICH) detection challenge. The dataset likely contains PNG format images derived from brain CT scans, specifically representing negative cases without hemorrhage. It is hosted on the Kaggle platform.
Use Cases
- Train a binary classifier to identify the absence of intracranial hemorrhage in brain CT scans (inferred from domain, verify after download)
- Benchmark image preprocessing pipelines for converting DICOM medical images to standard PNG format (inferred from domain, verify after download)
- Create synthetic data or augment training sets for medical AI models using confirmed negative cases (inferred from domain, verify after download)
Strengths
- Published on Kaggle, a major platform for data science competitions and datasets.
- Focuses on preprocessed PNG images, which may simplify initial model development compared to raw DICOM files.
Limitations
- Metadata is minimal; actual content, image count, and preprocessing details require verification after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
Provenance
- Source
- Radiological Society of North America (RSNA) Intracranial Hemorrhage detection challenge.
- Collection Method
- Likely derived from a public Kaggle competition dataset, with preprocessing applied to convert images to PNG format.