当前位置: X-MOL 学术Bio-Medical Mater. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Multiview framework using a 3D residual network for pulmonary micronodule malignancy risk classification.
Bio-Medical Materials and Engineering ( IF 1.0 ) Pub Date : 2020-09-04 , DOI: 10.3233/bme-206005
Yujie Yang 1, 2 , Qianqian Zhang 1, 2
Affiliation  

BACKGROUND:Pulmonary micronodules account for 80% of all lung nodules. Generally, pulmonary micronodules in the early stages can be detected on thoracic computed tomography (CT) scans. Early diagnosis is crucial for improving the patient’s survival rate. OBJECTIVE:This paper aims to estimate the malignancy risk of pulmonary micronodules and potentially improve the survival rate. METHODS:We extract 3D features of the CT images to obtain richer characteristics. Because superior performance can be achieved by having deep layers, we apply a 3D residual network (3D-ResNet) to classify the pulmonary micronodule. We construct a framework by using three parallel ResNets whose inputs are CT images in different regions of interest, i.e., the multiview of the image. To further evaluate the applicability of the framework, we make a five-category classification and achieve good performance. RESULTS:By fusing different characteristics from three views, we achieve the area under the receiver operating characteristic curve (AUC) of 0.9681. Based on the results of the experiments, our 3D-ResNet has a better performances than 3D-VGG and 3D-Inception in terms of precision (the increase rates are 13.7% and 7.4%), AUC (the increase rates are 15.8% and 5.3%), and accuracy (the increase rates are 14.3% and 4.5%). Meanwhile, the recall performance is close to that of the 3D-Inception network. CONCLUSION:Overall, the framework we propose has applicability and feasibility in pulmonary micronodule classification.

中文翻译:

使用 3D 残差网络进行肺微结节恶性肿瘤风险分类的多视图框架。

背景:肺微结节占所有肺结节的80%。一般来说,早期的肺部微结节可以在胸部计算机断层扫描 (CT) 扫描中检测到。早期诊断对于提高患者的生存率至关重要。目的:本文旨在评估肺微结节的恶性风险并提高生存率。方法:我们提取CT图像的3D特征以获得更丰富的特征。由于可以通过深层实现卓越的性能,我们应用 3D 残差网络 (3D-ResNet) 对肺微结节进行分类。我们通过使用三个并行的 ResNets 构建一个框架,其输入是不同感兴趣区域的 CT 图像,即图像的多视图。为了进一步评估该框架的适用性,我们进行了五类分类并取得了良好的性能。结果:通过融合三个视图的不同特征,我们获得了 0.9681 的受试者工作特征曲线下面积 (AUC)。根据实验结果,我们的 3D-ResNet 在精度(提升率分别为 13.7% 和 7.4%)、AUC(提升率分别为 15.8% 和 5.3 %)和准确率(增幅分别为 14.3% 和 4.5%)。同时,召回性能接近 3D-Inception 网络。结论:总体而言,我们提出的框架在肺微结节分类中具有适用性和可行性。我们实现了 0.9681 的受试者工作特征曲线下面积 (AUC)。根据实验结果,我们的 3D-ResNet 在精度(提升率分别为 13.7% 和 7.4%)、AUC(提升率分别为 15.8% 和 5.3 %)和准确率(增幅分别为 14.3% 和 4.5%)。同时,召回性能接近 3D-Inception 网络。结论:总体而言,我们提出的框架在肺微结节分类中具有适用性和可行性。我们实现了 0.9681 的受试者工作特征曲线下面积 (AUC)。根据实验结果,我们的 3D-ResNet 在精度(增长率分别为 13.7% 和 7.4%)、AUC(增长率分别为 15.8% 和 5.3 %)和准确率(增幅分别为 14.3% 和 4.5%)。同时,召回性能接近 3D-Inception 网络。结论:总体而言,我们提出的框架在肺微结节分类中具有适用性和可行性。召回性能接近 3D-Inception 网络。结论:总体而言,我们提出的框架在肺微结节分类中具有适用性和可行性。召回性能接近 3D-Inception 网络。结论:总体而言,我们提出的框架在肺微结节分类中具有适用性和可行性。
更新日期:2020-09-05
down
wechat
bug