AI is nearly twice as effective at grading the aggressiveness of some cancers compared to a biopsy, a new NHS and NIHR study has found.
The study, jointly conducted by the Royal Marsden and the Institute of Cancer Research showed that, for soft tissue sarcomas, AI accurately graded 82% of the analysed tumours while only 44% were correctly scored via a biopsy.
Soft tissue sarcoma is a type of cancer that develops in the areas of the body that connect, support and surround main body structures. They can form in fat, muscles, nerves, blood and lymph vessels.
Around 4,000 people a year are diagnosed with sarcoma cancer in England, although there are 50 different types. The researchers focused on retroperitoneal sarcoma for this study – a sarcoma that develops in the back of the abdomen, making it hard to diagnose and treat.
The researchers also looked at how effective the AI was at distinguishing between the two main types of retroperitoneal sarcoma. They found the AI accurately predicted the type in 84% of cases, with radiologists managing 65%. This was based on 90 patients across Europe and the US.
Experts have highlighted faster, more accurate and more detailed diagnostics all as benefits of using this AI approach. The algorithm could also help clinicians map more tailored treatment plans and thus improve patient outcomes.
Study lead, Professor Christina Messiou, who is a consultant radiologist at the Royal Marsden, said: “We’re incredibly excited by the potential of this state-of-the-art technology, which could lead to patients having better outcomes through faster diagnosis and more effectively personalised treatment.
“As patients with retroperitoneal sarcoma are routinely scanned with CT, we hope this tool will eventually be used globally, ensuring that not just specialist centres – who see sarcoma patients every day – can reliably identify and grade the disease.”
She continued: “In the future, this approach may help characterise other types of cancer, not just retroperitoneal sarcoma.
“Our novel approach used features specific to this disease, but by refining the algorithm, this technology could one day improve the outcomes of thousands of patients each year.”
The full study results were published in The Lancet.
Image credit: iStock