Thanks to artificial intelligence, researchers have discovered that patients with Parkinson’s disease speak differently than patients without the disease.
- AI enables early diagnosis of Parkinson’s disease through speech analysis.
- Patients with this degenerative disease have a different way of speaking compared to people who are not ill, and this, long before the cognitive decline characteristic of the disease appears.
- They use fewer nouns and fillers like hum and uh but more verbs based on AI analysis.
Speech would be a marker of Parkinson’s disease according to a study conducted by a team of Japanese researchers who used a branch of artificial intelligence to analyze human language (NLP).
Parkinson’s disease impairs the language of those affected
Parkinson’s disease (PD) is a degenerative brain disease associated with motor symptoms (slow movements, tremors, rigidity and imbalance) and other complications, including cognitive, mental health, sleep and pain and sensory disturbances, indicates theWorld Health Organization (WHO).
In addition, patients with Parkinson’s disease exhibit a variety of speech-related problems, including impaired speech production and language use, the researchers point out.
That’s why they used this AI that allows computers to understand and interpret large amounts of human language data using statistical models to identify patterns. They were thus able to analyze the differences in the speech patterns of the patients according to 37 characteristics, using texts from free conversations.
Parkinson’s: patients use more verbs in their sentences
According to the AI, patients with this disease use fewer common names and proper names. In contrast, they use a higher percentage of verbs and case particles (an important feature of the Japanese language) per sentence.
Patients with Parkinson’s disease also utter shorter sentences than people in the healthy group. Additionally, the healthy control group also uses more fillers, such as ‘GOOD‘ Or ‘um‘, to connect the sentences.
Study co-author Dr. Katsunori Yokoi, of Nagoya University Graduate School of Medicine, Japan, explains: “When I asked them to tell about their day in the morning, a patient with Parkinson’s disease could say something like this, for example: “I woke up at 4:50. I thought it was a little early, but I got up. It took me half an hour to go to the bathroom, so I washed and dressed around 5:30. My husband cooked breakfast. I had my breakfast after 6 o’clock. Then I brushed my teeth and got ready to go out“.
Dr. Yokoi continues: “Whereas a healthy control group member might say something like this: ‘Well, in the morning I woke up at six o’clock, got dressed and, yes, washed my face. Then I fed my cat and my dog. My daughter cooked a meal, but I told her I couldn’t eat, and I, uh, drank some water.“.
Parkinson’s: AI detects it at 80%
The experiment presents a particularly promising aspect: it was in fact carried out on patients who did not yet show the cognitive decline characteristic of Parkinson’s disease.
Therefore, their findings offer a potential early detection means to distinguish patients with Parkinson’s disease:
“Even in the absence of cognitive decline, the conversations of patients with Parkinson’s disease differ from those of healthy subjects. When we attempted to identify patients with Parkinson’s disease or healthy controls on the basis of these conversational changes, we were able to identify patients with Parkinson’s disease with an accuracy of more than 80%”concludes Professor Masahisa Katsuno of Nagoya University Graduate School of Medicine in Japan, the co-author of the study.
