Universal Loss Reweighting to Balance Lesion Size Inequality in 3D Medical Image Segmentation
Universal Loss Reweighting to Balance Lesion Size Inequality in 3D Medical Image Segmentation
复制标题
通用损失重新加权平衡 3D 医学图像分割中的病变大小不平等
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
M. Belyaev
中科院分区：
文献类型：
--
作者：
B. Shirokikh;A. Shevtsov;Anvar Kurmukov;A. Dalechina;Egor Krivov;V. Kostjuchenko;A. Golanov;M. Belyaev
在许多医学图像分割任务中，目标不平衡会影响最近的深度学习方法的性能。这是一个双重问题：类别不平衡-阳性类别（病变）大小与阴性类别（非病变）大小相比;病变大小不平衡-大病变遮盖小病变（在每张图像有多个病变的情况下）。虽然前者在多个作品中得到了解决，但后者缺乏调查。我们提出了一种损失重新加权的方法，以提高网络检测小病变的能力。在学习过程中，我们为每个图像体素分配权重。分配的权重与病变体积成反比，因此较小的病变获得较大的权重。我们报告了我们的方法对众所周知的损失函数的好处，包括骰子损失，焦点损失和非对称相似性损失。此外，我们比较我们的结果与其他重新加权技术：加权交叉熵和广义骰子损失。我们的实验表明，逆加权大大提高了检测质量，同时保留了国家的最先进的水平上的描绘质量。我们为两个公开的CT图像数据集发布了一个完整的实验管道：LiTS和LUNA 16（此https URL）。我们还显示了一个私人数据库的多个脑转移瘤划定任务的MR图像的结果。
Target imbalance affects the performance of recent deep learning methods in many medical image segmentation tasks. It is a twofold problem: class imbalance - positive class (lesion) size compared to negative class (non-lesion) size; lesion size imbalance - large lesions overshadows small ones (in the case of multiple lesions per image). While the former was addressed in multiple works, the latter lacks investigation. We propose a loss reweighting approach to increase the ability of the network to detect small lesions. During the learning process, we assign a weight to every image voxel. The assigned weights are inversely proportional to the lesion volume, thus smaller lesions get larger weights. We report the benefit from our method for well-known loss functions, including Dice Loss, Focal Loss, and Asymmetric Similarity Loss. Additionally, we compare our results with other reweighting techniques: Weighted Cross-Entropy and Generalized Dice Loss. Our experiments show that inverse weighting considerably increases the detection quality, while preserves the delineation quality on a state-of-the-art level. We publish a complete experimental pipeline for two publicly available datasets of CT images: LiTS and LUNA16 (this https URL). We also show results on a private database of MR images for the task of multiple brain metastases delineation.