Does Brain Atlas Choice Matter? An Empirical Study in Alzheimer’s Diagnosis Using FDG-PET Images
Brain mapping is essential for diagnosing neurological conditions and can greatly influence analytical outcomes. This study evaluates the performance of various brain atlases in classifying Alzheimer’s disease (AD) and mild cognitive impairment (MCI) using fluorodeoxyglucose positron emission tomography (FDG-PET) images, focusing on the impact of atlas choice and granularity. Our analysis reveals substantial performance differences among atlases, with data-tailored atlases demonstrating notably higher effectiveness. These findings highlight the importance of choosing atlases that align with specific tasks and datasets, offering valuable insights for improving diagnostic accuracy.
Authors
- Éric Guedj (ORCID: https://orcid.org/0000-0003-1912-6132)
- Nguyen Linh Trung (ORCID: https://orcid.org/0000-0002-3103-994X)
- Mouloud Adel (ORCID: https://orcid.org/0000-0003-2840-0172)
- Phạm Minh Tuấn
Institutions
- Vietnam National University, Hanoi (VN)
- Institut Fresnel (FR)
Publication Details
- Published
- 2024-12-03
- DOI
- https://doi.org/10.1109/apsipaasc63619.2025.10848822
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- TED