Convolutional neural network to predict the local recurrence of giant cell tumor of bone after curettage based on pre-surgery magnetic resonance images
Convolutional neural network to predict the local recurrence of giant cell tumor of bone after curettage based on pre-surgery magnetic resonance images
复制标题
DOI：
10.1007/s00330-019-06082-2
复制
发表时间：
2019-10-01
影响因子：
5.9
通讯作者：
Xie, Xueqian
中科院分区：
文献类型：
--
作者：
He, Yifeng;Guo, Jiapan;Xie, Xueqian
目的应用深层卷积神经网络(CNN)对骨巨细胞瘤(GCTB)的MR表现和刮除后的临床特征进行局部复发预测。方法对56例经手术病理证实的GCTB患者进行了随访，随访时间为5.8年(2.0~9.5年)。通过数据增强获得两类MR数据集(经常性和非经常性GCTB)，对INITIMATION v3 CNN结构进行微调，并使用四次交叉验证来评估其泛化能力。选择28例(50%)作为CNN和4名放射科医生的训练数据集，其余28例(50%)作为测试数据集。将CNN预测与患者特征(年龄和肿瘤部位)相结合，建立预测GCTB复发的二元Logistic回归模型。用准确度和敏感度来评价预测效果。结果比较CNN、CNN回归和放射科医师，CNN和CNN回归模型的准确率分别为75.5%(95%CI 55.1~89.3%)和78.6%(59.0~91.7%)，高于放射科医师的64.3%(44.1~81.4%)。敏感度分别为85.7%(42.1~99.6%)和87.5%(47.3~99.7%)，高于放射科医师的58.3%(27.7~84.8%)(p<；0.05)。结论CNN对GCTB刮除术后复发有预测价值。结合患者特征的二元回归模型提高了其预测精度。
Objective To predict the local recurrence of giant cell bone tumors (GCTB) on MR features and the clinical characteristics after curettage using a deep convolutional neural network (CNN). Methods MR images were collected from 56 patients with histopathologically confirmed GCTB after curettage who were followed up for 5.8 years (range, 2.0 to 9.5 years). The inception v3 CNN architecture was fine-tuned by two categories of the MR datasets (recurrent and non-recurrent GCTB) obtained through data augmentation and was validated using fourfold cross-validation to evaluate its generalization ability. Twenty-eight cases (50%) were chosen as the training dataset for the CNN and four radiologists, while the remaining 28 cases (50%) were used as the test dataset. A binary logistic regression model was established to predict recurrent GCTB by combining the CNN prediction and patient features (age and tumor location). Accuracy and sensitivity were used to evaluate the prediction performance. Results When comparing the CNN, CNN regression, and radiologists, the accuracies of the CNN and CNN regression models were 75.5% (95% CI 55.1 to 89.3%) and 78.6% (59.0 to 91.7%), respectively, which were higher than the 64.3% (44.1 to 81.4%) accuracy of the radiologists. The sensitivities were 85.7% (42.1 to 99.6%) and 87.5% (47.3 to 99.7%), respectively, which were higher than the 58.3% (27.7 to 84.8%) sensitivity of the radiologists (p < 0.05). Conclusion The CNN has the potential to predict recurrent GCTB after curettage. A binary regression model combined with patient characteristics improves its prediction accuracy.