The study was led by scientists from the Medical Research Council (MRC)-funded Dementia Research UK Institute at Cardiff University and Dr Kathryn Peall, MRC Clinical Research Fellow at Cardiff University.

The researchers analyzed the data collected by the smartwatches over a period of seven days by measuring the movement speed of the participants.

They found that using artificial intelligence (AI) they could accurately predict those who would later develop Parkinson’s disease.

The researchers say this could be used as a new screening tool for Parkinson’s disease, allowing the disorder to be detected at a much earlier stage than current methods allow.

The need for better detection

Parkinson’s disease affects cells in the brain called dopaminergic neurons, located in an area of ​​the brain known as the substantia nigra.

It causes motor symptoms such as tremors, rigidity (stiffness) and slowness of movement.

By the time these characteristic symptoms of Parkinson’s disease begin to appear and a clinical diagnosis can be made, more than half of the cells in the substantia nigra will have already died.

Therefore, there is a need for cheap, reliable and readily available methods for early detection of changes so that intervention can be carried out before the disease causes extensive brain damage.

UK Biobank data

Researchers analyzed data collected from 103,712 UK Biobank participants who wore a medical-grade smartwatch over a seven-day period from 2013 to 2016.

The devices continuously measured the average acceleration, which means the speed of movement, during a week.

They compared data from a subset of participants who had already been diagnosed with Parkinson’s disease with another group who received the diagnosis up to seven years after the smartwatch data was collected.

These groups were also compared with healthy people of the same age and sex.

Using AI

Researchers have shown that using artificial intelligence, it is possible to identify participants who will later develop Parkinson’s disease from data from their smartwatches.

Not only could these participants be distinguished from healthy control subjects in the study, but the researchers extended this to show that AI can be used to identify individuals who will later develop Parkinson’s disease in the general population.

They found that it was more accurate than any other risk factor or other recognized early sign of the disease in predicting whether someone would develop Parkinson’s disease.

The machine learning model was also able to predict time to diagnosis.

A valuable screening tool

Study leader Dr Cynthia Sandor, Emerging Leader at the UK Dementia Research Institute at Cardiff University, said:

Smartwatch data is readily available and inexpensive. As of 2020, around 30 percent of the UK population wears a smartwatch. Using this type of data, we could potentially identify individuals in the very early stages of Parkinson’s disease within the general population.

Here we have shown that one week of collected data can predict events up to seven years into the future. With these results, we could develop a valuable screening tool to aid in the early detection of Parkinson’s disease. This has implications both for research, in improving recruitment to clinical trials, and for clinical practice, allowing patients to access treatments at an earlier stage, in the future when such treatments become available.

A limitation of the study is the lack of replication using another data source, as there are currently no other comparable data sets that would allow for a similar analysis. However, extensive evaluation was performed to mitigate any biases.

The study was funded by the UK Dementia Research Institute, the Welsh Government and Cardiff University.