By Muizat Hameed
A recent study published in Journal Nature has suggested that Artificial Intelligence is more accurate than medical doctors in diagnosing breast cancer.
A trained computer model was formed from X-ray images of almost 29,000 women.
The resultant algorithm which was designed by an international team of researchers from Imperial College London and Google Health outperformed six radiologists in reading mammograms.
The researchers while stating the advantages of the machine over doctors said, “AI was still as good as two doctors working together. Unlike humans, AI is tireless, thus can improve detection.”
“Our team is really proud of these research findings, which suggests that we are on our way to developing a tool that can help clinicians spot breast cancer with greater accuracy,” said Dominic King, one of the researchers from Google Health.
He further stated that it takes over a decade of training as a doctor and specialist to become a radiologist who will be capable of interpreting mammograms.
He added that reading X-rays is vital but time consuming, and there is an estimated shortage of more than 1000 radiologist across the United Kingdom.
Sarah Hiom, the Director of Cancer Intelligence and Early Diagnosis at Cancer Research UK (CRUK) said, “This is promising early research which suggests that in future it may be possible to make screening more accurate and efficient , which means less waiting and worrying for patients and better outcomes.”
A patient representative on the CRUK panel, who was diagnosed with breast cancer at 44, said she was skeptical about what Google Health will do with her data, but found out eventually that the research was stripped of all identifiers.
Helen Edward who required surgery, chemotherapy and radiotherapy, but has been free from cancer for more than a decade said, “Artificial intelligent machines don’t get tired. They can work 24/7 whereas a human can’t do that, so to combine the two is a great idea.”
She concluded that the research will benefit women mostly, in the long run.”
The study however stated that the knowledge and expertise of human can’t be taken over by the machine model, as both will be needed in designing and training other artificial intelligence models.
