For many people, our daily morning routines are quite simple to the point of monotony. We get dressed, have a shower, make breakfast and leave the house to start our days.
Although for people with Parkinson’s disease, these simple tasks we take for granted can be debilitatingly difficult. Parkinson’s affects approximately 1 in 500 people in New Zealand and about 1 percent of people aged over 60. While there is no cure, getting medication dosages right can significantly improve symptoms, particularly the sometimes-violent tremors that Parkinson’s is typically associated with.
Using technology to track symptoms
Keeping track of tremors is an effective means of tracking the severity of the disease and adjusting medications accordingly, although this process can be inefficient; relying on manual check-ups that can leave patients waiting for weeks.
That’s where technology comes in. Led by neurologist Dr Yun Hwang at Gosford Hospital on Australia’s Central Coast, a team of registrars and clinicians have teamed up with health-tech provider Digital Aged Care—part of enterprise software giant SAP—to develop a machine learning (ML) platform designed to improve the lives of Parkinson’s patients.
The system being developed uses ML to analyse thousands of images of spirals, drawn by patients suffering various tremors (of those diagnosed with Parkinson’s and other illnesses).
Machine learning and medication
These images are used to train an ML algorithm, so that it can automatically detect the severity of the tremor and whether it’s Parkinson’s-related, giving physicians a clear idea of which medications to prescribe or adjust existing dosages.
“I’ve been working within the Parkinson’s space for some years now and one of the biggest challenges has always been around monitoring symptom fluctuations during the day,” Dr Yun explains.
“We’re not there all the time to check up on the patient, which is where AI can be very useful.”
It’s not the first-time spirals have been used in Parkinson’s research, as they offer a good indication of how intense the tremors—and thus the disease—can be, especially when paired with ML.
Although spirals are just the beginning for Dr Yun and SAP, as the project’s next phase involves using smart watches to provide real-time insight into Parkinson’s severity. When wearing these watches, patients can draw spirals in the same way, with the watch using an accelerometer and gyroscope to generate a 3D model of the tremor, providing even greater insight into severity.
The team is also looking to rig the watch to track tremors 24/7, showing progress visually through graphs that map the times when patients are ‘on or off’ in terms of the tremor’s severity.
SAP Business Technology Platform
While it’s a compelling project, the team isn’t looking to do everything on their own. SAP is making the application open source, to inspire collaboration and reach as many patients as possible.
“We’re also going to give the platform to the Open Source University and developer communities, so that people can join us in building these new innovations collaboratively,” says SAP Digital Aged Care co-founder Simon Grace.
“We encourage all organisations and individuals to contribute and join our open source ecosystem so that we can help break down barriers for all people suffering from Parkinson’s disease.”
“Building these innovations only takes days to weeks, thanks to the easy to use tools and templates on the SAP Business Technology Platform”.
This article first appeared in Health Information New Zealand.
To find out more about the SAP Business Technology Platform (BTP), visit the SAP Business Technology Platform homepage