The diagnosis of Parkinson’s disease has shaken many lives. More than 10 million people worldwide are living with it. There is no cure, but if symptoms are noticed early, the disease can be controlled. As Parkinson’s disease progresses, along with other symptoms, speech changes.
Lithuanian researchers from the University of Technology Kaunas (KTU), Rytis Maskeliūnas, along with colleagues from the Lithuanian University of Health Sciences (LSMU), tried to identify the early symptoms of Parkinson’s disease using sound data.
Parkinson’s disease is usually associated with loss of motor function – tremors, stiff muscles or balance problems. According to Maskeliūnas, a researcher at KTU’s Department of Multimedia Engineering, when motor activity decreases, so does the function of the vocal cords, valves, and lungs: “Speech changes often occur faster than motor function problems. “Reasons to change. Speech may be the first sign of the disease.”
Expanding the AI language database
According to Professor Virgilijus Ulozas at the Department of Ear, Nose and Throat at LSMU Medical College, patients with early-stage Parkinson’s disease may speak more calmly, may have more loneliness, slower speech. And there is more scattering, and this is hard to notice by ear. As the disease progresses, hoarseness, hoarseness, slurred pronunciation of words, and loss of pauses between words may become more pronounced.
Taking into account these symptoms, Lithuanian researchers have developed a system to look for the disease earlier.
Researcher KTU Maskeliūnas says: “We do not develop alternatives for routine patient examination. Our approach is designed to facilitate early diagnosis of the disease and to monitor the effectiveness of treatment.”
According to him, the link between Parkinson’s disease and speech disorders is not new to the world of digital signal analysis, it has been known and researched since the 1960s. However, thanks to advanced technology, it is possible to extract more information from speech.
In their study, researchers used artificial intelligence (AI) to analyze and evaluate speech signals, where calculations were performed and analyzed in seconds rather than hours. The study was also unique – the results were adapted to the specific characteristics of the Lithuanian language, thus expanding the AI language database.
Algorithms will be mobile applications in the future
Commenting on the progress of the study, Kipras Pribuišis, a professor in the Department of Ear, Nose and Throat at LSMU Medical College, stressed that it only applies to patients who have already been diagnosed with Parkinson’s disease. This our method is able to distinguish Parkinson from healthy people using speech samples.
In the soundproof booth, a microphone is used to record the speeches of healthy Parkinson’s patients, and an artificial intelligence algorithm “learned” to process the signal by evaluating these recordings. The algorithm does not require powerful hardware and can be transferred to mobile applications in the future, the researchers point out.
“Our published results have a high scientific potential. Certainly there is still a long and difficult way to go before it can be implemented in everyday clinical practice,” says Maskeliūnas.
According to the researchers, the next steps include increasing the number of patients to collect more data and determine if the proposed algorithm is better than the alternative method used for the initial diagnosis of Parkinson’s disease. In addition, it will be necessary to check whether the algorithm works well not only in a laboratory-like environment, but also in the doctor’s office or in the patient’s home.