Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography
Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography
复制标题
DOI：
10.1002/jmri.25842
复制
发表时间：
2018-04-01
影响因子：
4.4
通讯作者：
Abe, Osamu
中科院分区：
文献类型：
--
作者：
Nakao, Takahiro;Hanaoka, Shouhei;Abe, Osamu
背景计算机辅助检测（CAD）用于检测脑动脉瘤的有效性已有报道;目的开发一种基于深度卷积神经网络（CNN）和最大强度投影（MIP）算法的脑动脉瘤CAD系统，研究类型回顾性研究对象450例颅内动脉瘤患者，其中颅内动脉瘤患者100例，颅内动脉瘤患者100例，颅内动脉瘤患者100例，颅内动脉瘤患者100例。脑动脉瘤的诊断是基于MRA，这是作为大脑筛查程序的一部分进行的。场强/序列3 T MR扫描仪上的非对比增强3D飞行时间（TOF）MRA。评估在我们的CAD中，我们使用了CNN分类器，通过输入从体素周围的感兴趣体积（VOI）生成的MIP图像来预测每个体素是在动脉瘤内部还是外部。CNN使用手动输入的标签预先训练。我们评估了我们的方法，使用450例颅内动脉瘤，其中300例用于培训，50个参数调整，和100为最终evaluation.Statistical TestsFree-response receiver operating characteristics（FROC）analysis.ResultsOur CAD系统检测到94.2%（98/104）的动脉瘤与2.9假阳性每例（FPs/case）。在70%的灵敏度下，FP数/例为0.26。数据结论我们表明，CNN和MIP算法的组合是有用的颅内动脉瘤的检测。证据等级：4技术有效性：第1阶段J. Magn. Reson。Imaging 2018;47：948-953.
BackgroundThe usefulness of computer-assisted detection (CAD) for detecting cerebral aneurysms has been reported; therefore, the improved performance of CAD will help to detect cerebral aneurysms.PurposeTo develop a CAD system for intracranial aneurysms on unenhanced magnetic resonance angiography (MRA) images based on a deep convolutional neural network (CNN) and a maximum intensity projection (MIP) algorithm, and to demonstrate the usefulness of the system by training and evaluating it using a large dataset.Study TypeRetrospective study.SubjectsThere were 450 cases with intracranial aneurysms. The diagnoses of brain aneurysms were made on the basis of MRA, which was performed as part of a brain screening program.Field Strength/SequenceNoncontrast-enhanced 3D time-of-flight (TOF) MRA on 3T MR scanners.AssessmentIn our CAD, we used a CNN classifier that predicts whether each voxel is inside or outside aneurysms by inputting MIP images generated from a volume of interest (VOI) around the voxel. The CNN was trained in advance using manually inputted labels. We evaluated our method using 450 cases with intracranial aneurysms, 300 of which were used for training, 50 for parameter tuning, and 100 for the final evaluation.Statistical TestsFree-response receiver operating characteristic (FROC) analysis.ResultsOur CAD system detected 94.2% (98/104) of aneurysms with 2.9 false positives per case (FPs/case). At a sensitivity of 70%, the number of FPs/case was 0.26.Data ConclusionWe showed that the combination of a CNN and an MIP algorithm is useful for the detection of intracranial aneurysms. Level of Evidence: 4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;47:948-953.