Cancer is a serious illness that is made all the more frightening by the myriad forms it takes and the difficulty of diagnosing many of those forms. Although some types of the disease, such as skin cancer, can be spotted and treated quickly, others are stealthier and – as a result – much deadlier.

Pancreatic cancer is well-known for being a sickness whose symptoms manifest after the illness has largely taken hold. This is because the pancreas is buried deep within the body, making detection difficult.

As one medical scientist in the US put it, the organ is located in “high-price real estate” – a part of the body surrounded by several blood vessels and important parts of the gut, which can make surgery difficult.

With those suffering from pancreatic cancer facing a survival rate of less than five per cent, doctors and scientists are always working on better ways to detect and defeat the disease. They will have been buoyed by recent news that a Mayo Clinic-developed AI model can help detect pancreatic cancer on routine abdominal CT scans up to three years before it is diagnosed.

The Radiomics-based Early Detection Model looked at 2,000 CT scans, including those from patients later diagnosed with pancreatic cancer – all originally interpreted as normal. Identifying 73 per cent of prediagnostic cancers at a median of about 16 months before diagnosis is nearly double the detection rate of specialists reviewing the same scans without the assistance of AI, a study showed.

This model is part of a wave of medical and healthcare developments being powered by AI. The UAE was quick to see the technology’s potential, with government and companies investing heavily for years in AI to advance medical research, streamline administrative processes and give patients the best chance of a positive outcome. As AI looks poised to play a growing role in oncology and beyond – such as helping doctors to detect neurological conditions and infectious diseases – it is important to make sure that it can be accessed by as many people as possible.

Without widespread adoption, even revolutionary tools can fail to live up to their potential. Were medical AI to remain the preserve of leading hospitals in advanced economies, the technology’s impact will be limited. There are a number of challenges in getting AI from the research lab into clinics and hospitals wards the world over.

The end goal must be to make advanced AI tools a routine part of day-to-day medicine

Many hospitals lack the computing resources needed to run AI. Healthcare systems also require strong clinical validation of AI-powered diagnostics and treatment. Financing medical AI is also critical; hospital administrators and health ministries need to see cost-effective results.

Certainly, there are ways to meet these challenges. Commercially available cloud platforms can provide more of the computing power that AI needs. Making AI research readily available to medical centres globally, in addition to adjusting insurance schemes and government programmes to account for AI adoption could help to remove financial barriers.

The end goal must be to make advanced AI tools a routine part of day-to-day medicine. By taking steps now to not only develop the technology but popularise it, AI can truly live up to its potential – and save many more lives.