Department of Radiology, Stanford University School of Medicine, 300 Pasteur Drive, H3630, Stanford, CA, 94305.
Department of Radiology, Stanford University School of Medicine, 300 Pasteur Drive, H3630, Stanford, CA, 94305; Division of Interventional Radiology, Stanford University School of Medicine, 300 Pasteur Drive, H3630, Stanford, CA, 94305.
J Vasc Interv Radiol. 2020 Jan;31(1):66-73. doi: 10.1016/j.jvir.2019.05.026. Epub 2019 Sep 18.
To demonstrate the feasibility and evaluate the performance of a deep-learning convolutional neural network (CNN) classification model for automated identification of different types of inferior vena cava (IVC) filters on radiographs.
In total, 1,375 cropped radiographic images of 14 types of IVC filters were collected from patients enrolled in a single-center IVC filter registry, with 139 images withheld as a test set and the remainder used to train and validate the classification model. Image brightness, contrast, intensity, and rotation were varied to augment the training set. A 50-layer ResNet architecture with fixed pre-trained weights was trained using a soft margin loss over 50 epochs. The final model was evaluated on the test set.
The CNN classification model achieved a F1 score of 0.97 (0.92-0.99) for the test set overall and of 1.00 for 10 of 14 individual filter types. Of the 139 test set images, 4 (2.9%) were misidentified, all mistaken for other filter types that appear highly similar. Heat maps elucidated salient features for each filter type that the model used for class prediction.
A CNN classification model was successfully developed to identify 14 types of IVC filters on radiographs and demonstrated high performance. Further refinement and testing of the model is necessary before potential real-world application.
展示一种基于深度学习卷积神经网络（CNN）的分类模型用于自动识别 X 线下不同类型下腔静脉滤器的可行性，并对其性能进行评估。
共从一家中心静脉滤器注册中心的患者中收集了 14 种类型的 1375 张下腔静脉滤器的裁剪图像，其中 139 张图像被保留作为测试集，其余图像用于训练和验证分类模型。通过改变图像的亮度、对比度、强度和旋转来扩充训练集。使用固定的预训练权重的 50 层 ResNet 架构，通过 50 个周期的软边界损失进行训练。最终模型在测试集上进行评估。
该 CNN 分类模型对测试集的整体 F1 评分为 0.97（0.92-0.99），对 14 种滤器类型中的 10 种达到了 1.00 的准确率。在 139 张测试集中，有 4 张（2.9%）被错误识别，均被错误地识别为其他高度相似的滤器类型。热图阐明了模型用于分类预测的每个滤器类型的显著特征。
成功开发了一种用于识别 X 线下 14 种下腔静脉滤器的 CNN 分类模型，并表现出了较高的性能。在潜在的实际应用之前，还需要对模型进行进一步的细化和测试。
