Researchers have developed an AI tool that can detect Parkinson's disease from blood samples. Artificial intelligence can spot the possibility of the disease 15 years before the onset of symptoms with 96 per cent accuracy. Are we just witnessing a revolution in the medical field?
The technology developed could probably be adapted to diagnose other conditions, making it possible to detect and even prevent many incurable diseases earlier.
Parkinson's disease is the second most common neurodegenerative disease after Alzheimer's disease, and the number of cases is growing faster than other neurological conditions. It is characterised by the occurrence of unintentional movements such as tremor, rigidity or difficulty with balance, among others. It is caused by the loss of nerve cells in the brain, leading to reduced levels of dopamine, which in turn plays a key function in the motor system.
The researchers developed an AI tool called CRANK cells -MS, through which they identified combinations of metabolites unique to the study participants who developed Parkinson's disease. Then, based on these combinations alone, they identified up to 96 per cent of people who developed the disease over a 15-year period.
The experts also reported that CRANK cells -MS could diagnose Parkinson's disease in 84.3 per cent of cases based on skin sebum samples.
This tool will be a game changer in the diagnosis of Parkinson's disease, providing an objective, reliable and highly accurate way to predict conversion in patients at risk of developing the disease, said Dr Mya Schiess, Professor of Neurology and Director of Movement Disorders and Neurodegenerative Diseases.
Further research is needed to conclusively assess the tool's performance, but the availability of CRANK cells -MS provides an excellent opportunity for collaborative scientific experimentation and refinement of the tool's algorithms, potentially advancing disease knowledge.