An AI tool has succeeded in identifying signs of breast cancer in 11 women which were missed by human doctors at an NHS hospital trust, according to reports. The tool, Mia, was piloted alongside NHS clinicians studying mammograms of more than 10,000 women. Despite most being cancer-free, Mia successfully flagged up all those showing symptoms of the disease, as well as those that were missed by clinicians. Of the cases studied in the trial, none were analysed by Mia alone, with two radiologists studying every individual scan.
Barbara was one of the 11 patients whose cancer was flagged by Mia but not detected by hospital radiologists. As her 6mm tumour was caught so early, she only required an operation and five days of radiotherapy. She stated that she was pleased the treatment was less invasive than that received by her mother and sister, who previously suffered from the disease. Because of AI tools’ instantaneous nature, they can potentially reduce wait times for results from 14 days down to three, according to Kheiron, the company behind Mia.
Dr. Gerald Lip, clinical director of breast screening in the North West of Scotland and the project’s leader, suggested that in the future, AI could reduce the fatigue felt by doctors who regularly view up to 5,000 scans per year by halving the workload for each pair. Breast cancer doctors can view up to 100 in a single session. However, Mia does not yet have access to patient history, which means that some cysts were flagged despite being identified by previous scans and deemed harmless.
The Mia trial represents an early test of one product in one location. Although the University of Aberdeen has independently validated the results, they have yet to be peer reviewed. Dr Katharine Halliday, President of the Royal College of Radiologists, commented that the results were “encouraging” and demonstrated the “exciting potential AI presents for diagnostics”. Dr Julie Sharp, head of health information at Cancer Research UK, said that innovation would be “vital” to help reduce pressure on NHS services and staff as the number of cancer cases declines
Read the full article from The BBC here: Read More