Carleton University
Multimodal Dementia Screening Using Gaze and Speech: Biomarker Validation and Acoustic Modeling to Support Real-World Application
Abstract
dc:description.abstractThis thesis proposes and evaluates methods using gaze and speech as multimodal features for dementia assessment. Software methods were created to process and analyze eye-tracking and speech data from standardized tasks to extract robust multimodal features. These features were shown to characterize cognitive performance for healthy controls and persons with cognitive decline, supporting their use as digital biomarkers. Classification and regression models using these features were proposed and evaluated to support cognitive assessment and monitoring. To test practicality across realistic scenarios and improve system robustness, the performance of automatic speech-to-text systems was characterized across varying acoustic environments to quantify environmental impacts on transcription accuracy. To support this analysis, rapid acoustic simulation methods were proposed using handheld LiDAR scans and ray tracing to efficiently model reverberation. Collectively, these contributions assess the performance and robustness of multimodal gaze and speech biomarkers and support a solution for scalable remote dementia screening and monitoring.
Degree
thesis:*- Name thesis:degree_name
- Master of Applied Science (M.App.Sc.)
- Level thesis:degree_level
- Master's
- Discipline thesis:degree_discipline
- Engineering, Biomedical
- Grantor dc:publisher
- Carleton University
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Laghai, Amir
Subjects
dc:subject × 2Rights
dc:rights- Statement dc:rights
-
- Copyright © 2026 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, distribution to students, research and scholarship. Theses may only be shared by linking to the Carleton University Institutional Repository and no part may be copied without proper attribution to the author; no part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
- Language dc:language.iso
- en