Researchers in Lithuania have explored whether artificial intelligence (AI) could be used to distinguish between the voices of people with and without Parkinson’s.
The team built an AI learning model using vocal data from 50 Italian people, including 28 people with the condition and 22 without. Then, the AI model evaluated voice recording samples of 104 Lithuanian people saying the same sentence as the previous group.
This data enabled the researchers to identify instances of Parkinson’s with 90% accuracy among the Italian samples and 80% accuracy among the Lithuanian samples.
Commenting on the study, the scientists noted that the AI model has the potential to supplement traditional methods of diagnosis. Lead author Rytis Maskeliūnas said in a press release. “We are not creating a substitute for a routine examination of the patient – our method is designed to facilitate early diagnosis of the disease and to track the effectiveness of treatment.”
