Semisupervised learning using denoising autoencoders for brain lesion detection and segmentation
Semisupervised learning using denoising autoencoders for brain lesion detection and segmentation
复制标题
DOI：
10.1117/1.jmi.4.4.041311
复制
发表时间：
2017-10-01
影响因子：
2.4
通讯作者：
Krishnamurthi, Ganapathy
中科院分区：
文献类型：
--
作者：
Alex, Varghese;Vaidhya, Kiran;Krishnamurthi, Ganapathy
这项工作探讨了使用去噪自动编码器（DAE）进行脑病变检测，分割和假阳性减少。堆叠去噪自动编码器（SDAE）使用大量未标记的患者体积进行预训练，并使用从有限数量的患者（n = 20，40，65）中提取的补丁进行微调。结果显示，即使使用20名标记患者对SDAE进行微调，性能损失也可以忽略不计。低级别胶质瘤（LGG）分割是使用迁移学习方法实现的，其中使用LGG图像块对用高级别胶质瘤数据预训练的网络进行微调。该网络还表现出良好的泛化能力，并在看不见的BraTS 2013和BraTS 2015测试数据上提供了良好的分割。该手稿还包括使用单层DAE，称为新奇检测器（ND）。ND接受了准确重建非病变斑块的培训。使用测试数据的重建误差图来定位病变。错误地图被证明分配独特的错误分布的胶质瘤的各种成分，使本地化。ND准确地学习非病变大脑，因为它也被证明对来自不同数据库的图像中的缺血性脑病变提供良好的分割性能。(c)2017年，美国光电仪器工程师学会（SPIE）。
The work explores the use of denoising autoencoders (DAEs) for brain lesion detection, segmentation, and false-positive reduction. Stacked denoising autoencoders (SDAEs) were pretrained using a large number of unlabeled patient volumes and fine-tuned with patches drawn from a limited number of patients (n = 20, 40, 65). The results show negligible loss in performance even when SDAE was fine-tuned using 20 labeled patients. Low grade glioma (LGG) segmentation was achieved using a transfer learning approach in which a network pretrained with high grade glioma data was fine-tuned using LGG image patches. The networks were also shown to generalize well and provide good segmentation on unseen BraTS 2013 and BraTS 2015 test data. The manuscript also includes the use of a single layer DAE, referred to as novelty detector (ND). ND was trained to accurately reconstruct nonlesion patches. The reconstruction error maps of test data were used to localize lesions. The error maps were shown to assign unique error distributions to various constituents of the glioma, enabling localization. The ND learns the nonlesion brain accurately as it was also shown to provide good segmentation performance on ischemic brain lesions in images from a different database. (c) 2017 Society of Photo-Optical Instrumentation Engineers (SPIE).