Radiography imaging protocols focus on particular body regions, therefore producing images of great similarity and yielding recurrent anatomical structures across patients. Exploiting this structured information could potentially ease the detection of anomalies from radiography images. To this end, we propose a Simple Space-Aware Memory Matrix for In-painting and Detecting anomalies from radiography images (abbreviated as SimSID). We formulate anomaly detection as an image reconstruction task, consisting of a space-aware memory matrix and an in-painting block in the feature space. During the training, SimSID can taxonomize the ingrained anatomical structures into recurrent visual patterns, and in the inference, it can identify anomalies (unseen/modified visual patterns) from the test image. Our SimSID surpasses the state of the arts in unsupervised anomaly detection by +8.0%, +5.0%, and +9.9% AUC scores on ZhangLab, COVIDx, and CheXpert benchmark datasets, respectively. Code: https://github.com/MrGiovanni/SimSID

翻译：放射成像协议聚焦于特定身体区域，因此能生成高度相似的图像，并在患者间产生重复的解剖结构。利用这种结构化信息有望简化放射影像中异常检测的难度。为此，我们提出一种简单的空间感知记忆矩阵用于放射影像的修复与异常检测（简称为SimSID）。我们将异常检测形式化为图像重建任务，该任务由特征空间中的空间感知记忆矩阵和修复模块组成。训练时，SimSID能将固有的解剖结构分类为重复的视觉模式；推理时，它能从测试图像中识别异常（未见或修改的视觉模式）。我们的SimSID在ZhangLab、COVIDx和CheXpert基准数据集上的无监督异常检测AUC分数分别超越当前最优方法+8.0%、+5.0%和+9.9%。代码：https://github.com/MrGiovanni/SimSID