Computational Efficiency Comparison for Cardiac MRI Segmentation

Yafei Mi published a dataset comparing computational efficiency on the ACDC cardiac MRI segmentation dataset using only 10% labeled data. The dataset is a 5.5 KB XLS file, last updated in April 2026. It provides metrics for evaluating the speed and resource usage of segmentation models on a subset of the well-known ACDC benchmark.

Use Cases

• Benchmark training or inference time of segmentation models against reported metrics in the XLS file.
• Analyze the relationship between computational efficiency (e.g., runtime, memory) and model performance on the ACDC 10% subset.
• Compare efficiency results across different algorithm configurations documented in the comparison spreadsheet.

Strengths

• Provides a direct computational efficiency comparison on the established ACDC medical imaging benchmark.
• Dataset is small (5.5 KB), ensuring fast download and easy inspection in spreadsheet software.
• Uses a standardized subset (10% labeled data) of the ACDC dataset for controlled evaluation.

Limitations

• Extremely limited in scope at only 5.5 KB, likely containing only summary metrics, not raw image or model data.
• The underlying ACDC data subset is only 10% labeled, which is a very small sample for comprehensive model evaluation.
• Unknown row count and specific column structure limit immediate understanding of the comparison's depth.

Provenance

Source

Yafei Mi via figshare.

Collection Method

Derived from computational experiments on the ACDC (Automated Cardiac Diagnosis Challenge) dataset.

Freshness

Last updated April 2026, indicating recent analysis.