AI Tool Detects Parkinson’s up to 15 Years Early with 96% Accuracy
Researchers have developed an artificial intelligence (AI) tool that can predict Parkinson’s disease from blood samples with 96% accuracy, potentially up to 15 years before symptoms emerge. This breakthrough could revolutionize how this neurodegenerative condition, the second most common after Alzheimer’s disease, is diagnosed and managed.
Around 90,000 people are diagnosed with Parkinson’s in the United States every year, with over 10 million people living with the condition worldwide. Currently, Parkinson’s is diagnosed based on symptoms, medical history, and physical examination, as there are no specific tests for the disease.
Researchers used an AI tool called CRANK-MS to analyze blood samples from 78 individuals from Spain, who were followed for 15 years. Half of them were eventually diagnosed with Parkinson’s. CRANK-MS was able to identify unique combinations of metabolites—chemicals created when the body breaks down food, drugs, and its own tissue—in those who developed Parkinson’s with a high degree of accuracy.
The AI tool demonstrated superior accuracy compared to existing clinical assessments by movement disorder specialists, which have an accuracy rate of around 80%. The researchers also reported that CRANK-MS could diagnose Parkinson’s 84.3% of the time from skin sebum samples in a second cohort of 274 NHS patients.
Dr. Daniel Truong, a neurologist not involved in the study, highlighted that the AI tool’s ability to detect Parkinson’s up to 15 years before symptoms occur could provide the opportunity for intervention and treatment at an earlier stage. This could significantly improve patient outcomes and quality of life.
However, the research does have its limitations. Blood sample collection and processing might not always be feasible, and the accessibility of samples for early detection in asymptomatic individuals could be challenging. Furthermore, the specific biological mechanisms and associations of these metabolites with Parkinson’s disease are not yet fully understood, necessitating further research to validate and understand the underlying metabolic pathways and causative relationships.
Overall, this study represents a significant advancement in the detection and management of Parkinson’s disease and may pave the way for similar approaches to diagnosing other neurodegenerative conditions in the future.
— By Annie Lennon on May 17, 2023 — Fact checked by Hannah Flynn
