Artificial intelligence could be adopted by the NHS to spot ovarian cancer after new research suggested the software was more effective than doctors at catching the hard-to-find deadly disease.
The technology, known also as AI, can correctly spot tiny tumours, called lesions, on ultrasound images of the ovaries, nearly nine out of ten times, according to Swedish researchers.
Ovarian cancer specialists, on the other hand, were found to only spot the lesions in eight out of ten cases.
Experts say the findings are significant because ovarian cancer is considered one of the most difficult to diagnose form of tumours.
This is because the symptoms – which include bloating, frequent urination, vaginal discharge and constipation – are often mistaken for signs of less severe illnesses.
There is also currently no effective way to screen women for the disease. This means that, by the time it is spotted, the cancer has often spread around the body.
Research suggests as many as four in five cases are found only after the cancer has spread into other parts of the body.
Some 7,500 women in the UK are diagnosed with ovarian cancer every year – and it kills around 4,000 over the same time period.
Artificial intelligence could be adopted by the NHS to spot ovarian cancer after new research suggested the software was more effective than doctors at catching the hard-to-find deadly disease (Stock image)
The technology, known also as AI, can correctly spot tiny tumours on ultrasound images of the ovaries, nearly nine out of ten times (Stock image)
Research has already shown that artifical intelligence can speed up diagnosis of patients with skin and lung cancer.
Last year, the NHS announced a first-of-its-kind breast screening trial which will use AI to look for signs of cancer on mammograms, in an effort to improve the accuracy and speed of diagnosis.
The latest study, published by scientists at the Stockholm South General Hospital, uploaded more than 17,000 ultrasound images of ovaries to a self-learning AI computer programme – also known as a neural network model.
These images included some of patients with cancerous lesions and others with growths which were non cancerous – known as benign lesions.
After analysing all the images, the AI was able to correctly identify the signs of ovarian cancer in the vast majority of cases.
The researchers concluded that, due to its speed and accuracy, using the AI in hospitals could speed up the number of referrals doctors can see on a daily basis by around 60 per cent, and cut misdiagnoses by nearly a fifth.
‘Ovarian tumours are common and are often detected by chance,’ said Professor Elisabeth Epstein, senior obstetrics and gynecology consultant at Stockholm South General Hospital.
‘This suggests that neural network models can offer valuable support in the diagnosis of ovarian cancer, especially in difficult-to-diagnose cases and in settings where there’s a shortage of ultrasound experts.’