基于孪生神经网络潜空间的单类支持向量机用于脑 MRI 白质高信号异常无监督检测（英文版）

英文标题：One-Class SVM on siamese neural network latent space for Unsupervised Anomaly Detection on brain MRI White Matter Hyperintensities中文摘要：提出基于 Patch 自编码器的无监督异常检测方法，在多模态神经影像任务中，将单类支持向量机应用于异常检测以检测病变，评估了该模型在公共数据集 White Matter Hyperintensities (WMH) Challenge 上的性能和表现与两种最优秀的现有技术相当。英文摘要：Anomaly detection remains a challenging task in neuroimaging when little tono supervision is available and when lesions can be very small or with subtlecontrast. Patch-based representation learning has shown powerfu