Sign in to view source links and access this dataset
Description
1,574 invasive coronary angiogram studies, including X-ray multi-view videos, prepared for Nature Scientific Data by a team including Ivan Kruzhilov and Gleb Mazanov. The dataset is designed for the coronary dominance classification task, an essential aspect in assessing the severity of coronary artery disease. It was last updated on the Hugging Face platform on 2025-02 19.
Use Cases
Classifying coronary artery dominance patterns based on X-ray angiogram videos
Training machine learning models for automated cardiac anatomy assessment
Benchmarking medical imaging algorithms for a specific, clinically relevant classification task
Strengths
1,574 studies provide a substantial sample for model development
Includes X-ray multi-view videos, offering multiple perspectives per study
Dataset is prepared for publication in Nature Scientific Data, suggesting a curation standard
Limitations
Column-level documentation is absent; field semantics must be inferred after download
Row count per study is unknown, which may limit suitability assessment
Data may reflect bias inherent to the specific collection of angiograms used
Provenance
Source
Hugging Face dataset uploaded by author BearSubj13
Collection Method
Collection of invasive coronary angiograms, likely from clinical sources
Time Range
null
Freshness
Last updated 2025-02-19 16:39:18
Geography
null
License restrictions are unknown and should be verified before use.