Dynamic Deep Networks for Retinal Vessel Segmentation
Dynamic Deep Networks for Retinal Vessel Segmentation
复制标题
DOI：
10.3389/fcomp.2020.00035
复制
发表时间：
2020-08-26
影响因子：
2.6
通讯作者：
Estrada, Rolando
中科院分区：
文献类型：
--
作者：
Khanal, Aashis;Estrada, Rolando
深度学习最近在视网膜血管分割方面取得了令人印象深刻的成果。然而，最先进的方法往往是保守的，倾向于精确度而不是召回率。因此，他们倾向于对模糊的血管进行不足的分割，低估较厚的血管的宽度，甚至错过整个血管。为了解决这一局限性，我们提出了一种深度神经网络的随机训练方案，该方案稳健地平衡了精度和召回率。首先，我们用在每次训练迭代中波动的损失函数中的动态类权重来训练我们的深层网络。这种随机方法-我们相信这种方法也适用于许多其他机器学习问题-迫使网络学习平衡的分类。其次，我们将分割过程分解为两个步骤。在我们的流程的前半部分，我们估计每个像素的可能性，然后使用这些可能性来分割明显是血管或背景的像素。在流水线的后半部分，我们使用第二个网络对图像中的模糊区域进行分类。我们提出的方法通过学习假阳性率和假阴性率之间的稳健平衡，在五个视网膜数据集-DRIVE、STARE、Chase-DB、AV-Wide和VEVIO上获得了最先进的结果。我们新的训练范例使神经网络对类别比率的样本间差异更稳健，我们相信这将被证明对于具有稀疏训练数据的环境特别有效，例如医学图像分析。此外，我们是第一个报告AV范围数据集的分割结果的人，我们已经公开了该数据集的地面真实注释。这项工作的实现可以在上找到。
Deep learning has recently yielded impressive gains in retinal vessel segmentation. However, state-of-the-art methods tend to be conservative, favoring precision over recall. Thus, they tend to under-segment faint vessels, underestimate the width of thicker vessels, or even miss entire vessels. To address this limitation, we propose a stochastic training scheme for deep neural networks that robustly balances precision and recall. First, we train our deep networks with dynamic class weights in the loss function that fluctuate during each training iteration. This stochastic approach-which we believe is applicable to many other machine learning problems-forces the network to learn a balanced classification. Second, we decouple the segmentation process into two steps. In the first half of our pipeline, we estimate the likelihood of every pixel and then use these likelihoods to segment pixels that are clearly vessel or background. In the latter part of our pipeline, we use a second network to classify the ambiguous regions in the image. Our proposed method obtained state-of-the-art results on five retinal datasets-DRIVE, STARE, CHASE-DB, AV-WIDE, and VEVIO-by learning a robust balance between false positive and false negative rates. Our novel training paradigm makes a neural network more robust to inter-sample differences in class ratios, which we believe will prove particularly effective for settings with sparse training data, such as medical image analysis. In addition, we are the first to report segmentation results on the AV-WIDE dataset, and we have made the ground-truth annotations for this dataset publicly available. An implementation of this work can be found at .