Deep learning-based Alzheimer's disease detection: reproducibility and the effect of modeling choices
Our work underscores the critical importance of adhering to rigorous experimental practices in the field of ML applied to healthcare. The results clearly demonstrate how data augmentation and model depth-often overlooked factors- can dramatically impact final performance if not thoroughly investigated. This highlights both the necessity of exploring neglected modeling aspects and the need to comprehensively report all modeling choices to ensure reproducibility and facilitate meaningful comparisons across studies.
Authors
- Annalisa Barla (ORCID: https://orcid.org/0000-0002-3436-035X)
- Alessandro Verri (ORCID: https://orcid.org/0000-0001-9777-9986)
- Rosanna Turrisi (ORCID: https://orcid.org/0000-0002-5804-9566)
Institutions
- University of Genoa (IT)
Publication Details
- Journal
- Frontiers in Computational Neuroscience
- Published
- 2024-09-20
- DOI
- https://doi.org/10.3389/fncom.2024.1360095
- Citations
- 10
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 3.47
Funders
- Yale University
- Vanderbilt University
- University of Pennsylvania
- Ohio State University
- Emory University
- NYU Langone Medical Center
- Northwestern University
- Cleveland Clinic
- Johns Hopkins University
- University of Texas Southwestern Medical Center
- University of Pittsburgh
- Georgetown University
- University of Rochester
- Rush University
- University of South Florida
- Vanderbilt University Medical Center
- York University
- Jewish General Hospital
- University of California, Los Angeles
- University of California, San Diego
- Medical School, University of Michigan
- Wake Forest School of Medicine
- Houston Methodist Research Institute