ISIC 2018 Task 1 & 2, Interactive Atlas of Dermoscopy
收藏资源简介:
本研究使用了两个广泛应用于皮肤病变分析的公共数据集:ISIC 2018 Task 1 & 2和Interactive Atlas of Dermoscopy。ISIC 2018 Task 1 & 2数据集包含2594张皮肤病理图像,用于病变分割和皮肤病理属性分割。Interactive Atlas of Dermoscopy是一个教育资源,用于训练皮肤科医生,包含872张皮肤病理图像。这些数据集用于分析和减少皮肤病变分类模型中的偏差,旨在提高模型在实际医疗决策中的可靠性和准确性。
This study utilized two public datasets widely adopted in skin lesion analysis: ISIC 2018 Task 1 & 2 and the Interactive Atlas of Dermoscopy. The ISIC 2018 Task 1 & 2 dataset consists of 2,594 dermatopathological images, which are used for lesion segmentation and dermatopathological attribute segmentation. The Interactive Atlas of Dermoscopy, an educational resource designed for training dermatologists, contains 872 dermatopathological images. These datasets are employed to analyze and mitigate biases in skin lesion classification models, with the aim of enhancing the reliability and accuracy of such models in real-world clinical decision-making.




