- Charles Sturt computer vision researchers are working to devise effective ways to counter the COVID-19 challenge and serve the global community
- Their studies provide researchers with a time-saving preliminary review of available literature on computer vision efforts countering the COVID-19 pandemic
- Experimental results will provide a solid pathway for researchers and practitioners to develop improved models in the future
During the COVID-19 pandemic much political and media focus has been on front-line health and support services.
But behind the scenes computer scientists and mathematicians at Charles Sturt University have been working to expeditiously solve complex problems in healthcare.
Professor in Computer Science Manoranjan Paul (pictured) in the Charles Sturt School of Computing and Mathematics and his colleagues have recently posted the results of research online to serve the global community and contribute to the fight to control COVID-19.
Professor Paul said their two research papers, ‘Computer vision for COVID-19 control: A survey’ and
‘X-ray image based COVID-19 detection using pre-trained deep learning models’, represent the highly technical life-saving expertise that operates behind the scenes in modern healthcare systems.
“Computer Vision, as a subfield of Artificial Intelligence (AI), has enjoyed recent success in solving various complex problems in healthcare, and can help to fight COVID-19,” Professor Paul said.
“The COVID-19 pandemic has triggered an urgent need to contribute to the fight against this immense threat to the human population.
“In response to this call, computer vision researchers are putting their knowledge to work to devise effective ways to counter the COVID-19 challenge and serve the global community.
“New contributions are being shared with every passing day, and it motivated us to review the recent work, collect information about available research resources, and provide an indication of future research directions.
“Our study ‘Computer vision for COVID-19 control: A survey’ is intended to provide a preliminary review of the available literature on the computer vision efforts against the COVID-19 pandemic, and we want to make it available to computer vision researchers to save precious time.”
Professor Paul said the second study, ‘X-ray image based COVID-19 detection using pre-trained deep learning models’, offered the promise that early detection may help in devising appropriate treatment plans and disease containment decisions.
“In this study, we demonstrate how pre-trained computer deep learning models can be adopted to perform COVID-19 detection using X-ray images,” he said.
“The aim is to provide over-stressed medical professionals with a second pair of eyes through intelligent image classification models.
“We highlight the challenges, including dataset size and quality, in utilising current publicly available COVID-19 datasets for developing useful deep learning models.
“We propose a semi-automated image pre-processing model to create a trustworthy image dataset for developing and testing deep learning models.”
Professor Paul explained the new approach is aimed to reduce unwanted 'noise' from X-ray images so that deep learning models can focus on detecting diseases with specific features.
“Next, we devise a deep learning experimental framework, where we utilise the processed dataset to perform comparative testing for several popular and widely available deep learning model families, such as VGG, Inception, Xception, and Resnet.
“The experimental results highlight the suitability of these models for current available datasets and indicate that models with simpler networks such as VGG19 perform relatively better with up to 83 per cent precision.
“This will provide a solid pathway for researchers and practitioners to develop improved models in the future.”These studies have underpinned related current Charles Sturt research.