Detecting Parkinson's Disease Through Speech Analysis

Automatic Acoustic Speech Analysis and REM Sleep Behaviour Disorder for Detecting Subjects at High Risk for Parkinson's Disease and Other Alpha-synucleinopathies

General University Hospital, Prague, University Hospital, Montpellier, Mayo Clinic, McGill University Health Centre/Research Institute of the McGill University Health Centre, Medical University Innsbruck, San Raffaele University Hospital, Italy, Philipps University Marburg Medical Center

Speech is an important indicator of motor function and movement coordination and can be extremely sensitive to involvement in the course of neurologic diseases. The aim of this project is to discover for the first time using simple speech recording and high end pattern analysis preclinical stages of disabling central nervous system disorders including Parkinson's disease and other alpha-synucleinopathies in "at high risk" patients with REM sleep behavior disorder and thus provide one essential prerequisite for trials on REM sleep behavior disorder with preventive therapy.

Seven centers of excellence in sleep research will investigate speech and other clinical symptoms in more than 100 subjects with REM sleep behavior disorder. Analyses of a number of unique speech dimensions based upon three fundamental categories of simple speaking tasks will be used to search for specific prodromal alterations in speech patterns in REM sleep behavior disorder, compared to age- and gender-matched patients with early Parkinson's disease and healthy control subjects. Robust algorithms allowing automated speech analysis will be developed and optimized through English, German, French, Czech and Italian languages. Early motor dysfunction strongly predicts Parkinson's disease and other alpha-synucleinopathies. In this regard, vocal assessment has intriguing potential advances as is non-invasive, inexpensive, simple to administer and scalable to large population with possibility to perform recordings remotely, even by telephone from patients' home. Speech analysis may serve as a simple tool to screen large populations for the risk to develop Parkinson's disease. If speech impairment appears to be a strong biomarker of early motor dysfunction, the screening of speech changes may improve stratification for future neuroprotective therapies for Parkinson's disease and other synucleinopathies.

Each subject will undergo short non-invasive speech assessment lasting approximately 15 minutes that will be recorded using microphone.

Volunteers who enroll in the study will undergo various assessments lasting about 2 hours, including neurological examination of both motor and non-motor skills, autonomic testing and cognitive testing.

Quantitative acoustic assessment of several deviant speech dimensions connected with phonatory, articulatory, and prosodic abnormalities in hypokinetic dysarthria of Parkinson's disease.

The most commonly used scale for the clinical evaluation of Parkinson's disease assessing variety of motor and non-motor symptoms.

Inclusion Criteria for subjects with Parkinson's disease (PD): Diagnosed based on the MDS Clinical Diagnostic Criteria for Parkinson's Disease including the presence of bradykinesia in combination with either rest tremor, rigidity, or both; Hoehn & Yahr stage 1-2 in the defined OFF state; Disease duration from diagnosis < 5 years; No motor fluctuations or dyskinesias; On stable dose of medication in last four weeks; Onset of PD after 50 years; No history of communication or neurological disorders unrelated to PD; Not currently involved in any speech therapy; Inclusion Criteria for subjects with REM Sleep Behaviour Disorder (RBD): Fulfill criteria for RBD based on polysomnography according to the standard International Classification of Sleep Disorders diagnostic criteria, 3rd edition; Onset of RBD after 50 years; No history of communication or neurological disorders. Inclusion Criteria for healthy control subjects: No history of communication or neurological disorders, parasomnias and other sleep disorders with important severity; No regular benzodiazepines, hypnotics and melatonin intake. Exclusion criteria: (for all the arms) Inability to speak and read a text aloud.

De-identified clinical data (speech recordings and associated clinical scales) may be shared with the Michael J. Fox Foundation (the study funder). This data may be kept for storage at a central repository either hosted by the Michael J. Fox Foundation, its collaborators, or consultants.

Rusz J, Hlavnicka J, Tykalova T, Buskova J, Ulmanova O, Ruzicka E, Sonka K. Quantitative assessment of motor speech abnormalities in idiopathic rapid eye movement sleep behaviour disorder. Sleep Med. 2016 Mar;19:141-7. doi: 10.1016/j.sleep.2015.07.030. Epub 2015 Sep 14.

Postuma RB, Lang AE, Gagnon JF, Pelletier A, Montplaisir JY. How does parkinsonism start? Prodromal parkinsonism motor changes in idiopathic REM sleep behaviour disorder. Brain. 2012 Jun;135(Pt 6):1860-70. doi: 10.1093/brain/aws093. Epub 2012 May 4.