Parkinson’s severity assessed through drawing

0
IMAGE

Credit: Courtesy of Dinesh Kumar and Ms. Poonam Zham of the 'Affordable diagnostics' group in RMIT University, Melbourne, Australia

Researchers in Australia asked volunteers to draw a spiral on a sheet of paper. By analyzing how long it took them to draw the spiral and how hard they pressed on the paper with the pen, the team could not only tell which volunteers had Parkinson's disease, they could also tell how severe it was.

Parkinson's disease is a neurodegenerative disorder that causes shaking, muscle rigidity and difficulty with walking. Many treatment options for Parkinson's are only effective when doctors diagnose the disease early, and when symptoms are very noticeable it may be too late. It's also important for doctors to be able to tell how severe the disease is, to make the right treatment decisions, and to follow-up the progression of symptoms.

One way to contribute to the diagnosis of Parkinson's involves getting patients to use a pen. Certain symptoms that appear early in the disease, such as rigidity, can interfere with a patient's ability to write or sketch. Handwriting can be influenced by a person's level of education and language proficiency, so a better alternative involves sketching a shape, such as a spiral.

One drawback to this approach is that only an expert can interpret the sketches, meaning that routine check-ups at a doctor's surgery aren't possible. However, even for an expert, it can be difficult to tell how severe the disease is from the sketches alone, especially at the early stages of the disease.

Previous research has found that Parkinson's patients tend to move their pen more slowly when sketching, and they also use less pressure on the page. While these factors are useful for telling if someone has Parkinson's or not, so far researchers have not been able to reliably gauge how severe someone's disease is, using pen speed or pressure.

In a new study, recently published in Frontiers in Neurology, a team of researchers in Australia set out to develop an automatic system to contribute to the diagnosis of Parkinson's, and to asess its severity, from the comfort of a community doctor's office. "Our aim was to develop an affordable and automated electronic system for early-stage diagnosis of Parkinson's disease, which could be used easily by a community doctor or nursing staff," explains Poonam Zham, a researcher involved in the study.

The researchers developed specialized software and combined it with a tablet computer that can measure writing speed, and a pen that can measure pressure on a page. They used the system to measure pen speed and pressure during a simple spiral sketching task in a sample of healthy volunteers and Parkinson's patients with different levels of disease severity. In a world-first, the system also mathematically combines pen speed and pressure into one measurement, which the team calls the Composite Index of Speed and Pen-pressure (CISP) score.

The system measured slower pen speeds, pen pressures and CISP scores in the Parkinson's patients, compared with the healthy volunteers, and all three measurements clearly indicated whether a participant had Parkinson's or not. On their own, pen speed and pressure were not sufficiently different between patients with different levels of Parkinson's severity, for the system to distinguish between them.

However, using the new CISP score, the system could tell whether the patients had level 1 or level 3 Parkinson's, using a particular disease severity scale. "The system can automatically provide accurate Parkinson's diagnosis and could also be used by community doctors to monitor the effect of treatment on the disease," says Zham. "This simple device can be used by community doctors for routine screening of their patients every few years after the patients are above middle-age."

###

Media Contact

Melissa Cochrane
[email protected]
41-787-246-393
@frontiersin

http://www.frontiersin.org

Related Journal Article

http://dx.doi.org/10.3389/fneur.2017.00435

Leave A Reply

Your email address will not be published.