Unsupervised pathology detection in medical images using conditional variational autoencoders
Unsupervised pathology detection in medical images using conditional variational autoencoders
复制标题
DOI：
10.1007/s11548-018-1898-0
复制
发表时间：
2019-03-01
影响因子：
3
通讯作者：
Ehrhardt, Jan
中科院分区：
文献类型：
--
作者：
Uzunova, Hristina;Schultz, Sandra;Ehrhardt, Jan
目的
医学图像数据中的病理学检测是一项重要但相当复杂的任务。特别是，病理学的巨大变异性对自动检测方法甚至机器学习方法都是一个挑战。有监督算法通常会基于大量带注释的数据集学习单一病理结构的外观。由于此类数据通常不可得，尤其是大量数据，在这项工作中我们采用一种不同的无监督方法。
方法
我们的方法基于学习健康数据的整体变异性，并通过病理与所学习的正常情况的差异来检测病理。为此，我们使用条件变分自编码器，它学习健康图像的重建和编码分布，并且还能够整合关于数据（条件）的某些先验知识。
结果
我们在不同的2D和3D数据集上的实验表明，该方法适用于病理检测，并能提供合理的骰子系数和曲线下面积（AUC）。此外，这种方法可以估计病理图像中缺失的对应关系，因此可以用作配准方法的预处理步骤。我们的实验表明，在使用这种方法时，病理数据的配准结果有所改善。
结论
总体而言，所提出的方法适用于医学图像中的粗略病理学检测，并且可以成功地用作其他图像处理方法的预处理步骤。
PurposePathology detection in medical image data is an important but a rather complicated task. In particular, the big variability of the pathologies is a challenge to automatic detection methods and even to machine learning methods. Supervised algorithms would usually learn the appearance of a single pathological structure based on a large annotated dataset. As such data is not usually available, especially in large amounts, in this work we pursue a different unsupervised approach.MethodsOur method is based on learning the entire variability of healthy data and detect pathologies by their differences to the learned norm. For this purpose, we use conditional variational autoencoders which learn the reconstruction and encoding distribution of healthy images and also have the ability to integrate certain prior knowledge about the data (condition).ResultsOur experiments on different 2D and 3D datasets show that the approach is suitable for the detection of pathologies and deliver reasonable Dice coefficients and AUCs. Also this method can estimate missing correspondences in pathological images and thus can be used as a pre-step to a registration method. Our experiments show improving registration results on pathological data when using this approach.ConclusionsOverall the presented approach is suitable for a rough pathology detection in medical images and can be successfully used as a preprocessing step to other image processing methods.