AI can identify bowel cancer ‘in less than a second’
Project leader says system could cut deaths from the disease
Researchers have developed an artificial intelligence (AI) programme that can detect signs of bowel cancer in “less than a second”.
The system - developed by researchers at Showa University in Yokohama, Japan - detects potentially cancerous tumours by taking pictures during a colonoscopy and comparing them with 30,000 reference images almost instantaneously.
In recent tests, says Alphr, the system successfully identified 90% of colorectal adenomas - benign tumours that can evolve into bowel cancer - from a sample of 250 patients.
Presenting the AI system at the United European Gastroenterology event in Barcelona, project leader is that it can assess patients in real time, removing the need to examine results after a colonoscopy has been performed.
“We believe these results are acceptable for clinical application and our immediate goal is to obtain regulatory approval for the diagnostic system”, he adds.
The immediacy of the AI-powered system could help “decrease the risk of colorectal cancer and, ultimately, cancer-related death”, Dr Mori said.
It’s not just cancer treatment that can benefit from AI. Several universities have developed systems that can detect diseases before medical symptoms appear.
A team of researchers from the University of Bari, in Italy, has created an AI programme that is able to detect early signs of Alzheimer’s by comparing a patient’s MRI scans with a range of reference images. This can help doctors identify the disease up to a decade before visible symptoms show.
A similar programme that can detect the disease through an eye test has been created by scientists from several universities across California, says Digital Trends.