Abstract Details

Title Development of a novel digital speech composite measure for Frontotemporal Dementia

Topic Aging, Dementia, and Behavioral Neurology

Presentation(s) P7 - Poster Session 7 (8:00 AM-9:00 AM)

Poster/Presentation
Number 3-002

Objective

The objective for this study was to examine longitudinal change in speech patterns in Frontotemporal Dementia (FTD) and to develop a novel composite measure for assessing speech and language abilities.

Background Current tools to assess speech and language abilities in FTD often involve specialized neuropsychological testing, which can be lengthy, costly and burdensome to patients and their caregivers. Digital tools can be used to objectively measure speech abilities remotely with lower patient burden and higher functional relevance to everyday life.

Design/Methods 36 participants with subtypes of FTD were enrolled in a remote, longitudinal study collecting digital speech assessments over 12 months. Speech samples were analyzed using Winterlight’s speech analysis platform, generating >500 variables describing the acoustic and linguistic features of the speech sample. Using linear mixed models, we determined which speech variables demonstrated significant change over time and developed a novel composite score based on these variables.

Results Nine speech variables demonstrated significant change over time, controlling for effects of age, sex and stimulus. These nine selected variables were combined to create a novel speech composite score, reflecting different aspects of speech and language including the ratio of words to pauses, the types of words used, the information content and the complexity of sentences. When compared to healthy control participants, the FTD group had lower scores at baseline on the novel composite score. The composite score showed significant decline over time (β = -0.055, p < 0.001) and high test-retest reliability (ICC = 0.76, p < 0.001).

Conclusions This study demonstrated that digital speech assessments can be used to characterize speech and language abilities in FTD. We developed a novel composite measure sensitive to disease-related differences and progression. Future work will further validate this score using clinical standards and biomarkers, and integrate these measures into clinical trials.

Authors/Disclosures
Jessica Robin, PhD (Winterlight Labs) PRESENTER	Dr. Robin has received personal compensation for serving as an employee of Winterlight Labs.
	No disclosure on file
	No disclosure on file
	No disclosure on file
Stella McCaughey, PhD	Dr. McCaughey has received personal compensation for serving as an employee of Alector, Inc. Dr. McCaughey has received stock or an ownership interest from Alector, Inc.
	No disclosure on file
	No disclosure on file
Michael Ward	Michael Ward has received personal compensation for serving as an employee of Alector. Michael Ward has stock in Alector. Michael Ward has received intellectual property interests from a discovery or technology relating to health care. Michael Ward has a non-compensated relationship as a Reviewer with ADDF that is relevant to AAN interests or activities.

��ɫ��

��ɫ��