SMMILe: Whole-Slide Image Embeddings and Superpixel Data for Six Public Datasets

Six publicly available whole-slide image (WSI) datasets are included, with embeddings generated using two different encoders: ResNet-50 and Conch. The repository contains embedding superpixel segmentation and checkpoint data, authored by zeyugao and last updated on September 4,我们发现2025. Each dataset is available as a ZIP file containing both the embeddings and the corresponding superpixel segmentation results.

Use Cases

• Training or benchmarking WSI analysis models based on provided ResNet-50 and Conch encoder embeddings.
• Developing superpixel-based segmentation algorithms using the included segmentation results.
• Conducting comparative studies on different embedding methods for medical image representation.
• Fine-tuning pre-trained models using the provided checkpoint data for specific WSI datasets.

Strengths

• Includes data from six distinct publicly available whole-slide image datasets.
• Provides embeddings from two different encoders (ResNet-50 and Conch) for comparative analysis.
• Bundles both embeddings and corresponding superpixel segmentation results in each dataset file.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• License information is unavailable, which could restrict usage.

Provenance

Source

huggingface

Collection Method

Embeddings and superpixel segmentation generated from six public WSI datasets using ResNet-50 and Conch encoders.

Time Range

null

Freshness

Last updated 2025-09-04 12:42:10; freshness should be verified.

Geography

null