Cruz-Roa Angel Alfonso, Arevalo Ovalle John Edison, Madabhushi Anant, González Osorio Fabio Augusto
MindLab Research Group, Universidad Nacional de Colombia, Bogota, Colombia
Dept. of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA.
Med Image Comput Comput Assist Interv. 2013;16(Pt 2):403-10. doi: 10.1007/978-3-642-40763-5_50.
This paper presents and evaluates a deep learning architecture for automated basal cell carcinoma cancer detection that integrates (1) image representation learning, (2) image classification and (3) result interpretability. A novel characteristic of this approach is that it extends the deep learning architecture to also include an interpretable layer that highlights the visual patterns that contribute to discriminate between cancerous and normal tissues patterns, working akin to a digital staining which spotlights image regions important for diagnostic decisions. Experimental evaluation was performed on set of 1,417 images from 308 regions of interest of skin histopathology slides, where the presence of absence of basal cell carcinoma needs to be determined. Different image representation strategies, including bag of features (BOF), canonical (discrete cosine transform (DCT) and Haar-based wavelet transform (Haar)) and proposed learned-from-data representations, were evaluated for comparison. Experimental results show that the representation learned from a large histology image data set has the best overall performance (89.4% in F-measure and 91.4% in balanced accuracy), which represents an improvement of around 7% over canonical representations and 3% over the best equivalent BOF representation.
本文提出并评估了一种用于自动检测基底细胞癌的深度学习架构，该架构整合了：（1）图像表征学习；（2）图像分类；以及（3）结果可解释性。这种方法的一个新颖特点是，它扩展了深度学习架构，还包括一个可解释层，该层突出显示有助于区分癌组织和正常组织模式的视觉模式，其工作方式类似于数字染色，突出显示对诊断决策重要的图像区域。对来自皮肤组织病理学切片308个感兴趣区域的1417幅图像进行了实验评估，需要确定这些图像中是否存在基底细胞癌。为了进行比较，评估了不同的图像表征策略，包括特征袋（BOF）、规范（离散余弦变换（DCT）和基于哈尔的小波变换（Haar））以及提出的从数据中学习的表征。实验结果表明，从大型组织学图像数据集中学习到的表征具有最佳的整体性能（F值为89.4%，平衡准确率为91.4%），比规范表征提高了约7%，比最佳等效BOF表征提高了3%。
