The earliest indication of Parkinson’s disease may be altered speech

Study tried to identify early symptoms of Parkinson’s disease using voice data.


Parkinson’s disease (PD) is the second most prevalent neurodegenerative disease. The pathogenetic mechanisms of the disease are variable and not yet fully understood. The disease presents itself with a broad spectrum of motor and non-motor symptoms, including those that directly affect a person’s capability to operate.

Rytis Maskelinas, a Lithuanian researcher from Kaunas University of Technology (KTU) and associates from the Lithuanian University of Health Sciences (LSMU), attempted to recognize early symptoms of Parkinson’s disease using voice data.

Maskelinas, a researcher at KTU’s Department of Multimedia Engineering, claims that when motor activity declines, the ability of the vocal cords, diaphragm, and lungs to function also declines.

“Changes in speech often occur even earlier than motor function disorders, which is why the altered speech might be the first sign of the disease.”

Early-stage Parkinson’s disease individuals may speak more quietly, more monotonously, less expressively, slowly, and fragmentarily. It can be quite challenging to detect this by ear. Hoarseness, stuttering, mangled word pronunciation, and absence of pauses between words might become more noticeable as the disease progresses. 

Scientists considered all these symptoms to develop a system to detect the disease earlier. They used AI to analyze and assess speech signals, where calculations are done and diagnoses made in seconds rather than hours.

The results of this study are specifically suited to the features of the Lithuanian language, broadening the AI language database.

Kipras Pribuišis, lecturer at the Department of Ear, Nose, and Throat at the LSMU Faculty of Medicine, said, “The study was only carried out on patients already diagnosed with Parkinson’s: “So far, our approach can distinguish Parkinson’s from healthy people using a speech sample. This algorithm is also more accurate than previously proposed.”

The speech of Parkinson’s patients and healthy subjects were recorded using a microphone in a soundproof booth, and an artificial intelligence algorithm was “learned” to perform signal processing by evaluating these recordings. The algorithm could eventually be turned into a mobile app because it doesn’t require sophisticated hardware.

Maskeliūnas said“Our results, which have already been published, have a very high scientific potential. Sure, there is still a long and challenging way before it can be applied in everyday clinical practice.”

Scientists are further planning to include more pateints in the study to collect more data and determine whether the proposed algorithm is superior to alternative methods used for early diagnosis of Parkinson’s.

Journal Reference:

  1. Rytis Maskeliunas et al. A Hybrid U-Lossian Deep Learning Network for Screening and Evaluating Parkinson’s Disease. Appl. Sci. 2022, 12(22), 11601; DOI: 10.3390/app122211601
- Advertisement -

Latest Updates