Research using machine learning on images of everyday items is improving the accuracy and speed of detecting respiratory diseases, reports Tech Xplore. Researchers from Edith Cowan University in Australia trained algorithms on a database of more than 1 million commonplace images and transferred this knowledge to identify characteristics of medical conditions that can be diagnosed with an X-ray. Results of this technique, known as transfer learning, achieved a 99.24 percent success rate when detecting COVID-19 in chest X-rays. “Our technique has the capacity to not only detect COVID-19 in chest X-rays, but also other chest diseases such as pneumonia,” said researcher Dr. Shams Islam. “We have tested it on 10 different chest diseases, achieving highly accurate results. Normally, it is difficult for AI-based methods to perform detection of chest diseases accurately because the AI models need a very large amount of training data to understand the characteristic signatures of the diseases…. Our method bypasses this requirement and learns accurate models with a very limited amount of annotated data.”
https://techxplore.com/news/2021-06-photos-toasters-fridges-algorithms-covid-.html
