Alzheimer’s disease diagnosis using a multimodal approach with 3D MRI and PET

Βοζινάκη, Ανθή-Μαρία; Vozinaki, Anthi-Maria

dc.contributor.author	Βοζινάκη, Ανθή-Μαρία	el
dc.contributor.author	Vozinaki, Anthi-Maria	en
dc.date.accessioned	2025-07-30T07:28:40Z
dc.date.available	2025-07-30T07:28:40Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/62223
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.29919
dc.rights	Default License
dc.subject	Alzheimer’s Disease	en
dc.subject	Multimodal	en
dc.subject	Neuroimaging data	en
dc.subject	Convolutional Neural Networks	en
dc.subject	Mixture of Experts	en
dc.title	Alzheimer’s disease diagnosis using a multimodal approach with 3D MRI and PET	en
heal.type	bachelorThesis
heal.classification	Μηχανική Μάθηση	el
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2025-02-24
heal.abstract	Alzheimer’s disease is an irreversible brain disease that severely damages human thinking and is the seventh leading cause of death worldwide. Early diagnosis plays an important part es- pecially at the Mild Cognitive Impairment stage, where timely intervention can help slow its progression before it advances to AD. Neuroimaging data, like MRI and PET scans, can help detect brain changes early by providing structural and functional brain changes related to the disease. However, despite the availability of various imaging modalities for the same patient, the development of multi-modal models leveraging these modalities remains underexplored. This thesis aims to address this gap by proposing and evaluating classification models using 3D MRI and amyloid PET scans in a multimodal framework. We first employ a 3D Convolutional Neural Network, followed by three fusion techniques: feature concatenation, Gated Multimodal Unit, and Gated Self-Attention. To further improve classification performance and computational efficiency, we integrate a Mixture of Experts model, which dynamically selects the most relevant subnetworks for each prediction. Finally, we utilize Grad-CAM to visualize disease-related regions, ensuring model interpretability. The results show that the GMU-based model achieves 95.47% accuracy and specificity of 96.73% in the NC vs. AD classification task, outperforming state-of-the-art approaches. Additionally, the model successfully locates disease-related regions in both MRI and PET scans, with different activation patterns in each modality, according to Grad-CAM analysis. This result supports the effectiveness of a multimodal strategy in the diagnosis of AD by confirming the complementary nature of MRI and PET.	en
heal.advisorName	Askounis, Dimitrios	en
heal.committeeMemberName	Marinakis, Vangelis	en
heal.committeeMemberName	Psarras, John	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών. Τομέας Ηλεκτρικών Βιομηχανικών Διατάξεων και Συστημάτων Αποφάσεων	el
heal.academicPublisherID	ntua
heal.numberOfPages	96 σ.	el
heal.fullTextAvailability	false