The use of artificial intelligence detection tools has the potential to save the U.S. millions of dollars annually, according to a study published in The Lancet Digital Health.
The study administered the Markov model microsimulation of conducting colonoscopy with and without AI for colorectal cancer screening. The microsimulation screened for people at average risk for colorectal cancer and used a hypothetical cohort of 100,000 people in the U.S. ages 50-100.
The first analysis looked at colonoscopy screenings with and without AI every 10 years starting at age 50 and ending at 80, with follow-up until 100, assuming 60 percent screening population uptake. The second analysis modeled once-in-life colonoscopy screening at 65 in adults between 50-79 with an average risk for colorectal cancer.
The study estimated 3 percent annual discount rates for the costs of AI tools and the cost for downstream treatment of screening-detected disease.
The study found that AI detection tools decreased the discounted costs per screened person from $3,400 to $3,343, which saves each person $57. It also found that for the U.S. population, implementing AI detection during colonoscopy screening would result in annual additional prevention of 7,194 colorectal cancer cases and 2,089 related deaths, and yield annual savings of $290 million.