Cancer is one of the biggest killers in the world, accounting for nearly 10 million deaths in 2020, according to the World Health Organisation.
Although many cancers can be cured if detected early and treated effectively, new research “reveals that less than half of 18-to-24-year-olds can identify any of the five main warning signs of cancer in young people” says the Teenage Cancer Trust.
“You have the best chance at survival and a healthy quality of life if your cancer is diagnosed and treated in its early stages,” explains Healthline.
That said, a lack of awareness and education about cancer symptoms could be a thing of the past, thanks to new AI development.
“Harvard Medical School did a collaboration with the University of Copenhagen where they trained these AI models on millions of people’s health records in Denmark” explains Dr Karan Raj on TikTok.
Researchers at Harvard Medical School and the University of Copenhagen were able to develop an AI with the potential to identify people at risk of pancreatic cancer up to three years before diagnosis.
Dr Karan Raj explains the AI could pick up on subtle changes and medical notes, and whether a patient developed health risks such as gallstones, anaemia, and Type-2 diabetes – all potential signs of developing pancreatic cancer.
Why is it important?
A paper published in Nature Medicine suggested that AI-based population screening could be useful for “identifying high-risk individuals and could help accelerate detection of malignancy often diagnosed at advanced stages.”
“Only 12% of patients are alive after five years, and it’s pretty much the same all over the world,” Søren Brunak, PhD, professor of disease systems biology and director of research at the University of Copenhagen says. “It’s also on the rise – not for very well-known reasons. So the unmet need in this domain is quite significant.”
How does it work?
“We looked at whether it would pick up many of the known risk factors because, when we have trained an algorithm, we can do this explainability exercise through which we try to look at the parameters in a neural network that trigger it toward coming up with a diagnosis of pancreatic cancer,” explains Brunak.
“During the training, the AI algorithm identified patterns suggestive of pancreatic cancer risk based on disease trajectories. It identified diagnoses such as gallstones, anaemia, type 2 diabetes and other gastrointestinal conditions as predictive of risk for pancreatic cancer.”
Earlier this year, an AI tool designed by experts at the Royal Marsden NHS Foundation Trust, the Institute of Cancer Research, London, and Imperial College London found they could identify whether abnormal growths found on CT scans were cancerous.
“This is an important step forward in being able to use AI to understand which patients are at highest risk of cancer recurrence,” says Dr Richard Lee, a consultant physician in respiratory medicine and early diagnosis at the Royal Marsden NHS Foundation Trust.
Most importantly, “detecting this relapse sooner so that re-treatment can be more effective,” he adds.