For years, young people have used smoking as a way to look older. As it turns out – thanks to a first-of-its-kind study out of the University of Lethbridge using artificial intelligence (AI) to analyze blood biochemistry – it’s true, smoking truly does make you older. “We demonstrate for the first time that smoking status can be predicted using blood biochemistry and cell count results and the recent advances in AI and machine learning,” says Dr. Olga Kovalchuk, a professor in the Department of Biological Sciences. “By employing age-prediction models developed using supervised deep learning techniques, we found that smokers exhibited higher aging rates than non-smokers. In other words, we show that smoking makes people biologically older.”
This realization almost sounds like common sense but until now, through the use of AI, it has never been quantified and illuminated to this extent. “We all have a chronological age but then there is also our biological age, which is an indicator of general fitness,” says Kovalchuk. “If somebody is 35 but on a biological clock, through specific markers, it shows them at a biological age of 50, obviously they are doing something wrong. Smoking, specifically in younger people, those in their 20s, 30s and 40s, is truly harmful as it makes them biologically older.”
The study involved a team effort by top clinicians, AI researchers, and deep learning and aging experts led by the Canada Cancer and Aging Research Laboratories in collaboration with InSilico Medicine, the world leaders in AI and aging research, as well as several national and international institutions. Kovalchuk has a personal connection to the effects of smoking on the human body. She also understands how the warnings linking smoking to lung cancer and heart disease often ring hollow with the millennial generation. “Fighting smoking is kind of up close and personal for me because my dad was a smoker and even though he quit in 2003, it still caught up with him,” says Kovalchuk of her father’s passing. “People are somewhat tired of hearing about lung cancer and heart disease in the context of smoking prevention but this is a different story. Our study shows that smoking is definitely associated with aging and it also shows that some effects are more pronounced in females. Maybe that’s a message that could really resonate.”
Kovalchuk’s research group used data from 149,000 anonymous individual blood biochemistry records linked to smoking status from across the province. Through the use of AI, they were able to look at blood biochemistry markers and predict age. What they found was both remarkable and troubling. Age predictions showed that the biological age of male smokers was 1.5 times older than their chronological age while female smokers were nearly twice as old as their actual chronological age.
“What’s beautiful about AI is that we couldn’t run these calculations before because the human mind just can’t deal with these large data sets. What it looks like is a bunch of numbers, lines of numbers, and we train it what to do and then it looks for patterns,” says Kovalchuk. “We wanted to do this using nothing fancy, just general basic bloodwork that is done on every general checkup. But with this data, using AI, you can see major patterns and it’s just fascinating.”
How people will consume this new information is unknown. For Kovalchuk, it’s another weapon in an age-old war against smoking. If it’s appealing to young people’s vanity, then so be it. “Once you develop cancer, it doesn’t really matter how it developed, now you have to treat it, so shaming or blaming a person for smoking is not productive,” she says. “But if we can prevent the cancers from happening with a message that will resonate, specifically with millennials and what are they concerned about, their looks, then let’s use this information.”
Source: University of Lethbridge