Prostate MRI Segmentation Model Performance Metrics

Experimental results from 2026 compare the LM-UNet and UVM-UNet models for prostate MRI lesion segmentation. The data includes metrics on parameter count, computational memory requirements, and segmentation accuracy for lesion margins, evaluated on the PROMISE12 dataset. The study was authored by Kuncai Xu.

Use Cases

• Compare parameter count and computational memory requirements between LM-UNet and UVM-UNet models.
• Analyze segmentation accuracy for lesion margins across different model architectures.
• Benchmark model performance using the PROMISE12 dataset as a reference standard.

Strengths

• Data is recent, with a last update in March 2026.
• Results are derived from a comparative experiment on the established PROMISE12 dataset.
• The dataset is small (5.5 KB), enabling quick download and inspection.

Limitations

• The dataset is extremely small (5.5 KB), indicating very limited scope and likely a summary table, not raw data.
• Specific row counts, column names, and sample data are unavailable, limiting detailed analysis.
• The data focuses solely on model comparison results, not on the underlying medical images or patient records.

Provenance

Source

figshare, authored by Kuncai Xu.

Collection Method

Experimental results from comparing segmentation models.

Time Range

null

Freshness

Last updated on 2026-03-23.

Geography

null