Smartwatches Detect Parkinson’s Disease

Smartwatches have the potential to revolutionise the early detection of Parkinson’s disease, according to new research. Scientists at the UK Dementia Research Institute and Neuroscience and Mental Health Innovation Institute conducted a study using data collected from smartwatches worn by participants. They found that the watches could accurately predict the development of Parkinson’s disease using artificial intelligence (AI) analysis of movement speed.

The study, which involved thousands of patients, discovered that accelerometers in smartwatches could detect subtle changes in movement patterns. By analysing the data, researchers were able to identify individuals who would later be diagnosed with Parkinson’s disease. This groundbreaking discovery could lead to the development of a new screening tool for early detection of the disorder.

The research was conducted as part of a long-term study by UK health authorities, tracking the health of over half a million individuals over the age of 40. In this particular study, 103,712 participants wore smartwatches for a week to record their activity. The data collected helped scientists identify an objective marker of Parkinson’s disease, indicating its presence before visible symptoms appear.

Smart Watches Could Detect Parkinson’s Up To Seven Years Before Hallmark Symptoms Appear #Parkinson #𝒟𝒾𝑒𝒷𝒪1 #𝒟𝒾𝑒𝒷𝒪37 #Biomedical #Technology #SmartWatch #ArtificialIntelligence #AI https://t.co/HYcrH09oLH

— George Obeid (@_Georgeobeid) July 4, 2023

Parkinson’s disease is a neurodegenerative movement disorder characterised by slow disease progression and symptoms such as slowness of movement, stiffness, coordination difficulties, and tremors. These symptoms manifest long before a formal diagnosis is made, often going unnoticed by the affected individuals. However, smartwatches equipped with accelerometers, magnetometers, and gyroscopes can detect these subtle motor or non-motor symptoms.

To analyse the data collected from smartwatches, researchers used artificial intelligence systems to detect patterns in the participants’ movement graphs. They observed a decrease in mobility between 7 am and 12 pm in individuals diagnosed with Parkinson’s disease. By training the AI system on this pattern, they successfully identified nearly 200 people who were diagnosed with the disease an average of 4.33 years after their movements were recorded. In some cases, the detection occurred up to 7 years earlier.

The availability of smartwatch data and the high prevalence of smartwatch usage, especially in the UK, present an opportunity to establish a centralised platform for early Parkinson’s disease detection. However, challenges related to technology, privacy, and legal implications need to be addressed.

Current treatments for Parkinson’s disease are symptomatic and do not prevent its progression. Experimental treatments aimed at slowing down the disease’s progression have yet to be successful. Therefore, the ability to detect Parkinson’s disease earlier through smartwatch technology opens the door to potentially using these experimental treatments during the prodromal phase of the disease.

Further research is needed to refine the detection capabilities of smartwatches, particularly by accumulating data over an extended period. Other markers, such as brain imaging, lifestyle factors, and blood biochemistry, are also being investigated for early detection of Parkinson’s disease. The integration of these various markers and the power of predictive algorithms driven by artificial intelligence could significantly advance early diagnosis and intervention for Parkinson’s disease.