Researchers at the University of Adelaide have developed an artificial intelligence that can predict which patients will die within the next five years.
The AI developed by the researchers at the University of Adelaide was trained to detect signs of diseases in the chest area such as emphysema and congestive heart failure based on over 16,000 CT chest scans.
In the paper entitled “Precision Radiology: Predicting longevity using feature engineering and deep learning methods in a radiomics framework” published in the journal Nature, the University of Adelaide researchers showed that the AI they have developed, after analyzing CT chest scans of 48 patients, was able to predict which patients would die within five years, with 69% accuracy.
“Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual,” lead author of the paper Dr. Luke Oakden-Rayner, a radiologist and Ph.D. student at the University of Adelaide’s School of Public Health, said in a statement. “Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do, by incorporating large volumes of data and detecting subtle patterns.”
Authors of the paper wrote that routine radiological testing such as CT scans has been largely ignored in the context of precision medicine – an approach for disease treatment and prevention that takes into account differences in genes, environment, and lifestyle.
According to the University of Adelaide researchers, CT scans are the “most highly utilised cross-sectional medical imaging” with an estimated 69 million CT scans performed in the U.S. in 2007 and 2.1 million CT scans performed in Australia in 2009.