[论文解读] Method to Classify Skin Lesions using Dermoscopic images

Charan, Dusa Sai, Nadipineni, Hemanth+2 more|arXiv (Cornell University)|Aug 21, 2020

Cutaneous Melanoma Detection and Management被引用 28

一句话总结

本文提出了一种基于深度学习的自动皮肤病变分类方法，利用皮肤镜图像，采用双路径卷积神经网络（CNN），结合数据增强、基于U-Net的分割以及10折交叉验证，以提高准确性。该模型实现了0.886的峰值准确率，表明预处理和稳健的训练策略在不平衡医学图像数据上显著提升了性能。

ABSTRACT

Skin cancer is the most common cancer in the existing world constituting one-third of the cancer cases. Benign skin cancers are not fatal, can be cured with proper medication. But it is not the same as the malignant skin cancers. In the case of malignant melanoma, in its peak stage, the maximum life expectancy is less than or equal to 5 years. But, it can be cured if detected in early stages. Though there are numerous clinical procedures, the accuracy of diagnosis falls between 49% to 81% and is time-consuming. So, dermoscopy has been brought into the picture. It helped in increasing the accuracy of diagnosis but could not demolish the error-prone behaviour. A quick and less error-prone solution is needed to diagnose this majorly growing skin cancer. This project deals with the usage of deep learning in skin lesion classification. In this project, an automated model for skin lesion classification using dermoscopic images has been developed with CNN(Convolution Neural Networks) as a training model. Convolution neural networks are known for capturing features of an image. So, they are preferred in analyzing medical images to find the characteristics that drive the model towards success. Techniques like data augmentation for tackling class imbalance, segmentation for focusing on the region of interest and 10-fold cross-validation to make the model robust have been brought into the picture. This project also includes usage of certain preprocessing techniques like brightening the images using piece-wise linear transformation function, grayscale conversion of the image, resize the image. This project throws a set of valuable insights on how the accuracy of the model hikes with the bringing of new input strategies, preprocessing techniques. The best accuracy this model could achieve is 0.886.

研究动机与目标

• 开发一种自动化、准确且鲁棒的深度学习模型，用于从皮肤镜图像中分类皮肤病变。
• 通过替代人工临床评估，减少皮肤癌诊断中的人为错误和时间消耗。
• 通过仿射变换等数据增强技术，解决皮肤病变数据集中类别不平衡的问题。
• 通过应用基于U-Net的图像分割，提高模型对病变区域的关注度。
• 通过10折交叉验证，提升模型的泛化能力和鲁棒性。

提出的方法

• 设计了一种双路径CNN架构，用于从皮肤镜图像中提取分层特征，其中一条路径处理原始图像，另一条路径处理分割后的病变区域。
• 图像预处理包括灰度化、图像缩放以及分段线性变换，以增强亮度和一致性。
• 应用基于U-Net的语义分割，将病变区域与背景分离，减少非病变皮肤区域的干扰。
• 使用仿射变换（如旋转、缩放等）进行数据增强，以增加少数类样本数量并提升泛化能力。
• 采用10折交叉验证评估模型性能，最终准确率计算为各折准确率的平均值：$ Acc = (\sum_{i=1}^{K} Acc_i)/K $。
• 使用ISIC 2018皮肤病变数据集进行模型训练与验证，通过超参数调优以优化性能。

实验结果

研究问题

• RQ1 在深度学习中引入图像分割如何影响皮肤病变分类的准确率？
• RQ2 数据增强在不平衡皮肤病变数据集上在多大程度上提升了模型性能？
• RQ3 双路径CNN架构与单路径CNN在分类皮肤镜图像方面有何差异？
• RQ4 10折交叉验证在多大程度上显著提升了皮肤病变分类中模型的鲁棒性和泛化能力？
• RQ5 诸如亮度调节和图像缩放等预处理技术对模型准确率有何影响？

主要发现

• 采用完整预处理、分割、数据增强和10折交叉验证的模型实现了最高准确率0.886。
• 引入分割后，准确率从无分割时的0.583提升至0.886，表明聚焦于病变区域可显著增强特征学习。
• 当与交叉验证结合时，数据增强使准确率从0.583提升至0.886，表明其在处理类别不平衡问题中起着关键作用。
• 双路径CNN模型优于单路径模型，准确率从0.814提升至0.886，表明双路径特征提取可提升性能。
• 展平层的神经元数量与准确率之间不存在线性相关性，表明在某一性能阈值后，增加参数带来的收益递减。
• U-Net分割模型有效分离了病变区域，减少了噪声，提升了模型对相关特征的关注度。