Liver Tumor Classification: 2D CT Slices from the LiTS Challenge

Processed instances from the Liver Tumor Segmentation Challenge (LiTS) dataset, adapted for binary liver tumor classification. The original 3D CT scans and segmentation masks have been converted into individual 2D slices. The dataset was created by author 'seongun' and was last updated on October 27, 2025.

Use Cases

• Train binary classifiers for liver tumor detection based on processed CT slices.
• Benchmark image classification models using data derived from a known medical imaging challenge.
• Explore preprocessing techniques for converting 3D medical scans into 2D training data.

Strengths

• Derived from the established Liver Tumor Segmentation Challenge (LiTS) dataset.
• Includes both original and windowed (processed) CT slices for each data instance.
• Last updated on October 27, 2025, indicating recent maintenance.

Limitations

• Row count, file formats, and column-level documentation are unknown, limiting suitability assessment.
• The description is incomplete; actual data quality and preprocessing details require manual inspection after download.

Provenance

Source

Liver Tumor Segmentation Challenge (LiTS) dataset.

Collection Method

Original 3D CT scans and segmentation masks were processed into individual 2D slices.

Time Range

null

Freshness

Last updated 2025-10-27 16:36:44.

Geography

null