Wearables can detect Parkinson’s disease early on
Long before a clinical diagnosis of Parkinson’s is possible, movement tracking from wearables can provide enough data for early detection. One study showed that an evaluation by AI can identify the disease up to seven years prior to any visible symptoms.
Movement data collected by wearables such as smartwatches could be used in the future for early detection of Parkinson's disease, as was revealed in a study published in Nature Medicine, titled «Wearable movement-tracking data identify Parkinson’s disease years before clinical diagnosis». The researchers from Cardiff University were able to show that movement tracking data can detect Parkinson’s disease in its early stages better than other biomarkers such as genetics, laboratory values or lifestyle.
Slowly progressive movement changes typical of Parkinson’s disease
Parkinson’s disease is a degenerative nerve disease in which neurons in the brain progressively degrade. Around one per cent of people worldwide over the age of 60 suffer from the disease and the associated symptoms, such as muscle rigidity and lack of movement, or so-called resting tremor (e.g. involuntary movement of the hands). Two of the most famous people with Parkinson’s disease are boxing legend Mohammed Ali and actor Michael J. Fox (aka Marty Mcfly from Back to the Future). A diagnosis is usually made when the characteristic resting tremor becomes visible. The problem? By then, a significant proportion of the brain cells responsible for sending the signals needed to plan and initiate movement are already defective. Those affected will have been suffering from subtle motor disorders, such as a slowing of movements, for years by this point.
Motion data from wearables may serve as an early indicator
This is where early detection through smartwatches and other wearables comes into play. Most of these sports wristbands have integrated accelerometers, that is, sensors used to measure movement. From the data collected, conclusions can be drawn about movement changes that occur in the early stages of Parkinson’s disease – but have so far remained undetected. If the data from the wearables is uploaded to a server, an artificial intelligence can detect patterns indicative of early-stage Parkinson’s disease and advise seeing a doctor.
AI recognises disease pattern as early as seven years before diagnosis
For their study, the researchers drew on measurement data from the UK Biobank, a large-scale database and research resource that contains the detailed genetic and health information of half a million UK participants.
Between 2013 and 2015, 100,000 participants aged 40 to 69 had already agreed to wear a wristband for one week to record their movement data. By early 2021, 1,984 of these study participants had been diagnosed with Parkinson’s disease. The researchers were able to specifically look for patterns in the past movement data of the sufferers. This showed that the wearables had recorded the slowing of movements characteristic of Parkinson’s disease up to seven years before diagnosis, and that an AI could recognise these patterns. In other words, AI can detect Parkinson’s long before the individual affected notices any symptoms.
Risk of a frequent false alarm
Movement data proved better for making a diagnosis than other biomarkers such as genetic or blood tests, or even the evaluation of patients’ lifestyle questionnaires. The changes to mobility that precede Parkinson’s disease don’t resemble any other neurological disease, making the results relatively clear and reliable. However, even this can’t guarantee an absolutely reliable diagnosis. In other words, the large-scale analysis of movement data from smartwatches and the like would inevitably also result in many misdiagnoses.
How useful is early detection in the absence of treatment options?
And finally, there’s the question of how beneficial early diagnosis is of a disease for which there’s currently no treatment, let alone cure. Since there are no drugs or other therapies available that favourably affect the course of the disease, early diagnosis at this time could prove more harmful than good. But medical advances could quickly change that.
Header image: Ketut Subiyanto/pexels
Science editor and biologist. I love animals and am fascinated by plants, their abilities and everything you can do with them. That's why my favourite place is always outside - somewhere in nature, preferably in my wild garden.
