Hibi Atsuhiro, Cusimano Michael D, Bilbily Alexander, Krishnan Rahul G, Tyrrell Pascal N
Institute of Medical Science, University of Toronto, Toronto, ON, Canada.
Department of Medical Imaging, University of Toronto, 263 McCaul Street, 4th Floor rm 409, Toronto, ON, M5T 1W7, Canada.
Int J Comput Assist Radiol Surg. 2023 Nov;18(11):2001-2012. doi: 10.1007/s11548-023-02965-4. Epub 2023 May 29.
Current artificial intelligence studies for supporting CT screening tasks depend on either supervised learning or detecting anomalies. However, the former involves a heavy annotation workload owing to requiring many slice-wise annotations (ground truth labels); the latter is promising, but while it reduces the annotation workload, it often suffers from lower performance. This study presents a novel weakly supervised anomaly detection (WSAD) algorithm trained based on scan-wise normal and anomalous annotations to provide better performance than conventional methods while reducing annotation workload.
Based on surveillance video anomaly detection methodology, feature vectors representing each CT slice were trained on an AR-Net-based convolutional network using a dynamic multiple-instance learning loss and a center loss function. The following two publicly available CT datasets were retrospectively analyzed: the RSNA brain hemorrhage dataset (normal scans: 12,862; scans with intracranial hematoma: 8882) and COVID-CT set (normal scans: 282; scans with COVID-19: 95).
Anomaly scores of each slice were successfully predicted despite inaccessibility to any slice-wise annotations. Slice-level area under the curve (AUC), sensitivity, specificity, and accuracy from the brain CT dataset were 0.89, 0.85, 0.78, and 0.79, respectively. The proposed method reduced the number of annotations in the brain dataset by 97.1% compared to an ordinary slice-level supervised learning method.
This study demonstrated a significant annotation reduction in identifying anomalous CT slices compared to a supervised learning approach. The effectiveness of the proposed WSAD algorithm was verified through higher AUC than existing anomaly detection techniques.
目前用于支持CT筛查任务的人工智能研究要么依赖监督学习，要么依赖异常检测。然而，前者由于需要许多逐切片注释（真实标签）而涉及繁重的注释工作量；后者很有前景，但虽然它减少了注释工作量，但其性能往往较低。本研究提出了一种基于扫描级正常和异常注释训练的新型弱监督异常检测（WSAD）算法，以在减少注释工作量的同时提供比传统方法更好的性能。
基于监控视频异常检测方法，使用动态多实例学习损失和中心损失函数，在基于AR-Net的卷积网络上训练表示每个CT切片的特征向量。回顾性分析了以下两个公开可用的CT数据集：RSNA脑出血数据集（正常扫描：12862例；颅内血肿扫描：8882例）和COVID-CT集（正常扫描：282例；COVID-19扫描：95例）。
尽管无法获得任何逐切片注释，但成功预测了每个切片的异常分数。脑CT数据集的切片级曲线下面积（AUC）、灵敏度、特异性和准确率分别为0.89、0.85、0.78和0.79。与普通切片级监督学习方法相比，该方法将脑数据集中的注释数量减少了97.1%。
本研究表明，与监督学习方法相比，在识别异常CT切片时注释显著减少。通过比现有异常检测技术更高的AUC验证了所提出的WSAD算法的有效性。
