Combo loss: Handling input and output imbalance in multi-organ segmentation
Combo loss: Handling input and output imbalance in multi-organ segmentation
复制标题
DOI：
10.1016/j.compmedimag.2019.04.005
复制
发表时间：
2019-07-01
影响因子：
5.7
通讯作者：
Hamarneh, Ghassan
中科院分区：
文献类型：
--
作者：
Taghanaki, Saeid Asgari;Zheng, Yefeng;Hamarneh, Ghassan
同时分割来自不同医学成像模式的多个器官是一项至关重要的任务，因为它可用于计算机辅助诊断、计算机辅助手术和治疗计划。由于深度学习的最新进展，为此目的已成功引入了几种用于医学图像分割的深度神经网络。在本文中，我们专注于学习标记体素的深度多器官分割网络。特别是，我们检查了损失函数的关键选择，以处理困扰学习模型输入和输出的臭名昭著的不平衡问题。输入不平衡是指输入训练样本中的类不平衡（即嵌入大量背景体素中的小前景物体，以及不同大小的器官）。输出不平衡是指推理模型的误报和漏报之间的不平衡。为了解决训练和推理过程中的两种不平衡问题，我们引入了一种新的基于课程学习的损失函数。具体来说，我们利用 Dice 相似性系数来阻止模型参数保持在不良的局部最小值，同时通过使用交叉熵项惩罚误报/误报来逐渐学习更好的模型参数。我们在三个数据集上评估了所提出的损失函数：具有 5 个目标器官的全身正电子发射断层扫描 (PET) 扫描、磁共振成像 (MRI) 前列腺扫描以及具有单个目标器官（即左心室）的超声心动图图像。我们表明，具有所提出的综合损失函数的简单网络架构可以超越最先进的方法，并且当使用我们提出的损失时，可以改善竞争方法的结果。 (C) 2019 Elsevier Ltd. 保留所有权利。
Simultaneous segmentation of multiple organs from different medical imaging modalities is a crucial task as it can be utilized for computer-aided diagnosis, computer-assisted surgery, and therapy planning. Thanks to the recent advances in deep learning, several deep neural networks for medical image segmentation have been introduced successfully for this purpose. In this paper, we focus on learning a deep multi-organ segmentation network that labels voxels. In particular, we examine the critical choice of a loss function in order to handle the notorious imbalance problem that plagues both the input and output of a learning model. The input imbalance refers to the class-imbalance in the input training samples (i.e., small foreground objects embedded in an abundance of background voxels, as well as organs of varying sizes). The output imbalance refers to the imbalance between the false positives and false negatives of the inference model. In order to tackle both types of imbalance during training and inference, we introduce a new curriculum learning based loss function. Specifically, we leverage Dice similarity coefficient to deter model parameters from being held at bad local minima and at the same time gradually learn better model parameters by penalizing for false positives/negatives using a cross entropy term. We evaluated the proposed loss function on three datasets: whole body positron emission tomography (PET) scans with 5 target organs, magnetic resonance imaging (MRI) prostate scans, and ultrasound echocardigraphy images with a single target organ i.e., left ventricular. We show that a simple network architecture with the proposed integrative loss function can outperform state-of-the-art methods and results of the competing methods can be improved when our proposed loss is used. (C) 2019 Elsevier Ltd. All rights reserved.