In the land down under, deep-learning algorithms have analyzed cross-sectional chest CT scans from 48 patients and predicted, with impressive accuracy, who among them would die within five years.
The AI tool drew from routinely acquired CT images and used off-the-shelf machine-learning methods. It was correct 69 percent of the time—about the same as that of clinicians using manual methods—and was most confident when basing its predictions on images showing severe chronic diseases like emphysema and congestive heart failure.
The international team behind the research was led by a radiologist at the University of Adelaide in South Australia. Their open-access study is running in Scientific Reports.
Meanwhile the university’s press materials have been picked up by Insurance Journal, whose interest in the work may be interesting in its own right.
“Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do,” says lead author Luke Oakden-Rayner, MD.
Read the rest:
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.