Automated detection of Alzheimer’s disease: a multi-modal approach with 3D MRI and amyloid PET

Автоматическое выявление болезни Альцгеймера: мультимодальный подход с использованием трехмерной МРТ и амилоидной ПЭТ
Giovanna Castellano, Andrea Esposito, Eufemia Lella, Graziano Montanaro, Gennaro Vessio
2024-03-03

3D MRIAlzheimer’s disease detectionOASIS-3 cohortamyloid PETmultimodal deep learning
Recent advances in deep learning and imaging technologies have revolutionized automated medical image analysis, especially in diagnosing Alzheimer's disease through neuroimaging. Despite the availability of various imaging modalities for the same patient, the development of multi-modal models leveraging these modalities remains underexplored. This paper addresses this gap by proposing and evaluating classification models using 2D and 3D MRI images and amyloid PET scans in uni-modal and multi-modal frameworks. Our findings demonstrate that models using volumetric data learn more effective representations than those using only 2D images. Furthermore, integrating multiple modalities enhances model performance over single-modality approaches significantly. We achieved state-of-the-art performance on the OASIS-3 cohort. Additionally, explainability analyses with Grad-CAM indicate that our model focuses on crucial AD-related regions for its predictions, underscoring its potential to aid in understanding the disease's causes.
1
Combining MRI and amyloid PET significantly improves classification performance compared with single-modality models.
2
Grad-CAM analyses show that the model focuses on brain regions relevant to Alzheimer’s disease, supporting its interpretability and potential biological insight.
3
The proposed approach achieves state-of-the-art performance on the OASIS-3 cohort.
4
The study evaluates Alzheimer’s disease classification using 2D and 3D MRI and amyloid PET in unimodal and multimodal frameworks.
5
Volumetric 3D imaging data produce more effective learned representations than 2D images alone.

Alzheimer’s disease as represented by 2D/3D MRI images and amyloid PET scans

The effectiveness and explainability of uni-modal versus multi-modal volumetric imaging models for Alzheimer’s disease classification

Publication Details
Publication Date
2024-03-03
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Giovanna Castellano
Andrea Esposito
Eufemia Lella
Graziano Montanaro
Gennaro Vessio
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%