latest
In a study of GP practices in the east of England, using AI software to analyse patient data led to a higher cancer detection rate, while another study found that AI was able to speed up diagnosis of heart problems
“Our system has detected over 50 different types of cancers. The key thing is that it’s not only an earlier diagnosis, but a faster diagnosis.” Bea Bakshi, GP, and co-creator of C the Signs
GP practices have been using artificial intelligence (AI) software to scan patient records and find particular patterns to help them detect more cases of cancer.
The software, known as C the Signs, has enabled the rate of cancer detection to rise from 58.7% to 66% at those GP practices using it, researchers found. C the Signs works by analysing a patient’s medical record to pull together their past medical history, test results, prescriptions and treatments, as well as other personal characteristics that might indicate cancer risk, such as their postcode, age and family history.
The tool also prompts GPs to ask patients about any new symptoms. If it detects patterns in the data that indicate a higher risk of a particular type of cancer, then it recommends the tests or clinical pathway the patient should be referred to. Bea Bakshi, a GP who created the system with her colleague, Miles Payling, told the Guardian: “It could be a scan, an ultrasound, or they could need to be seen by a specialist at a clinic.”
C the Signs is used in about 1,400 GP practices in England – about 15% of the total. It was tested in 35 practices in the east of England between May 2021 and March 2022, covering a population of 420,000 patients. The results, which have been published in the Journal of Clinical Oncology, showed that while the cancer detection rate (CDR) improved from 58.7% in 2021-22 to 66% in 2022-23 in those practices, it remained at 58.4% in those practices that didn’t adopt the software.
Patients are tracked through the C the Signs system to remind doctors to check test results and referrals elsewhere. “Our system has detected over 50 different types of cancers,” Bakshi said. “The key thing is that it’s not only an earlier diagnosis, but a faster diagnosis.”
Bakshi and her colleagues validated the tool by assessing 118,677 patients in a previous study, which found that 7,295 were diagnosed with cancer and 7,056 were successfully identified by the algorithm.
In cases when the AI software concluded that it was unlikely a patient had cancer, only 239 out of 8,453 went on to have a confirmed cancer diagnosis within six months (about 2.8%). Bakshi developed the tool after meeting a patient in hospital who had been given a late diagnosis of pancreatic cancer and died three weeks later: “It stayed with me as a problem area. Why are patients with cancer being diagnosed so late?”
There are 200 different types of cancer. They are not always easy to diagnose, because they may be asymptomatic or have symptoms similar to other conditions.
“Two-thirds of deaths are in the non-screenable cancers and the ones that we aren’t screening for,” Bakshi said. “Patients visit GPs between three and five times before they are recognised as being at risk of cancer. GPs detect an average of eight cases of cancer a year.”
AI is also being used to analyse heart MRI scans to speed up diagnosis of heart conditions. Researchers from the Universities of East Anglia (UEA), Sheffield and Leeds created an AI tool that examines heart images from MRI scans to determine the size and function of the heart’s chambers. Whereas a standard analysis by a doctor of a heart scan takes about 45 minutes, the AI scan takes a few seconds, and produces comparable outcomes.
The study looked at data from 814 patients from Sheffield Teaching Hospitals NHS Foundation Trust and Leeds Teaching Hospitals NHS Trust, which was then used to train the AI model. To make sure the model’s results were accurate, scans and data from another 101 patients from the Norfolk and Norwich University Hospitals NHS Foundation Trust were then used for testing. The AI model provides a complete analysis of the entire heart using a view that shows all four chambers, while most earlier studies looked only at the heart’s two main chambers.
The lead researcher, Dr Pankaj Garg, of the University of East Anglia’s Norwich Medical School and a consultant cardiologist at the Norfolk and Norwich University Hospital, said: “This automated technique could offer speedy and dependable evaluations of heart health, with the potential to enhance patient care.”
FCC Insight
Both these studies show the value AI can bring to health care. Cancer can be notoriously difficult to detect, because many cancers are either asymptomatic, or their symptoms are easily confused with other diseases. The use of C the Signs to improve detection rates potentially helps to save the lives of patients who might otherwise be diagnosed too late. In the case of the AI tool used to analyse heart scans, the benefit arises from the much greater speed, rather than the accuracy, of diagnosis – but with similar potential to improve patient outcomes.