Save
A comparison of human and digital speech analysis to describe and monitor cerebellar dysfunction and disease severity in multiple sclerosis
ECTRIMS Online Library. Noffs G. Oct 12, 2018; 232027
Gustavo Noffs
Gustavo Noffs
Login now to access Regular content available to all registered users.

You may also access this content "anytime, anywhere" with the Free MULTILEARNING App for iOS and Android
Abstract
Discussion Forum (0)
Rate & Comment (0)

Abstract: 274

Type: Scientific Session

Abstract Category: Clinical aspects of MS - Clinical assessment tools

Background: Listener based assessments of speech in multiple sclerosis (MS) have identified time-related disturbances (e.g. slow speech rate, irregular rhythm) as key features of the disease. Objective digital analysis of speech has emerged as a promising tool to detect and monitor subclinical speech deficits. However, the relationship between speech measures with other disease severity measures in MS is not known.
Aim: Compare the sensitivity of two speech production measures: i) perceptual analysis (PA); ii) time-related acoustic analysis (AA) to predict MS-related responses: i) Expanded Disability Status Scale (EDSS) scores; ii) cerebellar function via Scale for the Assessment and Rating of Ataxia (SARA); iii) quality of life using the Multiple Sclerosis Impact Scale questionnaire (MSIS-29); iv) MRI-measured cerebellar atrophy and v) lesion load.
Methods: Speech samples from 118 people with MS were rated by two blinded speech experts (PA) and submitted to time-domain computerized acoustic analysis (AA). EDSS and SARA scores were documented for each participant; 80 participants completed the MSIS-29. 3T MRI to determine cerebellar volumes and lesion load were available in 60 participants. For each respective predictive method (PA and AA), four variables that best correlated with each of the five clinical outcomes were selected and entered into a stepwise linear regression model corrected for disease duration. Time to complete each predictive method was recorded.
Results: PA selected variables were Naturalness, Prolonged Intervals, Imprecise Consonants and Diadochokynetic Rate, which produced predictive R square adjusted (R2a) for EDSS= .548, SARA= .580 and MSIS= .259. AA selected variables were Diadochokinetic Rate, Reading Speech Rate, Free Speech Pause Percentage and Total Pause Time. AA produced predictive R2a adjusted for EDSS= .566, SARA= .597 and MSIS= .118. Neither PA nor AA predicted cerebellar atrophy or lesion load. PA ratings took 55 people-hours to complete by two trained specialists. AA took 3.5 people-hours and required access to semi-automated computer scripts and software.
Conclusion: AA performs similarly to expert perceptual analysis in relation to MS-related disability, cerebellar dysfunction and quality of life. AA has the potential to be fully automated and tested in real-life longitudinal monitoring of MS disease progression.
Disclosure: G Noffs: nothing to disclose
F Boonstra: nothing to disclose
T Perera: nothing to disclose
F Maldonado: nothing to disclose
A Vogel: consults to Takeda and Pfizer. He is Chief Science Officer of Redenlab. He receives grant and fellowship funding from the National Health and Medical Research Council of Australia.
A Evans: received honoraria from Novartis for giving presentations and providing consultancy services. He has participated in scientific advisory board meetings for Novartis, UCB Pharma, Allergan, and Boehringer Ingelheim. He has received conference travel support from Boehringer Ingelheim.
H Butzkueven: served on scientific advisory boards for Biogen, Roche, Merck, Novartis, Teva and Sanofi and has received conference travel support from Novartis, Biogen and Merck. He serves on steering committees for trials conducted by Merck, Biogen and Novartis, and has received research support from Novartis, Biogen, NHMRC Australia, MS Research Australia and the UKMS Trust.
S Kolbe: receives grant income from the National Health and Medical Research Council of Australia and has received honoraria from Novartis, Biogen and Merck
A van der Walt: received travel support from Biogen Idec, Novartis, Teva, Merck, Sanofi and served on advisory boards for Biogen Idec, Novartis and Merck. She receives grant and fellowship funding from the National Health and Medical Research Council of Australia.

Code of conduct/disclaimer available in General Terms & Conditions
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.


Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.



Google Analytics is used for user behavior tracking/reporting. Google Analytics works in parallel and independently from MLG’s features. Google Analytics relies on cookies and these cookies can be used by Google to track users across different platforms/services.


Save Settings