COVID-19 deep learning prediction model using publicly available radiologist-adjudicated chest X-ray images as training data: preliminary findings

• Main Authors: Che Azemin, Mohd Zulfaezal, Hassan, Radhiana, Mohd Tamrin, Mohd Izzuddin, Md. Ali, Mohd. Adli
• Format: Article
• Language: English; English; English
• Published: 2020
• Subjects: RC731 Specialties of Internal Medicine-Diseases of The Respiratory System; TK7885 Computer engineering
• Online Access: http://irep.iium.edu.my/82304/1/8828855.pdf; http://irep.iium.edu.my/82304/7/82304_COVID-19%20Deep%20Learning%20Prediction%20Model%20Using%20Publicly_Scopus.pdf; http://irep.iium.edu.my/82304/12/82272_Synthesis%20and%20characterization%20of%20bubble_wos.pdf; http://irep.iium.edu.my/82304/; https://www.hindawi.com/journals/ijbi/2020/8828855/; https://doi.org/10.1155/2020/8828855

The key component in deep learning research is the availability of training data sets. With a limited number of publicly available COVID-19 chest X-ray images, the generalization and robustness of deep learning models to detect COVID-19 cases developed based on these images are questionable. We aimed to use thousands of readily available chest radiograph images with clinical findings associated with COVID-19 as a training data set, mutually exclusive from the images with confirmed COVID-19 cases, which will be used as the testing data set. We used a deep learning model based on the ResNet-101 convolutional neural network architecture, which was pretrained to recognize objects from a million of images and then retrained to detect abnormality in chest X-ray images. The performance of the model in terms of area under the receiver operating curve, sensitivity, specificity, and accuracy was 0.82, 77.3%, 71.8%, and 71.9%, respectively. The strength of this study lies in the use of labels that have a strong clinical association with COVID-19 cases and the use of mutually exclusive publicly available data for training, validation, and testing.