Smartwatch data and AI aid early Detection of Parkinson's Disease
Researchers analysed data from more than 103,000 people who wore a medical-grade wearable device over seven-days measuring their speed of movement continuously. Slow movement is a hallmark symptom of the condition along with stiffness and shaking. The researchers found that using AI, they could accurately predict those who would later go on to develop Parkinson's disease. The accuracy of the AI model in identifying future Parkinson's cases surpassed that of traditional risk factors and early signs associated with the disease.
Current diagnostic methods often identify the disease at later stage when irreversible damage has already occurred. Parkinson's affects cells in the brain called dopaminergic neurons, located in an area of the brain known as the substantia nigra. By the time symptoms of Parkinson's begin to present, and a clinical diagnosis can be made, more than half of the cells in the substantia nigra will have died
The use of smart technology, specifically smartwatches, provides an accessible and cost-effective means of collecting data for early Parkinson's detection. With approximately 30 per cent of the UK population wearing smartwatches, this method holds promise for identifying those in the early stages of Parkinson's within the general population. Early detection allows for timely intervention, potentially improving patient outcomes and facilitating access to treatments once they become available.
The benefits of this research extend beyond clinical practice. The ability to identify people at risk of developing Parkinson's early can greatly enhance recruitment into clinical trials, thereby advancing research efforts in understanding and treating the disease. The study's findings, while significant, are limited by the lack of replication using alternative data sources.
The application of AI analysis to smartwatch data offers a promising avenue for early detection of Parkinson's as well as other diseases such as breast cancer, as discussed in a recent article, revolutionising diagnosis and generating timely intervention and improved patient outcomes.