A British-Pakistani scientist based in Qatar says artificial intelligence-powered eye scans could soon help doctors detect dementia and diabetic nerve damage years before symptoms emerge, potentially transforming early diagnosis and treatment for millions of patients worldwide.
Professor Rayaz Malik of Weill Cornell Medicine-Qatar explained that a non-invasive eye scanning technology known as corneal confocal microscopy (CCM) can identify microscopic nerve damage linked to neurodegenerative diseases and diabetes in just two to three minutes.
The breakthrough highlights the growing convergence of artificial intelligence, medical imaging, and digital healthcare innovation as researchers increasingly use advanced analytics to improve early disease detection.
Eye-Based AI Diagnostics Expanding Healthcare Possibilities
Professor Malik said the cornea provides a unique diagnostic window because it contains one of the body’s densest networks of sensory nerves.
Originally developed for diagnosing eye infections and abnormalities, CCM technology has evolved into a powerful tool capable of identifying nerve fibre damage associated with diabetes, dementia, Parkinson’s disease, multiple sclerosis, schizophrenia, autism, and other neurological conditions.
According to Malik, one of the technology’s most important applications lies in detecting dementia long before clinical symptoms become severe.
“When patients come to the doctor with memory loss and are diagnosed with dementia, the underlying nerve damage has usually been developing for 10 to 15 years,” he explained.
Researchers have shown that some patients with mild cognitive impairment already display abnormal corneal nerve patterns on CCM scans years before developing dementia.
AI Dramatically Enhancing Diagnostic Accuracy
Artificial intelligence has significantly accelerated the technology’s capabilities by allowing automated analysis of thousands of microscopic imaging features within seconds.
Professor Malik noted that while human experts may identify only a handful of visible nerve characteristics, AI systems can evaluate more than 2,500 image features simultaneously to recognize disease-specific patterns.
He said AI models are now capable of identifying underlying neurodegenerative diseases with accuracy rates between 90 and 95 percent, with some studies in diabetic neuropathy and Parkinson’s disease approaching near-perfect sensitivity and specificity.
The technology could become particularly valuable for countries facing rising diabetes rates and ageing populations, including Pakistan and many Gulf states.
Early Detection Could Improve Long-Term Outcomes
Professor Malik emphasized that diabetic nerve damage can potentially be identified up to five years earlier through CCM scans.
Early diagnosis is becoming increasingly important because lifestyle interventions and metabolic management may help repair nerve damage when treatment begins at earlier stages.
The research also underscores broader global trends around AI-enabled diagnostics, predictive healthcare systems, and personalized medicine.
Despite the promising results, Professor Malik acknowledged that adoption has been slowed by limited device availability and resistance from parts of the medical community. However, increasing production of CCM devices by additional manufacturers may improve affordability and global accessibility.
Editor’s Note
AI-powered diagnostics are rapidly transforming healthcare from reactive treatment models toward predictive and preventive medicine. Technologies capable of identifying diseases years before symptoms appear could fundamentally reshape healthcare systems, particularly as ageing populations and chronic disease burdens continue rising globally. Early detection combined with AI-driven analysis may become one of the most strategically important frontiers in future digital healthcare ecosystems.
