With an estimated 12 million people worldwide currently living with Parkinson’s disease, and projections indicating a doubling of that number by 2050, the search for earlier detection methods and improved treatments is increasingly critical.The neurodegenerative disorder, characterized by motor impairments and stemming from dopamine-producing nerve cell loss, currently lacks definitive diagnostic tests. However, a Swiss physician is pioneering a new approach, leveraging artificial intelligence and sensor technology to identify subtle indicators of the disease-potentially years before traditional symptoms manifest.
The number of people living with Parkinson’s disease is rapidly increasing worldwide, and researchers are exploring new ways to detect the condition earlier and improve quality of life for those affected. Currently, an estimated 12 million people globally have Parkinson’s, and that number is projected to more than double to around 25 million by 2050, making it the fastest-growing neurodegenerative disease.
Parkinson’s disease is a progressive disorder that affects the nervous system, leading to motor impairments like slowed movement, stiff muscles, and tremors, often in the hands. These symptoms arise from the loss of nerve cells in the brain that produce dopamine, a chemical messenger crucial for transmitting signals between nerve cells. While the exact cause remains unknown, increasing attention is being paid to environmental factors such as air pollution and pesticide exposure as potential triggers. The rising prevalence is also linked to increasing life expectancy, as most individuals are diagnosed after age 60.
A Unique Path to Parkinson’s Research
Early detection is key to managing Parkinson’s and potentially slowing its progression, according to Robert Ilesan, a chief physician at the Lucerne Cantonal Hospital in Switzerland. Ilesan’s journey to Parkinson’s research began with a broader focus on the impact of neurological conditions on public health. He currently serves as a chief physician in the Department of Oral and Maxillofacial Surgery at the hospital, specializing in facial and jaw asymmetries and sleep apnea. He also leads the hospital’s new 3D laboratory and treats patients with facial injuries.
Ilesan’s interest in Parkinson’s stemmed from his previous work as a health policy advisor in his native Romania, where he observed the widespread impact of neurodegenerative diseases. “I realized how common Parkinson’s and other neurodegenerative diseases are,” he said. “Besides the great suffering for those affected and their families, the disease also causes significant costs in healthcare.” He also noted that, as a specialist in oral and maxillofacial health, he frequently encounters patients with Parkinson’s who experience symptoms like difficulty swallowing and speaking, as well as dry mouth, all of which impact oral hygiene.
Driven by these observations, Ilesan launched a research project with specialists in neurology and computer science to explore the use of artificial intelligence (AI) and other technologies for earlier diagnosis and improved monitoring of Parkinson’s disease. The goal is to enable earlier treatment initiation and personalized adjustments to care based on individual patient health status.
How Algorithms and Sensors Can Reveal Early Signs
More than 200 years after James Parkinson first described the “Shaking Palsy” in 1817, diagnosis still relies primarily on clinical assessment – evaluating typical symptoms and neurological findings. Currently, there is no definitive laboratory test or imaging scan to confirm a diagnosis. However, researchers are increasingly investigating imaging techniques to develop more objective criteria for early detection and disease monitoring.
“Moreover, artificial intelligence and modern sensor technology now make it possible to capture symptoms with precision,” Ilesan explained. His team has developed specialized sensors and algorithms to analyze movement patterns while walking, speech patterns via smartphone recordings, and handwriting samples from scanned images. These symptoms – a shuffling gait with reduced arm movement, a monotonous voice, and changes in handwriting – often appear in the early stages of the disease, making them ideal for early detection.
Initial tests of the algorithms and sensors developed by Ilesan and his colleagues have shown promising results. However, Ilesan emphasized that “artificial intelligence can only calculate probabilities that Parkinson’s disease is present. The actual diagnosis is still made by a doctor.”
While the new tools are not yet ready for widespread clinical use, Ilesan is focused on refining the analysis software and sensor prototypes. He hopes to develop smaller, more wearable sensors, potentially integrated into clothing. He also anticipates that companies will become interested in his research and eventually take over the production of these medical devices.
Identifying Risks Before Diseases Develop
Ilesan plans to apply the lessons learned from his Parkinson’s research to other conditions. He is already working on using AI and sensors to improve the early diagnosis of sleep apnea. “Corresponding technologies could even be designed in the future to detect numerous diseases simultaneously,” he said, envisioning smartphone app-based tests that could help healthy individuals identify potential health risks. “We should take advantage of the opportunities offered by new technologies – while of course ensuring privacy and data protection,” Ilesan concluded.
