Detect Parkinson’s by laughter: the discovery of Spanish scientists that can stop its progress

Screening and early detection is very important to stop the progression of Parkinson’s because, although it has treatment, it has no cure Laughter “can give interesting clues” and serve as support for doctors’ decisions, says Rubén San Segundo Spanish specialists are now seeking financing to develop the program and a mobile applicationDetecting Parkinson’s through laughter is possible. This has been demonstrated by a group of Spanish researchers, in a recognized and promising project that is now seeking funding to expand the database and begin clinical trials, Rubén San Segundo, one of its developers, tells NIUS. The program is capable, through artificial intelligence, of finding, at an early stage, symptoms of this disease in people’s laughter. Something of great importance since, although there is no cure for Parkinson’s, its early detection makes it possible to stop its progress and improve the quality of life of those affected. And they are many. In the world, there are 10 million people with Parkinson’s. Specialists from around the world are investigating this neurodegenerative disease and characteristic tremor that occurs when neurons do not produce enough dopamine. There are many methods for its detection and analysis: from the study of the breathing of patients during sleep to their smell. So does a Scottish nurse who claims to be able to detect sick people by how they smell. Now, thanks to Spanish researchers, Parkinson’s can also be analyzed through laughter. In addition to diagnosing new cases, laughter can help to closely monitor already diagnosed patients and better adjust medication, avoid unnecessary trips to the consultation and prioritize the waiting list, if there is one. This is how Rubén San Segundo, professor and director of the Information Processing and Telecommunications Center of the Polytechnic University of Madrid, details it. But how is Parkinson’s disease detected through laughter? The system discovers motor abnormalities in the vocal tract, in the set of muscles that articulate speech and laughter and, based on these alterations, can determine the stage of Parkinson’s. To do this, laughter is analyzed with voice recognition methods and automatic classification techniques. During the trial, they collected the laughter of 120 healthy people and people with Parkinson’s disease, in a database of more than 20,000 samples. The laughs were recorded in a studio while people watched humorous half-hour videos with jokes and monologues. The laughs recorded during that session were manually segmented to generate the data with which to train the developed system. Funding needed to develop the project The program achieved an accuracy rate of 83% when detecting laughter from sick and healthy people. The application can be, therefore, -assure those responsible for the trial- perfectly valid to support clinical decision, so it is expected that, soon, the analysis of laughter can be carried out through an application installed in the mobile phone that is capable of performing the analysis and producing a reliable result in a few minutes. The teams from the Polytechnic University of Madrid, the Complutense University and the University of Zaragoza have seen that it is possible to do so. Now they require financing to carry it out. To do this, they need to record a larger database and involve a medical team to carry out clinical trials. If they succeed -they are working on it, announces Rubén San Segundo-, in a few months the technical application could be developed, although later we will have to wait for the clinical trials and, until the doctors validate the system and it is incorporated into the protocols, several years may pass, but “little by little”, the expert in voice signal processing and artificial intelligence is encouraged. It is not the first time that artificial intelligence has been applied to medicine and more specifically to Parkinson’s, with learning algorithms applied to extract biomarkers of movement, but it is the first to be done with laughter. A biological response with great advantages, says San Segundo: it is easy to record and it is a natural act, little controlled, which “can give interesting clues”. “Laughter that, depending on how it sounds, can give very unpleasant news, but the sooner you know it, the sooner you can be treated and you can have less degradation. In these things you have to take the bull by the horns and the sooner you know, the better “, says the professor of the Polytechnic University of Madrid.