A recent study published on nature.com entitled "Precision Radiology: Predicting longevity using feature engineering and deep learning methods in a radiomics framework" is intriguing the internet and everyone who is interested in figuring out when they might kick the bucket. But what does it all mean?
The main takeaway from the study is this: Using CT scans and the data gathered from those scans can more accurately predict life longevity, as opposed to using risk factors and environmental influences alone, which are the typical ways scientists and doctors use to predict how long you will live currently.
Genetic information alone isn't enough, so these scientists are interested in using AI to examine the current body (and all that ails it) in order to determine the effects of things like environmental and risk factors. That is, they're looking for disease. Then the AI collects data from thousands of scans and can put together an average life expectancy for, well, anyone.
While the study does have its drawbacks (it's invasive, for one), it's a great proof of concept with the goal of figuring out how healthy people. If you're interested in the science, definitely check out the study to explore all of the scientific details.