Reliable Learning with PDE-Based CNNs and DenseNets for Detecting COVID-19, Pneumonia, and Tuberculosis from Chest X-Ray Images

It has recently been shown that the interpretation by partial differential equations (PDEs) of a class of convolutional neural networks (CNNs) supports definition of architectures such as parabolic and hyperbolic networks. These networks have provable properties regarding the stability against the perturbations of the input features. Aiming for robustness, we tackle the problem of detecting changes in chest X-ray images that may be suggestive of COVID-19 with parabolic and hyperbolic CNNs and with domain-specific transfer learning. To this end, we compile public data on patients diagnosed with COVID-19, pneumonia, and tuberculosis, along with normal chest X-ray images. The negative impact of the small number of COVID-19 images is reduced by applying transfer learning in several ways. For the parabolic and hyperbolic networks, we pretrain the networks on normal and pneumonia images and further use the obtained weights as the initializers for the networks to discriminate between COVID-19, pneumonia, tuberculosis, and normal aspects. For DenseNets, we apply transfer learning twice. First, the ImageNet pretrained weights are used to train on the CheXpert dataset, which includes 14 common radiological observations (e.g., lung opacity, cardiomegaly, fracture, support devices). Then, the weights are used to initialize the network which detects COVID-19 and the three other classes. The resulting networks are compared in terms of how well they adapt to the small number of COVID-19 images. According to our quantitative and qualitative analysis, the resulting networks are more reliable compared to those obtained by direct training on the targeted dataset.

[1]  Joseph Paul Cohen,et al.  COVID-19 Image Data Collection , 2020, ArXiv.

[2]  Shi Qiu,et al.  The ensemble deep learning model for novel COVID-19 on CT images , 2020, Applied Soft Computing.

[3]  Mamun Bin Ibne Reaz,et al.  Can AI Help in Screening Viral and COVID-19 Pneumonia? , 2020, IEEE Access.

[4]  Jon Kleinberg,et al.  Transfusion: Understanding Transfer Learning for Medical Imaging , 2019, NeurIPS.

[5]  Phillip M Cheng,et al.  Tackling the Radiological Society of North America Pneumonia Detection Challenge. , 2019, AJR. American journal of roentgenology.

[6]  Eldad Haber,et al.  Deep Neural Networks Motivated by Partial Differential Equations , 2018, Journal of Mathematical Imaging and Vision.

[7]  Bram van Ginneken,et al.  COVID-19 on Chest Radiographs: A Multireader Evaluation of an Artificial Intelligence System , 2020, Radiology.

[8]  F. Pfeiffer,et al.  Efficient Deep Network Architectures for Fast Chest X-Ray Tuberculosis Screening and Visualization , 2019, Scientific Reports.

[9]  Stefan Jaeger,et al.  Two public chest X-ray datasets for computer-aided screening of pulmonary diseases. , 2014, Quantitative imaging in medicine and surgery.

[10]  Michael H. Goldbaum,et al.  Labeled Optical Coherence Tomography (OCT) and Chest X-Ray Images for Classification , 2018 .

[11]  Yifan Yu,et al.  CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison , 2019, AAAI.

[12]  B. Hamm,et al.  Comparing different deep learning architectures for classification of chest radiographs , 2020, Scientific Reports.

[13]  Alexander Wong,et al.  COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images , 2020, Scientific reports.

[14]  Jose Ruiz-Pinales,et al.  Transfer Learning for Stenosis Detection in X-ray Coronary Angiography , 2020, Mathematics.

[15]  Jan Sölter,et al.  On the Composition and Limitations of Publicly Available COVID-19 X-Ray Imaging Datasets , 2020, ArXiv.

[16]  Juan Eduardo Luján-García,et al.  Fast COVID-19 and Pneumonia Classification Using Chest X-ray Images , 2020, Mathematics.

[17]  Wenjie Yang,et al.  Clinical characteristics and imaging manifestations of the 2019 novel coronavirus disease (COVID-19):A multi-center study in Wenzhou city, Zhejiang, China , 2020, Journal of Infection.