Modeling Aphasia Type in Greek Speaking Patients Utilizing Deep Neural Networks

Chitos, Andreas (2025) Modeling Aphasia Type in Greek Speaking Patients Utilizing Deep Neural Networks. Bachelor's Thesis, Artificial Intelligence.

	Text BachelorProjectFinal.pdf Restricted to Registered users only Download (1MB)
Preview	Text akkoord chitos.pdf Download (348kB) | Preview

Abstract

Aphasia is a language disorder resulting from damage to specific brain regions responsible for language processing (Beharelle et al., 2010), most commonly within the left hemisphere. Accurate diagnosis and classification of aphasia subtypes present significant challenges (’Lazar & Boehme (2017)) due to the variability in individual symptomatology and the number of parameters influencing each patient’s linguistic profile. In this study, a deep neural network (DNN) was developed to classify transcribed speech data from Greek native speakers diagnosed with aphasia, specifically distinguishing between Broca’s aphasia, characterized by non-fluent speech and relatively preserved comprehension, and Wernicke’s aphasia, which is marked by fluent but often nonsensical speech and impaired comprehension. The DNN was trained on 45 Greek aphasia transcripts, achieving an accuracy of 62.2%, indicating that there is potential for further development with a need for a larger dataset. This approach aims to address the complexities inherent in aphasia classification by leveraging machine learning techniques to identify distinguishing linguistic patterns within the Greek speaking population.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Scheffer, S.D. and Tashu, T.M.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	13 Aug 2025 07:34
Last Modified:	26 Aug 2025 13:29
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/36666

Actions (login required)

View Item