Ataxia is a collection of rare, neurodegenerative brain diseases that cause speech, swallowing, and movement problems. “Slurred speech, irregular rhythm. Speech is slow but can speed up. “All of this impairs communication,” says DZNE researcher and UKB neurology specialist Dr. Marcus Grobe-Einsler. “A six-level classification system assesses speech disturbance severity.
Clinical practitioners have manually classified this. Time-consuming and subjective. We demonstrated in a proof-of-concept study that computer technology can automate and objectify the categorisation. Our method could simplify ataxia severity assessment.”
PeakProfiling GmbH assisted Grobe-Einsler and colleagues in these experiments. Berlin-based firm analyzes voices and noises. 67 patients with mild or severe ataxia provided voice recordings for this investigation. Standardized questions were answered.
The study participants discussed their hobbies and counted from 1 to 10 and back again. The researchers used sound analysis software and “machine learning” techniques to identify over 100 typical traits in the participants’ speech rhythm and loudness modulations.
Next, the automated analysis system was reduced to match the severity grade offered by a panel of three experts who had evaluated the voice samples. Expert opinion was used. The computer-assisted approach hit 80% of recordings that were excluded from the software’s optimization procedure.
“We now intend to further refine our method in larger studies and transfer it from German to other languages in international cooperation,” says Prof. Thomas Klockgether, DZNE Director of Clinical Research and UKB Department of Neurology head.
The neurologist says ataxia patients’ speech impairment is a key health indicator. Thus, an objective and automated assessment approach could benefit research and therapeutic practice. Our method could help track the disease’s progression, and its automation makes it efficient for large-scale studies. Drug trials benefit from this. New, albeit experimental, treatment techniques have given ataxia a boost.
He adds that smartphone apps can incorporate suitable software. Ataxia causes health changes that can only be tracked through clinic visits. Smartphones and digital technologies may do this more accurately, and the software could notify patients how logopedics or other treatments affected their speech. Klockgether said many patients desire direct input.