Skin Lesion Classification Using CNNs With Patch-Based Attention and Diagnosis-Guided Loss Weighting
Skin Lesion Classification Using CNNs With Patch-Based Attention and Diagnosis-Guided Loss Weighting
复制标题
DOI：
10.1109/tbme.2019.2915839
复制
发表时间：
2020-02-01
影响因子：
4.6
通讯作者：
Schlaefer, Alexander
中科院分区：
文献类型：
--
作者：
Gessert, Nils;Sentker, Thilo;Schlaefer, Alexander
目的：本文解决了皮肤病变分类的两个关键问题。第一个问题是有效使用高分辨率图像和预先训练的标准架构进行图像分类。第二个问题是现实世界的多类数据集中遇到的高级不平衡。方法：为了使用高分辨率图像，我们提出了一种新颖的基于补丁的注意力架构，该架构提供小型高分辨率补丁之间的全局上下文。我们修改了三种预训练的架构并研究了基于补丁的注意力的性能。为了解决类别不平衡问题，我们比较了过采样、平衡批量采样和特定于类别的损失加权。此外，我们提出了一种新颖的诊断引导损失加权方法，该方法考虑了用于地面实况注释的方法。结果：我们基于补丁的注意力机制优于以前的方法，并将平均灵敏度提高了 $\text{7}\%$。类别平衡显着提高了平均灵敏度，并且我们表明，与正常损失平衡相比，我们的诊断引导损失加权方法将平均灵敏度提高了 $\text{3}\%$。结论：新颖的基于补丁的注意力机制可以集成到预训练的架构中，并提供局部补丁之间的全局上下文，同时优于其他基于补丁的方法。因此，预训练的架构可以轻松地与高分辨率图像一起使用，而无需下采样。新的诊断引导损失加权方法优于其他方法，并且可以在面临类别不平衡时进行有效的训练。意义：所提出的方法改进了自动皮肤病变分类。它们可以扩展到与高分辨率图像数据和类别不平衡相关的其他临床应用。
Objective: This paper addresses two key problems of skin lesion classification. The first problem is the effective use of high-resolution images with pretrained standard architectures for image classification. The second problem is the high-class imbalance encountered in real-world multi-class datasets. Methods: To use high-resolution images, we propose a novel patch-based attention architecture that provides global context between small, high-resolution patches. We modify three pretrained architectures and study the performance of patch-based attention. To counter class imbalance problems, we compare oversampling, balanced batch sampling, and class-specific loss weighting. Additionally, we propose a novel diagnosis-guided loss weighting method that takes the method used for ground-truth annotation into account. Results: Our patch-based attention mechanism outperforms previous methods and improves the mean sensitivity by $\text{7}\%$. Class balancing significantly improves the mean sensitivity and we show that our diagnosis-guided loss weighting method improves the mean sensitivity by $\text{3}\%$ over normal loss balancing. Conclusion: The novel patch-based attention mechanism can be integrated into pretrained architectures and provides global context between local patches while outperforming other patch-based methods. Hence, pretrained architectures can be readily used with high-resolution images without downsampling. The new diagnosis-guided loss weighting method outperforms other methods and allows for effective training when facing class imbalance. Significance: The proposed methods improve automatic skin lesion classification. They can be extended to other clinical applications where high-resolution image data and class imbalance are relevant.