I was fully fit and healthy so it wasn’t a surprise when a fancy ‘medical MOT’ found no problems. But, eight months later, I suffered a catastrophic stroke. This is the sign everyone missed… and what YOU need to know
Sean Clifford, a physically fit 35-year-old, had no reason to believe he was at risk of stroke or any other serious medical problem.
But wanting to know more about his health in 2023, the New York father-of-two decided to pay £2,500 to have a full-body MRI.
Often called a health exam, the screening is supposed to check the body for early signs of disease. They rely on artificial intelligence (AI) software to detect subtle changes that might suggest something isn’t right.
Sean also believed he was in safe hands. Prenuvo, the company behind the scan, has an impressive roster of celebrity backers, including reality TV star Kim Kardashian, supermodel Cindy Crawford and actress Gwyneth Paltrow.
Moreover, the results were good. Prenuvo concluded that he showed no obvious signs of disease.
But just eight months later, Sean suffered a devastating stroke that left him partially paralyzed and brain damaged.
In September 2024, after a radiologist was contracted to re-evaluate Sean’s scan, his family filed a lawsuit against Prenuvo after finding visible signs that the arteries in Sean’s brain had narrowed, significantly increasing his risk of a stroke in the near future. According to the ongoing lawsuit, Sean’s stroke could have been prevented if Prenuvo’s AI software had flagged these narrowed arteries.
Experts say thousands of NHS patients now face similar life-threatening misdiagnoses.
Gwyneth Paltrow says she’s interested in Prenuvo’s ability to detect health problems before symptoms appear
Kim Kardashian touts benefits of scanning on Instagram and says it’s saved some friends and family
In recent years the NHS has invested millions in AI scanning technology to speed up diagnoses. Describing the technology as a ‘game changer’, ministers claim the move could reduce NHS waiting lists as AI can read scans faster than doctors.
Last year it was revealed that AI is now being used to analyze and interpret brain scans in every NHS stroke unit in England, and is also helping diagnose lung cancer in half of all hospitals.
But growing evidence shows AI can miss early signs of the disease, leading to tragic misdiagnoses like Sean’s, an expert told The Mail on Sunday.
Dr. New Jersey-based psychologist and founder of tech company Evidify, who analyzes the impact of artificial intelligence on radiographers’ decision-making process. “AI is a powerful tool, but it is unreliable,” says Joshua Henderson.
‘Research consistently shows that these systems fail in unpredictable ways.’
The need for drastic measures to speed up NHS MRI scans is clear. Scans, which use powerful magnetic and radio waves to create detailed images of organs and tissues, are vital for detecting early signs of serious diseases such as cancer, heart problems, stroke and fractures. Around five million procedures are carried out on the NHS every month but there are currently huge backlogs.
Patients should receive results back within six weeks. But last year’s data revealed that at any given time nearly 400,000 patients were waiting for their results for longer than that period. Research shows that for every month a cancer patient waits for MRI results, the risk of dying from cancer increases by approximately 10 percent.
Singer Rita Ora was also asked to undergo screening due to her mother’s breast cancer history
Research suggests that this backlog is largely due to a severe shortage of radiologists (specialists trained to analyze scans). According to the Royal College of Radiologists, there are around 3,000 vacancies in the NHS; This is a shortfall of roughly 30 percent. Experts say this has led to the Government adopting AI radiology software.
In theory, AI should not be used alone to diagnose patients. Instead, radiologists need to check the results to make sure the AI made the right decision.
In a study published in the medical journal Insights Into Imaging, AI was able to detect signs of stroke on an MRI scan in approximately 93 percent of cases. That means he misses about one in 14 shots. And most importantly, according to a 2024 study in the journal Radiology, experts were able to detect that in about a quarter of cases, AI simply made the wrong decision.
Dr Henderson argues that in the UK, where technology is being rapidly adopted across the NHS, these overlooked errors could put lives at risk. He says: ‘When a scan result is shaped by AI, patients deserve to know that the doctor is making independent clinical judgment and not simply following what the algorithm says.’
A spokesman for Prenuvo said: ‘We take any allegations seriously and are committed to dealing with them through the legal process.’
A spokesperson for the Department of Health and Social Care said: ‘AI tools are being used to assist, not assist, clinical decision-making, and all technologies used in the NHS must meet robust safety, effectiveness and regulatory standards before they are implemented.’
