Research Shows Artificial Intelligence More Accurate Than Doctors In Diagnosing Disease

A mum who underwent a double mastectomy after doctors wrongly diagnosed her with breast cancer has welcomed a study’s findings which show artificial intelligence is more accurate than doctors in diagnosing the disease.

Sarah Boyle was 25 when doctors at Royal Stoke Hospital told her she had triple negative breast cancer.

She underwent extensive treatment, which included chemotherapy, a bilateral mastectomy and reconstructive surgery. However, several months later she was informed that that her biopsy had been incorrectly reported and it was confirmed that she did not have cancer.

Following the news, Sarah instructed specialist medical negligence lawyers at Irwin Mitchell to investigate the case.  The legal experts secured an admission of liability from University Hospitals of North Midlands NHS Trust, which runs Royal Stoke, regarding the failings. The Trust stated that the misdiagnosis was down to ‘human error’

Sarah has joined her legal team at Irwin Mitchell in welcoming the publication of research in the medical journal Nature that suggests artificial intelligence (AI) is more accurate than doctors in diagnosing breast cancer from mammograms.

Expert Opinion

"What Sarah and her family have had to endure is truly shocking and the effect of what happened continues to impact on their lives.

“Sarah has suffered significant psychological trauma as a result of what she has been through, and also continues to endure ongoing symptoms caused by her treatment.

“The use of technology should not replace human input but technological advances that can complement and assist medical professionals to improve care and lessen waiting times and anxiety for patients should be welcomed. While this research is in its infancy the results of this study appear to be very promising.

“We are continuing to support Sarah to help her try and come to terms with what happened to her the best she can.” Sarah Sharples - Senior Associate Solicitor

Find out more about Irwin Mitchell's expertise in handling medical negligence cases 

Sarah, now 28, lives in Stoke-on-Trent with sons Teddy, Louis and her husband of five years, Steven, 31.

She was diagnosed with breast cancer at the end of 2016 with doctors saying how the disease was uncommon for a woman of her age.

Following extensive treatment she was told in June 2017 that her cancer had been misdiagnosed.

Researchers, including from Imperial College London and Google Health designed and trained a computer model on X-ray images from around 29,000 women whose details had been anonymised.

The computer programme outperformed six radiologists in interpreting mammogram results, according to the BBC.

AI was as good as two doctors working together. The results also showed the algorithm was better at correctly reading mammograms compared to one doctor working alone.

The NHS currently uses a system where two radiologists analyse radiology. If they disagree on the findings a third doctor assesses the images.

Sarah Boyle said: “Even now it is so difficult to try and describe what has happened to me. To be told you have cancer and it’s uncommon for someone your age was hard enough to take in. But then to be told after months of horrific treatment that it was all unnecessary is something I’m not sure I’ll ever fully come to terms with.

“It’s not just the physical effects that I have been left with but also the mental torture of what I’ve been through.

“A misdiagnosis of cancer can ruin people’s lives and some people may not be as fortunate to survive.

“It is vital to raise awareness of the consequences that families can be left to face because of errors. Anything that helps reduce the number of people affected by a misdiagnosis or allows others to receive treatment more quickly has to be welcomed.”