LM-UNet: Encoder Input and Output Dimensions for Prostate MRI Segmentation

A tabular dataset detailing the encoder input and output dimensions for the LM-UNet neural network architecture. The data was uploaded by author Kuncai Xu to figshare on March 23, 2026. The LM-UNet method is designed for segmenting lesions in prostate magnetic resonance images (MRI) and was evaluated on the PROMISE12 dataset.

Use Cases

• Replicating the LM-UNet architecture based on the specified encoder dimensions.
• Benchmarking segmentation model efficiency based on the reported parameter count and computational memory requirements.
• Developing feature fusion methods for medical images based on the described edge feature extraction and multi-level skip connections.
• Improving segmentation of low-contrast, fuzzy-boundary lesions based on the techniques described in the paper.

Strengths

• File size is 5.5 KB, indicating a compact and focused dataset.
• Released under the permissive CC-BY-4.0 license.
• Dataset is associated with a published method that outperformed seven comparative segmentation methods on the PROMISE12 dataset.

Limitations

• Row count and specific column definitions are unknown.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset appears to contain only architectural dimensions, not the underlying image data or segmentation labels.

Provenance

Source

figshare

Collection Method

Likely derived from the experimental setup and model architecture of the proposed LM-UNet method.

Time Range

null

Freshness

Last updated 2026-03-23 17:33:24; freshness should be verified.

Geography

null