KUALA LUMPUR, May 23 — Researchers of a study at the Northwestern University in Illinois and Google in the US, are suggesting that artificial intelligence (AI) is better than specialist doctors at diagnosing lung cancer.
The study published in Nature Medicine, focused on lung cancer, which claims the lives of 1.8 million a year, more than any other cancer.
Heavy smokers are considered of high risk. Screening them earlier for tumours and catching them at an early stage would make existing treatment options more effective.
However, indiscriminate screening including invasive biopsies, may produce negative or even false results, resulting in tumours going undetected.
The study trained computer software utilising 42,290 CT lung scans from nearly 15,000 patients by indicating which patients went on to get cancer and which did not.
It was then tested against a team of six radiologists.
The AI was found to be more effective than the radiologists when examining both a single or multiple CT scans.
It demonstrated that software could boost cancer detection by 5 percent while cutting false-positives (saying that the patients had cancer when they did not) by 11 percent.
However, the current AI setup is not yet ready or available for clinical trials. This will be the next step.