[论文解读] Coronavirus (COVID-19) Classification using CT Images by Machine Learning Methods

Mücahid Barstuğan, Umut Özkaya+1 more|arXiv (Cornell University)|Mar 20, 2020

COVID-19 diagnosis using AI参考文献 14被引用 334

一句话总结

该论文利用传统特征提取方法和 SVM 分类器，从腹部 CT 图像块中进行 COVID-19 的早期检测，并评估多种交叉验证方案。在 10 折 CV 下，使用 GLSZM 特征获得最佳结果，准确率为 99.68%。

ABSTRACT

This study presents early phase detection of Coronavirus (COVID-19), which is named by World Health Organization (WHO), by machine learning methods. The detection process was implemented on abdominal Computed Tomography (CT) images. The expert radiologists detected from CT images that COVID-19 shows different behaviours from other viral pneumonia. Therefore, the clinical experts specify that COVİD-19 virus needs to be diagnosed in early phase. For detection of the COVID-19, four different datasets were formed by taking patches sized as 16x16, 32x32, 48x48, 64x64 from 150 CT images. The feature extraction process was applied to patches to increase the classification performance. Grey Level Co-occurrence Matrix (GLCM), Local Directional Pattern (LDP), Grey Level Run Length Matrix (GLRLM), Grey-Level Size Zone Matrix (GLSZM), and Discrete Wavelet Transform (DWT) algorithms were used as feature extraction methods. Support Vector Machines (SVM) classified the extracted features. 2-fold, 5-fold and 10-fold cross-validations were implemented during the classification process. Sensitivity, specificity, accuracy, precision, and F-score metrics were used to evaluate the classification performance. The best classification accuracy was obtained as 99.68% with 10-fold cross-validation and GLSZM feature extraction method.

研究动机与目标

• 使用 CT 图像和机器学习推动早期阶段 COVID-19 的检测。
• 从 CT 图像形成基于 patch 的特征提取管线。
• 评估多种特征提取器和交叉验证方案，以评估分类性能。
• 比较不同 patch 大小和特征集下的性能，以识别表现最好的配置。

提出的方法

• 从 150 张 CT 图像中提取的 16x16、32x32、48x48、64x64 大小的 CT 图像块构建四个数据集。
• 对图像块应用包括 GLSZM、GLCM、LDP、GLRLM 和 DWT 的特征提取方法。
• 使用支持向量机（SVM）对提取的特征进行分类。
• 使用 2-fold、5-fold 和 10-fold 交叉验证进行评估。
• 以敏感性、特异性、准确率、精确度和 F-score（F1 分数）等指标评估性能。

实验结果

研究问题

• RQ1 CT 图像块是否能利用选定的纹理和变换特征通过 SVM 将 COVID-19 与其他病毒性肺炎区分开？
• RQ2 在交叉验证下，哪种块大小和哪种特征提取方法能获得最高的分类准确率？
• RQ3 不同的交叉验证方案如何影响在 CT 块上 COVID-19 检测的报告性能指标？

主要发现

• 最佳分类准确率为 99.68%，使用 GLSZM 特征并进行 10 折交叉验证。
• 在所有实验中，GLSZM 是该任务的表现最佳的特征提取方法。
• 使用 2 折、5 折、10 折的交叉验证方案来验证结果的鲁棒性。
• 该研究使用多种纹理和变换特征以增强对 COVID-19 与其他病毒性肺炎在 CT 块上的区分能力。