RTS-Net: Thyroid Ultrasound Nodule Segmentation Data for Model Training

A collection of thyroid ultrasound images used to train and evaluate the RTS-Net segmentation model. The data includes images from the TN3K and DDTI datasets, as well as a large-scale clinical dataset, with model performance metrics provided. The dataset was uploaded by Xiaojie Sun in April 2026.

Use Cases

• Training segmentation models for thyroid nodule detection based on ultrasound images.
• Benchmarking new segmentation architectures against reported performance metrics (e.g., 81.66% F1-score).
• Studying model generalization across different clinical datasets (TN3K, DDTI).
• Investigating the impact of attention mechanisms and graph convolutions on boundary integrity for small nodules.

Strengths

• Model performance is quantitatively reported with specific metrics (e.g., 71.87% IoU on TN3K).
• The dataset was used in extensive experiments, including cross-dataset validation.
• The accompanying research paper describes a novel network architecture with ablation studies.

Limitations

• The exact number of images and rows is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect bias inherent to the specific clinical sources used.

Provenance

Source

figshare, uploaded by Xiaojie Sun.

Collection Method

Likely contains ultrasound images and associated segmentation masks used for training the RTS-Net model.

Time Range

null

Freshness

Last updated 2026-04-13 04:35:24.

Geography

null