SPFF–UNet and Baseline Models: Macro-averaged Medical Image Segmentation Performance

Macro-averaged segmentation performance on an external test set, comparing five baseline models to the proposed SPFF–UNet model. The dataset contains mean ± standard deviation metrics computed over 12 foreground classes, excluding background voxels, across three random seeds. This 5.5 KB Excel file was authored by Nadine Francis and last updated on April 9, 2026.

Use Cases

• Benchmarking segmentation model performance based on macro-averaged metrics over 12 target classes.
• Comparing a proposed model (SPFF–UNet) against baseline models based on external test set results.
• Analyzing Intersection-over-Union (IoU) and other segmentation metrics for foreground-only voxel data.
• Assessing the stability of model performance based on the reported mean and standard deviation across three seeds.

Strengths

• Metrics are computed over 12 distinct foreground classes, providing a multi-class evaluation.
• Results include mean and standard deviation across three random seeds, indicating performance stability.
• Dataset is small (5.5 KB) and in a common format (XLS), facilitating quick inspection.

Limitations

• Row count and specific column-level documentation are unknown; field semantics must be inferred after download.
• The dataset is limited to aggregated performance metrics; it does not contain the underlying image or voxel data.
• Description metadata is limited; actual data quality and context require manual inspection after download.

Provenance

Source

figshare, authored by Nadine Francis.

Collection Method

Likely contains results from evaluating medical image segmentation models on an external test set.

Time Range

null

Freshness

Last updated 2026-04-09 17:38:26; freshness should be verified.

Geography

null