Published on Lancet Digital Health a study coordinated by Humanitas: the diagnosis of lesions increases by 44%. With fewer diagnostic errors and greater sustainability
Let’s face it: artificial intelligence (Ia) he can see things that we humans can’t even imagine. Is exactly in diagnostic imaginghis formidable analytical ability finds one of the
most promising fields of application. An example is colonoscopy, the endoscopic examination routinely used in screening for the prevention of colon and rectal cancers. Colonoscopy is done by passing an endoscope, which is a thin, flexible tube with a camera at the end, through the rectum and along the entire length of the colon. Through this tool, the experienced doctor can view any signs of cancer or precancerous lesions. In Italy, colon cancer ranks second in the ranking of the deadliest cancers and causes about 19,000 deaths a year and, to date, this test is the best tool for its correct diagnosis.
Greater diagnostic capacity
Well established scientific studies have shown that
the use of endoscopy and artificial intelligence together increases the ability to diagnose the presence of polyps by 10-13%
, even very small in size otherwise difficult to notice, either because they are not well exposed or because of their color often very similar to that of the healthy mucosa. But the benefits also translate into terms of greater economic sustainability: the data from a multicenter study published in The Lancet Digital Healthcoordinated by Alessandro Repici with Cesare Hassanboth professors of Humanitas University, in collaboration with the University of Oslo.
What escapes the human eye
The device, applied to the colonoscopy, helps doctors identify changes in the colon that may escape the attention of the human eye and therefore minimizes the risk of a missed diagnosis, explains Professor Alessandro Repici, director of the Department of Gastroenterology at Humanitas and professor at Humanitas University. International studies confirm that about 20% of injuries can escape diagnosis.
The cost-benefit ratio
The scientific work just published in The Lancet Digital Health investigates, for the first time, the cost-effectiveness of artificial intelligence technologies in the field of colonoscopy. The use of these software, in fact, requires very high economic investments from hospitals: the exam with Ia requires about 19 dollars (18 euros) more. per patient compared to routine. Scientific work has shown that, in 30 years, these investments are not only fully amortizedi but also generate savings when compared to medical bills for colon cancer patients.
Clinical advantages and economic savings
In fact, AI tools improve the identification rate of colon lesions by 44%, reducing the risk of diagnostic errors. The execution of screening colonoscopies with the aid of AI associated, on the one hand, with a lower incidence of colon cancers equal to 8.4% and on the other to a economic savings of 57 dollars (54 euros) per person thanks to the reduction in therapy costs linked to greater prevention. The PNRR funds, all those dedicated to digital innovation, should provide for the introduction of AI in medicine – Repici concludes -. As a researcher and clinician, my expectation that there is a massive implementation of physician support systems to improve diagnostics.
May 9, 2022 (change May 9, 2022 | 10:28)
© REPRODUCTION RESERVED
#colon #cancers #costs #endoscopy #performed #artificial #intelligence