Artificial intelligence-based solutions have achieved significant successes in the field of health in recent years. These solutions have been started to be used for pre-diagnosis and decision support for a virus that spreads rapidly such as COVID-19 and thus creates fear and panic among the public. These solutions have augmented clinical expertise and thus have great potential to mitigate the virus outbreak burden of health experts. In this context, the load of healthcare workers can be significantly reduced through the help of an automatic diagnosis system of a high number of patients who apply to healthcare organizations with suspicion of disease. In this study, a machine-learning automatic diagnosis system exploiting x-ray images is proposed to detect diseases caused by COVID-19. The proposed system employs powerful texture features (Histogram of Oriented Gradients, Law’s Texture Energy Measure, Gabor Wavelet Transform, Gray Level Co-Occurrence Matrix, and local binary pattern) for the x-ray images to training a randomized neural network, a fast network, to establish a robust and fast diagnosis process for the virus. This study has raised the thesis that the mentioned image texture features extracted from the virus patients' images contain determinative indicators in two-dimensional space that make it possible to diagnose the disease. The proposed system contributes to the literature by using the tissue properties of x-ray images for the diagnosis of the virus. The disease is detected with an accuracy of 100 utilizing Law’s Texture Energy Measure feature and randomized neural network approach.
Primary Language | English |
---|---|
Subjects | Electrical Engineering |
Journal Section | Research Article |
Authors | |
Publication Date | December 30, 2021 |
Published in Issue | Year 2021 |
All articles published by EJT are licensed under the Creative Commons Attribution 4.0 International License. This permits anyone to copy, redistribute, remix, transmit and adapt the work provided the original work and source is appropriately cited.