Infrared Chemical Images of a Breast Cancer Tissue Microarray

Mid-infrared spectroscopy chemical images of a breast cancer tissue microarray, specifically sample BR20832 from Biomax. The data relates to an open-access paper published in Analyst by researchers from the University of Manchester exploring machine learning for infrared pathology. Processed versions of the data are available in a separate Zenodo archive.

Use Cases

• Training classifiers for cancer pathology based on infrared spectral features.
• Developing image segmentation models for tissue microarray analysis.
• Benchmarking machine learning approaches like AdaBoost and Random Forests on unbalanced biomedical datasets.
• Studying the correlation between chemical composition from spectroscopy and tissue morphology.

Strengths

• Data is linked to a peer-reviewed open-access paper published in Analyst, providing academic context.
• The tissue microarray source is specified as BR20832 from Biomax, allowing for sample traceability.
• Processed data versions are available in a separate, citable Zenodo archive.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Manchester, associated with a published paper in Analyst.

Collection Method

Mid-infrared spectroscopy imaging of a commercial tissue microarray.

Time Range

null

Freshness

null

Geography

null