BMAD
收藏资源简介:
BMAD数据集是由阿尔伯塔大学和卡内基梅隆大学的研究团队开发的,旨在为医学图像异常检测提供一个全面的评估基准。该数据集包含六个来自五个不同医学领域的重新组织的数据集,包括脑部MRI、肝脏CT、视网膜OCT、胸部X光和数字病理学。这些数据集用于评估和比较不同的异常检测算法,特别是在无监督学习环境下的性能。BMAD数据集的设计考虑了医学图像的多样性和复杂性,旨在推动更通用和鲁棒的异常检测方法的发展。
The BMAD dataset was developed by research teams from the University of Alberta and Carnegie Mellon University, aiming to provide a comprehensive evaluation benchmark for medical image anomaly detection. This dataset comprises six reorganized datasets spanning five distinct medical domains, including brain MRI, liver CT, retinal OCT, chest X-ray, and digital pathology. These datasets are utilized to evaluate and compare the performance of various anomaly detection algorithms, especially under unsupervised learning scenarios. The BMAD dataset was designed with full consideration of the diversity and complexity of medical images, with the goal of promoting the development of more generalizable and robust anomaly detection methods.




