Bai Jie, Jiang Huiyan, Li Siqi, Ma Xiaoqi
Northeastern University, Shenyang 110819, China.
Software College, Northeastern University, Shenyang 110819, China.
Biomed Res Int. 2019 Mar 21;2019:1065652. doi: 10.1155/2019/1065652. eCollection 2019.
Accurate classification for different non-Hodgkin lymphomas (NHL) is one of the main challenges in clinical pathological diagnosis due to its intrinsic complexity. Therefore, this paper proposes an effective classification model for three types of NHL pathological images, including mantle cell lymphoma (MCL), follicular lymphoma (FL), and chronic lymphocytic leukemia (CLL).
There are three main parts with respect to our model. First, NHL pathological images stained by hematoxylin and eosin (H&E) are transferred into blue ratio (BR) and Lab spaces, respectively. Then specific patch-level textural and statistical features are extracted from BR images and color features are obtained from Lab images both using a hierarchical way, yielding a set of hand-crafted representations corresponding to different image spaces. A random forest classifier is subsequently trained for patch-level classification. Second, H&E images are cropped and fed into a pretrained google inception net (GoogLeNet) for learning high-level representations and a softmax classifier is used for patch-level classification. Finally, three image-level classification strategies based on patch-level results are discussed including a novel method for calculating the weighted sum of patch results. Different classification results are fused at both feature 1 and image levels to obtain a more satisfactory result.
The proposed model is evaluated on a public IICBU Malignant Lymphoma Dataset and achieves an improved overall accuracy of 0.991 and area under the receiver operating characteristic curve of 0.998.
The experimentations demonstrate the significantly increased classification performance of the proposed model, indicating that it is a suitable classification approach for NHL pathological images.
由于其内在的复杂性，对不同的非霍奇金淋巴瘤（NHL）进行准确分类是临床病理诊断中的主要挑战之一。因此，本文提出了一种针对三种类型的NHL病理图像的有效分类模型，包括套细胞淋巴瘤（MCL）、滤泡性淋巴瘤（FL）和慢性淋巴细胞白血病（CLL）。
我们的模型主要有三个部分。首先，将苏木精和伊红（H&E）染色的NHL病理图像分别转换为蓝色比率（BR）和Lab空间。然后，从BR图像中提取特定的斑块级纹理和统计特征，并从Lab图像中以分层方式获取颜色特征，从而产生一组对应于不同图像空间的手工制作表示。随后训练随机森林分类器进行斑块级分类。其次，裁剪H&E图像并将其输入预训练的谷歌Inception网络（GoogLeNet）以学习高级表示，并使用softmax分类器进行斑块级分类。最后，讨论了基于斑块级结果的三种图像级分类策略，包括一种计算斑块结果加权和的新方法。在特征1和图像级别融合不同的分类结果以获得更满意的结果。
所提出的模型在公共IICBU恶性淋巴瘤数据集上进行了评估，总体准确率提高到0.991，受试者工作特征曲线下面积达到0.998。
实验证明了所提出模型的分类性能显著提高，表明它是一种适用于NHL病理图像的分类方法。
