Sager Philipp, Näf Lukas, Vu Erwin, Fischer Tim, Putora Paul M, Ehret Felix, Fürweger Christoph, Schröder Christina, Förster Robert, Zwahlen Daniel R, Muacevic Alexander, Windisch Paul
Department of Radiation Oncology, Kantonsspital Winterthur, 8400 Winterthur, Switzerland.
Department of Radiology, Kantonsspital St. Gallen, 9007 St. Gallen, Switzerland.
Diagnostics (Basel). 2021 Sep 14;11(9):1676. doi: 10.3390/diagnostics11091676.
: Many proposed algorithms for tumor detection rely on 2.5/3D convolutional neural networks (CNNs) and the input of segmentations for training. The purpose of this study is therefore to assess the performance of tumor detection on single MRI slices containing vestibular schwannomas (VS) as a computationally inexpensive alternative that does not require the creation of segmentations. : A total of 2992 T1-weighted contrast-enhanced axial slices containing VS from the MRIs of 633 patients were labeled according to tumor location, of which 2538 slices from 539 patients were used for training a CNN (ResNet-34) to classify them according to the side of the tumor as a surrogate for detection and 454 slices from 94 patients were used for internal validation. The model was then externally validated on contrast-enhanced and non-contrast-enhanced slices from a different institution. Categorical accuracy was noted, and the results of the predictions for the validation set are provided with confusion matrices. : The model achieved an accuracy of 0.928 (95% CI: 0.869-0.987) on contrast-enhanced slices and 0.795 (95% CI: 0.702-0.888) on non-contrast-enhanced slices from the external validation cohorts. The implementation of Gradient-weighted Class Activation Mapping (Grad-CAM) revealed that the focus of the model was not limited to the contrast-enhancing tumor but to a larger area of the cerebellum and the cerebellopontine angle. : Single-slice predictions might constitute a computationally inexpensive alternative to training 2.5/3D-CNNs for certain detection tasks in medical imaging even without the use of segmentations. Head-to-head comparisons between 2D and more sophisticated architectures could help to determine the difference in accuracy, especially for more difficult tasks.
许多提出的肿瘤检测算法依赖于2.5/3D卷积神经网络（CNN）以及用于训练的分割输入。因此，本研究的目的是评估在包含前庭神经鞘瘤（VS）的单个MRI切片上进行肿瘤检测的性能，作为一种计算成本低廉的替代方法，该方法不需要创建分割。：根据肿瘤位置，对来自633名患者MRI的总共2992张包含VS的T1加权对比增强轴向切片进行了标记，其中来自539名患者的2538张切片用于训练CNN（ResNet-34），以根据肿瘤的侧别对其进行分类作为检测的替代指标，来自94名患者的454张切片用于内部验证。然后，该模型在来自不同机构的对比增强和非对比增强切片上进行外部验证。记录分类准确率，并通过混淆矩阵提供验证集的预测结果。：该模型在外部验证队列的对比增强切片上的准确率为0.928（95%CI：0.869-0.987），在非对比增强切片上的准确率为0.795（95%CI：0.702-0.888）。梯度加权类激活映射（Grad-CAM）的实施表明，该模型的关注点不仅限于对比增强的肿瘤，还包括小脑和小脑脑桥角的更大区域。：即使不使用分割，单切片预测对于医学成像中的某些检测任务可能构成一种计算成本低廉的替代方法，以替代训练2.5/3D-CNN。二维和更复杂架构之间的直接比较有助于确定准确率的差异，特别是对于更困难的任务。
