Imaging analysis tells Parkinson’s from atypical parkinsonism

Technique looks at water molecules in white brain matter

Patricia Inácio, PhD avatar

by Patricia Inácio, PhD |

Share this article:

Share article via email
An illustration shows the human brain in profile.

BrainTale’s noninvasive imaging analysis software, BrainTale-care, may be able to distinguish Parkinson’s disease from atypical parkinsonism.

That’s according to data from an advanced MRI technique, called diffusion tensor imaging, that measures the diffusion of water molecules in the brain’s white matter. White matter contains nerve fibers connecting different brain regions.

“The easy access to diffusion tensor markers in the clinical setting with BrainTale’s platform paves the way for improving differential diagnosis and accelerating treatment development,” Stéphane Lehéricy, MD, PhD, head of the neuroradiology department at Pitié-Salpêtrière Hospital in Paris, said in a company press release. “This technology also enables us to better understand the pathophysiology [processes] of different diseases and facilitates patient care in a non-invasive and reliable way.”

The findings were shared in a presentation titled, “Differential diagnosis model in parkinsonian syndromes derived from calibrated diffusion tensor imaging,” at the 10th European Academy of Neurology congress held June 29-July 2 in Helsinki, Finland. The data were presented by Vincent Perlbarg, PhD, BrainTale’s co-founder and scientific director, and Martin Grange, a data scientist for the company.

Recommended Reading

How a herky-jerky, wonky walk led to a Parkinson’s diagnosis

Diagnosis can be complex

Parkinson’s diagnosis is a complex process, requiring a complete medical history, along with physical examination and tests.

The fact that other disorders share some Parkinson’s symptoms can make diagnosis even harder. These conditions, called atypical parkinsonism, include multiple system atrophy (MSA), dementia with Lewy bodies, and progressive supranuclear palsy (PSP).

Changes in the brain’s microstructure, specifically white matter, have been linked to poor motor control and affect brain regions important for executive function — skills involved in daily functioning, planning complex behaviors, and solving problems. Other neurological diseases with reported changes in the white matter include multiple sclerosis and amyotrophic lateral sclerosis.

The researchers analyzed MRI scans from 189 patients with Parkinson’s-like symptoms enrolled in three distinct clinical trials. A total of 92 patients had Parkinson’s, 45 had MSA, and 42 had PSP.

Receiver operating function analysis, a test with a range of values from 0.5 to 1 in which higher scores reflect a better ability to accurately differentiate between two groups, showed an area under the ROC (receiver operating characteristic) curve of 0.87. Mean specificity was 0.74, and mean sensitivity was 0.90. A test’s sensitivity is its ability to correctly identify those with a given disease, while specificity refers to correctly identifying those without it.

The company said the results show that BrainTale’s white matter markers may be able to generate “an efficient and robust classification of parkinsonian syndromes, more specifically, differentiating PD vs MSA and PSP, two frequent confusing syndromes.”