Scientists from Charles Darwin University in Australia developed the new AI. They trained the algorithm using lung ultrasound video data and concluded that the AI can accurately detect problems and even differentiate one respiratory disease from another.
AI identifies specific patterns indicative of various lung diseases. Tests have shown diagnostic accuracy to date to be 97%: the AI was able to successfully differentiate between pneumonia, COVID-19, and other illnesses. Unlike humans, AI can detect subtle changes in the pixels of data, which, taken together, have a significant impact on the outcome.
It is especially important that, in addition to the results, the scientists trained the AI to provide explanations to doctors in order to increase the trust of medical professionals.
In addition to the diagnosis, the AI provides doctors with a report with specific examples in the images from which the corresponding conclusions were made.
The team is currently working to improve the accuracy of the algorithm to make it universally applicable to the diagnosis of tuberculosis, cancer, asthma, fibrosis, and many other lung diseases. This diagnostic could potentially be used for population screening and when specific diseases are suspected.
One of the common symptoms following COVID-19 is brain fog. Scientists recently discovered that this condition is caused by impaired lung function. Currently, such lung changes are diagnosed using MRI, but it's possible that, as new AI learns, it could also help with this issue during the rehabilitation phase.
