Expert Evaluation of Deep Learning Approaches to White Matter Hyperintensity Segmentation in Older Adults
This study highlights significant variability in the quality of WMH segmentation across commonly used deep learning pipelines when benchmarked against expert evaluation. Among the evaluated pipelines, ANTSx demonstrated superior performance, producing clinically plausible segmentations with high anatomical fidelity. These findings underscore the importance of expert validation in selecting and refining automated segmentation tools for research and clinical applications in aging and neurodegenerative disease.
Authors
- Amanda Cook Maher (ORCID: https://orcid.org/0000-0001-6229-4743)
- Adam Martersteck (ORCID: https://orcid.org/0000-0001-6997-9390)
- Siobhan McDermott (ORCID: https://orcid.org/0000-0002-7263-653X)
- Emily Rogalskı (ORCID: https://orcid.org/0000-0002-6472-1363)
- Rhiana Schafer (ORCID: https://orcid.org/0000-0003-0328-2499)
- Elizabeth Finger (ORCID: https://orcid.org/0000-0003-4461-7427)
- Maria Kharitonova (ORCID: https://orcid.org/0000-0002-8831-3321)
- Karthik Sreenivasan
- C W VanDyke
- Angela C. Roberts
- Sophia Moore
- Ozioma C. Okonkwo
- Felicia C. Goldstein
Institutions
- Western University (CA)
- Alzheimer's Association (US)
- University of Wisconsin–Madison (US)
- Emory University (US)
- University of Michigan (US)
- University of Chicago (US)
Publication Details
- Journal
- Alzheimer s & Dementia
- Published
- 2025-12-01
- DOI
- https://doi.org/10.1002/alz70862_110016
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00