Deep learning for segmentation of brain tumors: Impact of cross-institutional training and testing
Deep learning for segmentation of brain tumors: Impact of cross-institutional training and testing
复制标题
DOI：
10.1002/mp.12752
复制
发表时间：
2018-03-01
期刊：
影响因子：
3.8
通讯作者：
Mazurowski, Maciej A.
中科院分区：
文献类型：
--
作者：
AlBadawy, Ehab A.;Saha, Ashirbani;Mazurowski, Maciej A.
背景和目的卷积神经网络（CNN）通常用于脑肿瘤的分割。在这项工作中，我们评估跨机构的培训对CNNs.MethodsWe的性能的影响选择了44胶质母细胞瘤（GBM）患者从两个机构的癌症成像档案数据集。通过勾勒每个肿瘤成分来手动注释图像以形成地面实况。为了自动分割每个患者的肿瘤，我们训练了三个CNN：（a）一个使用来自同一机构的患者数据作为测试数据，（B）一个使用来自其他机构的患者数据，（c）一个使用来自两个机构的患者数据。使用Dice相似性系数以及地面实况和自动分割之间的平均Hausdorff距离来评估训练模型的性能。10倍交叉验证方案被用来比较不同方法的性能。结果模型的性能显着下降（P < 0.0001）当它接受来自不同机构的数据训练时（骰子系数：0.68 0.19和0.59 +/- 0.19）与来自同一机构的数据进行训练相比（骰子系数：0.72 +/- 0.17和0.76 +/- 0.12）。这种趋势持续分割的整个肿瘤，以及它的个别components.ConclusionsThere是一个非常强大的影响选择数据的训练性能的CNN在多机构设置。要确定这种影响背后的原因，需要进行更多的全面调查。
Background and purposeConvolutional neural networks (CNNs) are commonly used for segmentation of brain tumors. In this work, we assess the effect of cross-institutional training on the performance of CNNs.MethodsWe selected 44 glioblastoma (GBM) patients from two institutions in The Cancer Imaging Archive dataset. The images were manually annotated by outlining each tumor component to form ground truth. To automatically segment the tumors in each patient, we trained three CNNs: (a) one using data for patients from the same institution as the test data, (b) one using data for the patients from the other institution and (c) one using data for the patients from both of the institutions. The performance of the trained models was evaluated using Dice similarity coefficients as well as Average Hausdorff Distance between the ground truth and automatic segmentations. The 10-fold cross-validation scheme was used to compare the performance of different approaches.ResultsPerformance of the model significantly decreased (P < 0.0001) when it was trained on data from a different institution (dice coefficients: 0.68 0.19 and 0.59 +/- 0.19) as compared to training with data from the same institution (dice coefficients: 0.72 +/- 0.17 and 0.76 +/- 0.12). This trend persisted for segmentation of the entire tumor as well as its individual components.ConclusionsThere is a very strong effect of selecting data for training on performance of CNNs in a multi-institutional setting. Determination of the reasons behind this effect requires additional comprehensive investigation.