A US study suggests that Artificial Intelligence (AI) is better than specialist doctors at identifying lung cancer. It’s a finding that could revolutionize cancer screening in the future, potentially allowing tumors to be found at an earlier stage and improving treatment outcomes.
According to the study - which was conducted by researchers from Northwestern University in Illinois and the Google Health Research Group – Artificial Intelligence was able to outperform six specialist cancer doctors when it came to identifying cancer from a single CT scan. When multiple CT scans were used, the AI and the doctors were equally effective.
Prior to the tests, the AI was trained with 42,290 CT lung scans from nearly 15,000 patients. It was not told what to look for in a CT scan, merely which patients went on to develop cancer and which didn’t.
The results of the study, published in Nature Medicine, show that AI can not only boost cancer detection by 5%, but can also reduce false-positives by 11%.
Speaking about the findings of the research, Dr Mozziyar Etemadi, from Northwestern University, said: “Not only can we better diagnose someone with cancer, we can also say if someone doesn't have cancer, potentially saving them from an invasive, costly, and risky lung biopsy.”