An AI model developed by a team of researchers, scientists, and doctors can quickly detect cancer, potentially speeding up diagnosis and treatment of the disease, which is responsible for almost one in six deaths globally.
Experts from the Royal Marsden NHS Foundation Trust, the Institute of Cancer Research, London, and Imperial College London designed the AI tool that identifies cancerous growths found on CT scans. The team used radionics to extract essential information from medical images and tested the AI model on about 500 patients with large lung nodules.
The model’s accuracy of 0.87 on the area under the curve measure showed that it performed better than the current methods. Although the Libra study, which is supported by Cancer Research UK and other organizations, is still in its early stages, its potential benefits include accelerating cancer detection, fast-tracking treatment, and simplifying CT scan analysis. Dr. Richard Lee, the chief investigator of the study, stated that the team hopes to employ innovative technologies like AI to hasten cancer detection.
Meanwhile, the team leader at the Institute of Cancer Research and consultant physician in respiratory medicine at the Royal Marsden highlighted the urgency for new initiatives to expedite cancer detection, particularly in the case of lung cancer, the leading cause of cancer-related deaths worldwide, accounting for 21% of such deaths in the UK. Early diagnosis is vital for effective lung cancer treatment, but over 60% of lung cancer cases in England are diagnosed at stage three or four.