Researchers at the University of Technology Sydney (UTS) have made a breakthrough in using artificial intelligence (AI) to rapidly detect COVID-19 infection through the analysis of chest X-rays. Their deep learning-based algorithm, known as the Custom Convolutional Neural Network (Custom-CNN), achieved an impressive accuracy rate of over 98%.
Real-time reverse transcription-polymerase chain reaction (RT-PCR) testing, the current gold standard for diagnosing COVID-19, has limitations such as high costs, slow results, and false negatives. As a complement to RT-PCR, chest X-rays help in the timely detection and management of contagious infections, particularly when RT-PCR returns a negative result.
COVID-19 displays specific radiological characteristics in chest X-rays that radiologists rely on for diagnosis. However, this manual examination of X-rays is time-consuming and prone to errors. To address these challenges, the UTS researchers turned to AI to streamline the diagnostic process.
Amir Gandomi, corresponding author of the study, explained, “The most widely used COVID-19 test, real-time polymerase chain reaction (PCR), can be slow and costly and produces false negatives. To confirm a diagnosis, radiologists need to manually examine CT scans or X-rays, which can be time-consuming and prone to error.”
The researchers utilized two datasets of chest X-ray images, with categories including normal X-rays, coronavirus-positive X-rays, and X-rays of viral pneumonia. The Custom-CNN model, designed specifically for diagnosing COVID-19, was trained on 80% of the images and tested on the remaining 20%. The results showed an impressive classification accuracy of 98.19% in differentiating between COVID-19, normal, and pneumonia X-ray samples. In fact, the Custom-CNN outperformed other existing models in the evaluation.
Gandomi emphasized the advantages of the AI system, stating, “Deep learning offers an end-to-end solution, eliminating the need to manually search for biomarkers. The Custom-CNN model streamlines the detection process, providing a faster and more accurate diagnosis of COVID-19.”
Early detection of COVID-19 through chest X-rays enables prompt treatment, including antiviral medication, which has the best efficacy within the first five days of symptom onset. Additionally, early detection can encourage patients to isolate and prevent further transmission.
Gandomi also highlighted the potential global impact of this AI system, stating, “The new AI system could be particularly beneficial in countries experiencing high levels of COVID-19 where there is a shortage of radiologists. Chest X-rays are portable, widely available, and provide lower exposure to ionizing radiation than CT scans.”
This groundbreaking research, published in Scientific Reports, not only revolutionizes COVID-19 diagnosis but also showcases the potential of AI in transforming the field of medicine by providing faster and more accurate diagnostic tools.
The source of the article is from the blog windowsvistamagazine.es