How Might Parkinson’s Progress in a Person? Machine Learning Project Seeks Clues
A research team at The Australian National University (ANU) in Canberra is planning to collect information on subtleties like eye movement and posture sway in Parkinson’s patients, then use machine learning technology to spot indicators of how Parkinson’s disease (PD) might progress in individuals diagnosed with the neuromuscular disorder.
Dr. Deborah Apthorp of the ANU Research School of Psychology was awarded AU$138,930 (about $105,000) to conduct a study that will track a range of early PD symptoms, in order to determine if any can be used as an indicator of PD progression. The award, given by Perpetual Impact Philanthropy Grant program, will fund the work over five years.
“There are different types of Parkinson’s that can look similar at the point of onset,” Apthorp said in a press release. “But they progress very differently. We are hoping the information we collect will differentiate between these different conditions. Ultimately we’d like doctors to be able to conduct tests that can predict how the disease is likely to progress.”
Machine learning is defined by the global business analytics software and services firm SAS as “a method of data analysis that automates analytical model building. Using algorithms that learn from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look.” In other words, it involves computers with the ability to learn without being explicitly programmed, enabling them to grow and change when exposed to new data.
Using available medical technology, it can be difficult to determine what type of PD newly diagnosed patients have, or how quickly their condition will progress.
“The thing about Parkinson’s disease is that some people can be OK for quite a long time, while others progress more rapidly,” said Apthorp, whose research interests include motion perception, self-motion illusions (vection), and motion and form combinations in vision, time perception, temporal integration, attention, EEG, and EME bioeffects. She said the researchers will measure brain imaging, eye tracking, visual perception, and postural sway.
Observing that “human posture is an inherently unstable system, so you’re constantly making little corrections,” Apthorp added: “When you get Parkinson’s disease it becomes harder and harder to maintain that upright posture, and you have to think more about it. Eventually, as the disease gets further along, you start to fall and have trouble walking. There is also evidence that control of eye movement is related to parts of the brain that are impacted by Parkinson’s. We plan to look at all of these measures together.”
Once the data is collected, Apthorp’s team at the Research School of Psychology will collaborate with researchers from the ANU College of Engineering and Computer Science to analyze the findings.
“We’ll be using machine learning techniques to find patterns in the data that indicate degradation of motor function correlating with Parkinson’s,” Apthorp said. “The huge quantitative data set available from real time sway measurements will allow us to develop much more precise diagnostic indicators on the disease progression than can be obtained from traditional questionnaires.”
The investigation will be based at Canberra Hospital and involve 120 persons with Parkinson’s and an equal number of participants who don’t have the disease as a control group. Research will be conducted over five years, and is projected to begin in September.
Australian National University (ANU)