Prostate MRI Lesion Segmentation Method Comparisons

Kuncai Xu's 2026 study presents experimental results comparing a proposed LM-UNet method against seven other segmentation methods on the PROMISE12 dataset. The analysis focuses on metrics like parameter count, computational memory requirements, and lesion margin segmentation accuracy. The supporting data file is 5.5 KB in size.

Use Cases

• Compare the parameter count and computational memory requirements of eight segmentation methods for model efficiency analysis.
• Benchmark lesion margin segmentation accuracy across different methods using the PROMISE12 dataset.
• Analyze the performance of the proposed LM-UNet method relative to seven comparative methods on specific evaluation metrics.

Strengths

• Directly compares eight distinct segmentation methods, providing a clear performance benchmark.
• Focuses on concrete, clinically relevant metrics for prostate MRI analysis, such as lesion margin segmentation accuracy.
• The data is associated with a recent 2026 publication, indicating contemporary methodological relevance.

Limitations

• The dataset is extremely small at 5.5 KB, suggesting it contains only summary results or a limited set of metrics, not raw image data.
• No raw image data, patient records, or detailed tabular columns are provided, limiting the scope to methodological comparison results.
• The provenance and specific composition of the underlying PROMISE12 benchmark data used for evaluation are not detailed within this resource.

Provenance

Source

Kuncai Xu via figshare

Collection Method

Experimental results from training and evaluating segmentation models on the PROMISE12 dataset.

Freshness

Last updated in March 2026.