[论文解读] Multiple Sclerosis Lesion Segmentation from Brain MRI via Fully Convolutional Neural Networks

Snehashis Roy, John A. Butman+3 more|arXiv (Cornell University)|Mar 24, 2018

Brain Tumor Detection and Classification参考文献 61被引用 62

一句话总结

一个全卷积 CNN（FLEXCONN）通过并行多对比通道实现对多对比 MRI 的 MS 病灶分割，在 ISBI-61 和 MS-100 数据集上优于多种最新方法。

ABSTRACT

Multiple Sclerosis (MS) is an autoimmune disease that leads to lesions in the central nervous system. Magnetic resonance (MR) images provide sufficient imaging contrast to visualize and detect lesions, particularly those in the white matter. Quantitative measures based on various features of lesions have been shown to be useful in clinical trials for evaluating therapies. Therefore robust and accurate segmentation of white matter lesions from MR images can provide important information about the disease status and progression. In this paper, we propose a fully convolutional neural network (CNN) based method to segment white matter lesions from multi-contrast MR images. The proposed CNN based method contains two convolutional pathways. The first pathway consists of multiple parallel convolutional filter banks catering to multiple MR modalities. In the second pathway, the outputs of the first one are concatenated and another set of convolutional filters are applied. The output of this last pathway produces a membership function for lesions that may be thresholded to obtain a binary segmentation. The proposed method is evaluated on a dataset of 100 MS patients, as well as the ISBI 2015 challenge data consisting of 14 patients. The comparison is performed against four publicly available MS lesion segmentation methods. Significant improvement in segmentation quality over the competing methods is demonstrated on various metrics, such as Dice and false positive ratio. While evaluating on the ISBI 2015 challenge data, our method produces a score of 90.48, where a score of 90 is considered to be comparable to a human rater.

研究动机与目标

• 展示一种用于多对比 MR 图像的 MS 病灶分割的全卷积 CNN 架构。
• 评估使用大尺寸二维补丁和基于卷积的成员预测对分割精度的影响。
• 在公开和私有数据集上将 FLEXCONN 与成熟的 MS 病灶分割方法进行比较。

提出的方法

• 使用对多对比输入（MPRAGE 和 FLAIR）的两条并行卷积通道来提取特征。
• 拼接通道输出并应用第二条卷积通道以生成病灶成员映射。
• 以基于补丁的监督进行训练，其中训练成员通过高斯模糊的手动掩模生成。
• 相比于使用最终全连接层的预测，更倾向于全卷积预测以提高空间一致性并减少假阳性。
• 使用 VAL-28 验证集对补丁大小（35x35）和通道深度（五个滤波器组）进行优化。
• 对成员映射进行阈值化（0.30）以得到最终的二值分割；使用 Dice、LFPR、PPV 和 VD 进行评估。

实验结果

研究问题

• RQ1 全卷积 CNN 能否在 MS 病灶分割上提升相对于传统方法的准确性？
• RQ2 哪种补丁大小和网络深度能够在 Dice 分数和假阳性之间获得最佳平衡？
• RQ3 FLEXCONN 与 LesionTOADS、OASIS、LST、S3DL 在 ISBI-61 和 MS-100 数据集上的比较如何？

主要发现

• FLEXCONN 在 MS-100 上的 Dice 更高、病灶假阳性率（LFPR）和体积差异（VD）更低，优于竞争方法。
• 在 MS-100 上，FLEXCONN 的中位数指标为：Dice 0.5639，LFPR 0.3077，PPV 0.6040，VD 0.1978。
• 在 ISBI-61 上，FLEXCONN 报告了所评估方法中最高的分数：Score 90.48，表明分割精度可与人工多评者一致。
• 与基于 FC 的预测相比，卷积成员方法在皮质区域的假阳性更少且对整个体积推断更快（每个 3D 体积几秒）。
• 在 35x35 的补丁和 5 路径深度情况下获得最佳性能；较小的补丁在 Dice 和 LFPR 上表现较差。