2 weeks Ago By Yahoo Life UK
A woman has revealed how artificial intelligence (AI) successfully detected her cancer after a routine mammogram came back as "normal". Sheila Tooth, 68, from Littlehampton, West Sussex, was initially given the all-clear after her most recent mammogram. As is normal procedure, two experienced radiologists reviewed the scan and determined that there were no signs of cancer.
However, University Hospitals Sussex, the trust where Sheila had her appointment, was at the time partaking in a scheme which used an AI system to further analyse mammograms - to see if it could improve early detection. The technology spotted cancerous cells which were undetectable by the human eye, meaning Sheila could get treatment as early as possible. The retired nurse, who 15 years ago was diagnosed with a non-invasive early breast cancer, said: "I remember the shock of the letter.
Having had cancer before I was very frightened. But I knew that whatever they could see on my scan must have been incredibly small if it wasn’t picked up the first time." Sheila, a mother-of-one, was diagnosed with the same early non-invasive breast cancer as before, but because it was detected so early, she was able to have a lumpectomy and didn’t need any further treatment.
Now, recovering from the surgery, she describes being so "grateful" for the AI technology. "It’s extraordinary and I’m amazed," Sheila said. "When I talk to friends, we just can’t believe this AI can detect what the human eye can’t always see.
I just feel so lucky. Being 68, this may have been my last mammogram, so my early cancer might have developed into invasive cancer in my 70s. So, I’m deeply grateful for it to have been caught so early.
" University Hospitals Sussex is one of 15 trusts across the country that participated in the project, to test if AI can spot cancers than human scan "readers" might miss. The project used an AI system developed by Kheiron Medical Technologies called Mammography Intelligent Assessment - having been funded by the NIHR and NHS England's Artificial Intelligence in Health and Care Award. Throughout the two-month project, more than 12,000 mammograms considered as "normal" by radiologists were reviewed using the AI system.
The technology suggested that just under 10% of those mammograms should be re-read by a clinical panel to identify any potential cancers that were not detected in the initial screening. Upon further review, 11 women were asked to come back for investigation. Five of these were found to have breast cancer.
Commenting on the scheme, Dr Olga Strukowska, a consultant radiologist and director at the West Sussex Breast Screening Programme, says: "We are still in early stages of AI evaluation in clinical scenarios but based on current trials and research projects, AI should find its place within the breast screening programme. "The earlier and more accurately we detect cancer, the better the chance our patients will have a positive outcome," Dr Strukowska continues. "That’s why this is so exciting.
Using AI increases accuracy while reducing the number of missed cancers and lowering false positives. It empowers screening services to deliver confident, accurate, timely results through deep learning technology that works with radiologists and promotes high-quality standards of care for our patients." Steve Dixon, Senior AI Project Lead for Breast Services, adds: "I feel privileged to have been part of the UHSussex breast screening teams in this groundbreaking evaluation project to demonstrate one of the potential uses of AI within the NHS breast screening programme.
I have no doubt that, in time, integrating AI with clinicians’ expertise will enhance the effectiveness of patient care, improving both outcomes and the quality of service for patients." University Hospitals Sussex is now planning to take part in a national randomised controlled trial involving AI, which forms part of the next phase of introducing the technology to breast cancer screening. Last year, prospective evidence, , found that an AI tool, called Mia, could significantly increase the early detection of breast cancers in a European healthcare setting by up to 13%.
Currently, 55,000 women in the UK are diagnosed with breast cancer every year. The NHS invites women between the ages of 50 and 71 to routine mammogram appointments every three years. These screening measures to detect cancer early do reduce the amount of intensive treatments and deaths from breast cancer.
However, despite these initiatives, it is estimated that about 20% of breast cancers might be missed at this stage. A published in The Lancet Oncology describes how researchers used AI to help screen mammograms of more than 80,000 women in Sweden. Half of these women had their mammogram read by AI before it was looked at by a radiologist, while the other half had theirs read by two radiologists.
The study revealed that the AI group had 20% more cancers detected than the radiologist-only group. Another study in Germany and the US, published in the , that used AI to look at nearly 1.2 million mammograms found that having a radiologist and AI system working together was 2.
6% better at detecting breast cancer than a radiologist alone. AI technology could also help reduce false positive breast cancer diagnoses. According to a false positive result occurs when the radiologist detects an abnormal finding on a mammogram that doesn't ultimately prove to be a cancer.
A study, published in , of more than 91,000 mammograms from women in the US and the UK found that the use of an AI system lowered the rate of false positives by 5.7% in the US and by 1.2% in the UK.
AI may also lead to improvements in doctors’ ability to predict those people at greatest risk of developing breast cancer. A study published in June last year in found that AI was more accurate in predicting breast cancer risk than the Breast Cancer Surveillance Consortium (BCSC) risk model. (Yahoo Life UK, 3-min read) (Yahoo Life UK, 6-min read) (Yahoo Life UK, 5-min read).
