The Application of Artificial Intelligence Technique (CNN-Alexnet) in Diagnosing COVID-19 Using Chest X-ray Images
Background: The coronavirus which initially appeared in China in December 2019 was later declared global pandemic in the year 2020. It has caused a devastating effect on daily lives, public health, and the global economy. Early detection of positive cases is overly critical to prevent further spread of the pandemic and to quickly treat affected patients in isolation. Which is why introduction to fast and accurate alternative of diagnosing the virus is very vital. Methods: An AI technique called deep learning which is most applied to analyze visual imagery like radiological images, This AI technique uses convolutional neural networks (CNN) to analyze the images, AlexNet is the CNN model used for this research. Several studies suggest that medical images contain salient information about the Covid-19 virus, which is why applying such advanced artificial intelligence (AI) techniques coupled with radiological imaging can be helpful for the accurate detection of this disease with a huge potential to address the problem of a limited to no specialized physicians in remote areas like Nigeria’s most vulnerable regions. Results: Initially, the model gave high accuracy of 97.97%, this was suspected to be overfitting. This was corrected by increasing the dataset and applying cross validation thereby reducing noise by giving a lower accuracy to 85% and also increasing its specificity. Conclusions: The aim of the study was to introduce an alternative way of diagnosing the Covid-19 asides from the PCR that is currently the most popular one, this has been archived by our working system and the waiting time has been reduced from 24-48hours to 58 minutes. Secondly, to identify a suitable model in Deep learning in medical science and to measure the performance and to access the effectiveness of the chosen model Alexnet in terms of accuracy, precision, recall &F1score. We archived this by striking a balance in the high percentile number of the following terms and reducing it to a more believable, reliable, and accurate figure.