YOLO-Seg Burn Classification Model Performance Metrics

358 retrospective images of skin burns were used to evaluate YOLO-Seg models for burn depth classification. The dataset contains performance metrics from a comparative analysis of different architectural sizes, with results evaluated by expert general surgeons. It was authored by Uğur Şevik and last updated in March 2026.

Use Cases

• Benchmarking object detection and segmentation models based on performance metrics for burn classification
• Comparing the effectiveness of different YOLO architecture sizes for medical image analysis
• Training or validating models for automated burn depth assessment based on the described image dataset
• Conducting statistical analysis to confirm model performance improvements using various data augmentation techniques

Strengths

• Performance metrics are based on 358 retrospective medical images.
• Evaluation was conducted by expert general surgeons, suggesting clinical relevance.
• Comparative analysis includes different model architectures and data augmentation strategies.
• Dataset is openly licensed under CC-BY-4.0.

Limitations

• The dataset is very small at 5.5 KB, indicating limited scope.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Uğur Şevik via figshare.

Collection Method

Likely contains performance metrics from evaluating YOLO-Seg models on a proprietary collection of burn images.

Freshness

Last updated 2026-03-18 17:28:40; freshness should be verified.