Departament de Matemàtiques i Informàtica, Universitat de Barcelona (UB), Barcelona, Spain.
Departament de Matemàtiques i Informàtica, Universitat de Barcelona (UB), Barcelona, Spain.
Artif Intell Med. 2023 Sep;143:102606. doi: 10.1016/j.artmed.2023.102606. Epub 2023 Jun 15.
While deep learning has displayed excellent performance in a broad spectrum of application areas, neural networks still struggle to recognize what they have not seen, i.e., out-of-distribution (OOD) inputs. In the medical field, building robust models that are able to detect OOD images is highly critical, as these rare images could show diseases or anomalies that should be detected. In this study, we use wireless capsule endoscopy (WCE) images to present a novel patch-based self-supervised approach comprising three stages. First, we train a triplet network to learn vector representations of WCE image patches. Second, we cluster the patch embeddings to group patches in terms of visual similarity. Third, we use the cluster assignments as pseudolabels to train a patch classifier and use the Out-of-Distribution Detector for Neural Networks (ODIN) for OOD detection. The system has been tested on the Kvasir-capsule, a publicly released WCE dataset. Empirical results show an OOD detection improvement compared to baseline methods. Our method can detect unseen pathologies and anomalies such as lymphangiectasia, foreign bodies and blood with AUROC>0.6. This work presents an effective solution for OOD detection models without needing labeled images.
虽然深度学习在广泛的应用领域中表现出了优异的性能，但神经网络仍然难以识别它们从未见过的东西，即分布外（OOD）输入。在医学领域，构建能够检测 OOD 图像的强大模型非常关键，因为这些罕见的图像可能会显示出应该检测到的疾病或异常。在这项研究中，我们使用无线胶囊内窥镜（WCE）图像提出了一种新的基于补丁的自监督方法，该方法包含三个阶段。首先，我们训练一个三重网络来学习 WCE 图像补丁的向量表示。其次，我们对补丁嵌入进行聚类，根据视觉相似性对补丁进行分组。最后，我们使用簇分配作为伪标签来训练补丁分类器，并使用用于神经网络的离群检测（ODIN）进行 OOD 检测。该系统已经在公开发布的 WCE 数据集 Kvasir-capsule 上进行了测试。实验结果表明，与基线方法相比，该方法在 OOD 检测方面有所改进。我们的方法可以检测到未见过的病理和异常，如淋巴管扩张、异物和血液，AUROC>0.6。这项工作为不需要标记图像的 OOD 检测模型提供了一种有效的解决方案。
