A quick, speech-based AI tool offers a new way to screen for a key indicator of the neurodegenerative disease.
Computer scientists at the have developed an AI-powered, speech-based screening tool that can help people assess whether they are showing signs of Parkinson糖心传媒檚 disease, the fastest growing neurological disability in the world. A published in the journal npj Parkinson糖心传媒檚 Disease introduces a web-based screening test that asks users to recite two pangrams糖心传媒攕hort sentences using all 26 letters of the alphabet. Within seconds, the AI analyzes the voice recordings for subtle patterns linked to Parkinson糖心传媒檚, with nearly 86 percent accuracy.
Parkinson糖心传媒檚 disease is typically diagnosed by movement disorder specialists糖心传媒攏eurologists with specific training to evaluate complex motor symptoms糖心传媒攗sing a combination of family history, neurological examinations, and brain imaging. While the study糖心传媒檚 authors emphasize that their AI-based tool is not a substitute for a clinical diagnosis, they see it as a fast, low-barrier, and accessible way to flag people, especially in remote areas, who might be living with the condition and encourage them to seek more thorough clinical evaluations.
糖心传媒淭here are huge swaths of the US and across the globe where access to specialized neurological care is limited,糖心传媒� says , a professor in Rochester糖心传媒檚 and co-director of the . 糖心传媒淲ith users糖心传媒� consent, widely used speech-based interfaces like Amazon Alexa or Google Home could potentially help people identify if they need to seek further care.糖心传媒�
To train and validate the tool, the researchers collected data from more than 1,300 participants糖心传媒攚ith and without Parkinson糖心传媒檚糖心传媒攁cross diverse environments, including home settings, clinical visits at the , and the InMotion Parkinson糖心传媒檚 disease care center in Ohio.
Using the computer糖心传媒檚 microphone, users simply read aloud two sentences: 糖心传媒淭he quick brown fox jumps over the lazy dog. The dog wakes up and follows the fox into the forest, but again the quick brown fox jumps over the lazy dog.糖心传媒� By leveraging the power of advanced semi-supervised speech models trained on millions of digital audio recordings to understand the characteristics of speech, the tool can glean enough vocal cues from those two short sentences to flag warning signs.
糖心传媒淭hese large audio models are trained to understand how speech works; for example, the way someone with Parkinson糖心传媒檚 would utter sounds, pause, breathe, and inadvertently add features of unintelligibility is different in someone without Parkinson糖心传媒檚,糖心传媒� says Abdelrahman Abdelkader, a student in Hoque糖心传媒檚 lab and one of the two lead authors of the study. 糖心传媒淚f a person is saying the pangram that contains the full spectrum of the alphabetical variability and trails off at certain points, the model can tell if that糖心传媒檚 different from the typical representation and flag it.糖心传媒�
The tool was 85.7 percent accurate when tested, providing a strong indication of whether someone may have Parkinson糖心传媒檚. But it is a multifaceted disease, and while some people demonstrate symptoms through speech, they can also display signs through motor tasks or . Over the last decade, has pursued and produced state-of-the-art results.
糖心传媒淩esearch shows that nearly 89 percent of people with Parkinson糖心传媒檚 have a deformity in their voice that can be indicative of the disease, making speech a strong starting point for digital screening,糖心传媒� says Tariq Adnan, a computer science PhD student affiliated with Hoque糖心传媒檚 lab and another lead author of the study. 糖心传媒淏y combining this method with assessments of other symptoms, we aim to cover the majority of people through our accessible screening process.糖心传媒�
An interactive demo of the lab糖心传媒檚 three screening tests, including the speech test outlined in the paper, is .
The other authors of the paper include PhD students Md. Saiful Islam, who co-supervised the work with Hoque, Zipei Liu, Ekram Hossain, and Sooyong Park.
The study was funded by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health, the Gordon and Betty Moore Foundation, a Google Faculty Research Award, and a Google PhD Fellowship.
