[论文解读] Automated Diagnosis of Lymphoma with Digital Pathology Images Using Deep Learning

Hanadi El Achi, Т. А. Белоусова+7 more|PubMed|Oct 30, 2018

AI in cancer detection被引用 48

一句话总结

本文提出基于卷积神经网络（CNN）的系统，能够从数字病理图像在四个类别中诊断淋巴瘤，实现逐图像的高准确性，并在工作流程增强方面展示概念可行性。

ABSTRACT

Recent studies have shown promising results in using Deep Learning to detect malignancy in whole slide imaging, however, they were limited to just predicting a positive or negative finding for a specific neoplasm. We attempted to use Deep Learning with a convolutional neural network (CNN) algorithm to build a lymphoma diagnostic model for four diagnostic categories: (1) benign lymph node, (2) diffuse large B-cell lymphoma, (3) Burkitt lymphoma, and (4) small lymphocytic lymphoma. Our software was written in Python language. We obtained digital whole-slide images of Hematoxylin and Eosin stained slides of 128 cases including 32 cases for each diagnostic category. Four sets of 5 representative images, 40x40 pixels in dimension, were taken for each case. A total of 2,560 images were obtained from which 1,856 were used for training, 464 for validation, and 240 for testing. For each test set of 5 images, the predicted diagnosis was combined from the prediction of five images. The test results showed excellent diagnostic accuracy at 95% for image-by-image prediction and at 100% for set-by-set prediction. This preliminary study provided a proof of concept for incorporating automated lymphoma diagnostic screen into future pathology work-flow to augment the pathologists' productivity.

研究动机与目标

• 激发并评估在全片扫描（WSI）上的深度学习用于自动淋巴瘤诊断。
• 开发一种卷积神经网络方法，用于分类四个诊断类别。
• 在从WSIs提取的标注图像块数据集上评估性能。
• 为将自动筛查集成到病理工作流程中提供概念验证。

提出的方法

• 使用用 Python 实现的基于 CNN 的模型来处理图像块。
• 构建一个包含四个诊断类别（良性淋巴结、DLBCL、伯基特淋巴瘤、淋巴细胞性小淋巴瘤）的 128 例数据集。
• 提取四组代表性的 40x40 像素图像（每例 5 张），总计 2,560 个图像块。
• 将数据划分为训练集 1,856、验证集 464 和测试集 240 个图像块。

实验结果

研究问题

• RQ1 基于 CNN 的方法是否能够从 40x40 像素的图像块区分四个淋巴组织诊断类别？
• RQ2 在图像块层面与病例/集合层面的诊断性能如何？
• RQ3 将基于图像块的自动筛查作为病理工作流程中的辅助工具是否可行？

主要发现

• 测试集的逐图像预测准确率为 95%。
• 测试集的集合级（5 张图像）预测准确率为 10%。
• 数据集包含 128 例，且各类别表现均衡（每类 32 例）。
• 证明了在病理工作流程中整合自动淋巴瘤诊断的概念验证。
• 该方法为深度学习能够协助病理学家筛查淋巴瘤提供初步证据。