Explainable COVID-19 Detection Using Chest CT Scans and Deep Learning

• Main Authors: Hammam Alshazly, Christoph Linse, Erhardt Barth, Thomas Martinetz
• Format: Article
• Language: English
• Published: MDPI AG 2021-01-01
• Series: Sensors
• Subjects: coronavirus; COVID-19 detection; SARS-CoV-2; explainable deep learning; feature visualization
• Online Access: https://www.mdpi.com/1424-8220/21/2/455

This paper explores how well deep learning models trained on chest CT images can diagnose COVID-19 infected people in a fast and automated process. To this end, we adopted advanced deep network architectures and proposed a transfer learning strategy using custom-sized input tailored for each deep architecture to achieve the best performance. We conducted extensive sets of experiments on two CT image datasets, namely, the SARS-CoV-2 CT-scan and the COVID19-CT. The results show superior performances for our models compared with previous studies. Our best models achieved average accuracy, precision, sensitivity, specificity, and F1-score values of <inline-formula><math display="inline"><semantics><mrow><mn>99.4</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>99.6</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>99.8</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>99.6</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and <inline-formula><math display="inline"><semantics><mrow><mn>99.4</mn><mo>%</mo></mrow></semantics></math></inline-formula> on the SARS-CoV-2 dataset, and <inline-formula><math display="inline"><semantics><mrow><mn>92.9</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>91.3</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>93.7</mn><mo>%</mo></mrow></semantics></math></inline-formula>, <inline-formula><math display="inline"><semantics><mrow><mn>92.2</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and <inline-formula><math display="inline"><semantics><mrow><mn>92.5</mn><mo>%</mo></mrow></semantics></math></inline-formula> on the COVID19-CT dataset, respectively. For better interpretability of the results, we applied visualization techniques to provide visual explanations for the models’ predictions. Feature visualizations of the learned features show well-separated clusters representing CT images of COVID-19 and non-COVID-19 cases. Moreover, the visualizations indicate that our models are not only capable of identifying COVID-19 cases but also provide accurate localization of the COVID-19-associated regions, as indicated by well-trained radiologists.