Attention Blocks Improve White Matter Hyperintensity Semantic Segmentation using U-Nets

Topic: Brain proteins, microglia and lab models

White matter hyperintensities (WMHs) are a common finding on magnetic resonance (MR) images in older individuals, appearing as high-signal intensity regions on fluid-attenuated inversion recovery (FLAIR) imaging. People with high WMH volume are at increased risk for dementia and stroke, controlling for vascular risk factors, but WMH burden is not reliably assessed in clinical practice. Manual segmentation of WMHs is accepted as the gold standard (or ground truth), however, it is a laborious and time-consuming method. Newer machine learning (ML)-based approaches are being proposed as alternatives to manual segmentation. Among these approaches, U-Net convolutional neural networks have demonstrated good WMH segmentation performance. However, even state-of-the-art ML models sometimes fail to correctly identify WMHs and their boundaries with sufficient accuracy. Attention blocks have emerged as a potential solution for improving the performance of U-Net models by enhancing the ability of the model to focus on relevant features in the data. We investigated the effectiveness of attention blocks in U-Net models for WMH segmentation compared to three other models (U-Net++, U-Net3+, and a standard U-Net). Attention blocks significantly improved the F-measure score for WMH segmentation (0.811 vs 0.789 for next best model, p=0.04) in a diverse brain imaging dataset. This study demonstrates that attention blocks enhance U-Net models used for WMH identification and classification.

Key findings

• People with high WMH volume are at increased risk for dementia and stroke, controlling for vascular risk factors, but WMH burden is not reliably assessed in clinical practice.
• Among these approaches, U-Net convolutional neural networks have demonstrated good WMH segmentation performance.
• We investigated the effectiveness of attention blocks in U-Net models for WMH segmentation compared to three other models (U-Net++, U-Net3+, and a standard U-Net).
• Attention blocks significantly improved the F-measure score for WMH segmentation (0.811 vs 0.789 for next best model, p=0.04) in a diverse brain imaging dataset.

Why it matters

This study tested a machine learning method for measuring white matter hyperintensities, a marker linked to higher dementia and stroke risk, and found that adding attention blocks improved segmentation accuracy. Better measurement tools may help research and clinical assessment of this risk marker, but the study does not test any treatment or prevention for Alzheimer's disease.

Study details

• Evidence strengthLow

Read the original

• DOI: 10.1109/tla.2025.11072499
• Original source →

May need a subscription OpenAlex found no free copy of the full text (checked 6 Feb 1020), so the publisher may charge for it.

@article{duarte2025-e066ebde,
  author  = {Duarte, Kauê Tartarotti Nepomuceno and de Barros, Murilo Costa and Sidhu, Abhijot Singh and Gobbi, David G. and McCreary, Cheryl R. and Saad, Feryal and Camicioli, Richard M. and Smith, Eric Edward and others},
  title   = {{Attention Blocks Improve White Matter Hyperintensity Semantic Segmentation using U-Nets}},
  journal = {IEEE Latin America Transactions},
  year    = {2025},
  doi     = {10.1109/tla.2025.11072499},
  url     = {https://doi.org/10.1109/tla.2025.11072499}
}

Related studies in Vascular contributions

• Harmonization Remains Unresolved: The Case of White Matter Hyperintensity Using ComBat And Deming Regression Alzheimer s & Dementia · 2025-12-01 · usefulness score 40/100
• WMH ‐ DualTasker : A Weakly Supervised Deep Learning Model for Automated White Matter Hyperintensities Segmentation and Visual Rating Prediction Human Brain Mapping · 2025-04-15 · usefulness score 48/100
• White Matter Hyperintensities in Behavioral Variant Frontotemporal Dementia and Semantic Variant Primary Progressive Aphasia Neurology Open Access · 2026-01-23 · usefulness score 43/100
• Investigating links between white matter hyperintensities and menopausal status using robust age-correction methods in UK Biobank Brain Communications · 2025-01-01 · usefulness score 49/100
• Characterizing spatiotemporal white matter hyperintensity pathophysiology in vivo to disentangle vascular and neurodegenerative contributions Nature Communications · 2026-03-31 · usefulness score 58/100
• Insomnia, early and late rising are associated with small hippocampal volume and large white matter hyperintensity burden Alzheimer s Research & Therapy · 2025-04-05 · usefulness score 57/100

Also tagged “cerebral”

900 studies in the library carry the topic tag “cerebral”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 900 tagged studies.

• Plasma clusterin concentration is associated with longitudinal brain atrophy in mild cognitive impairment NeuroImage · 2011-07-28 · usefulness score 64/100 · Mild cognitive impairment
• Cerebrospinal fluid YKL‐40 relates to white matter hyperintensity progression in females but not males over a 6‐year period Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring · 2025-04-01 · usefulness score 43/100 · Vascular contributions
• Domain-specific versus generalized cognitive screening in acute stroke Journal of neurology · 2015-11-20 · usefulness score 61/100 · Cognitive tests and assessment

Similar studies

• Models to Identify Small Brain White Matter Hyperintensity Lesions 2025 · Vascular contributions
• Predicting White Matter Hyperintensity: Leveraging Portable MRI for Accessible Brain Health Screening 2025 · Vascular contributions
• Association of Data-Driven White Matter Hyperintensity Spatial Signatures With Distinct Cerebral Small Vessel Disease Etiologies 2022 · Biomarkers
• A fully automated method for quantifying and localizing white matter hyperintensities on MR images 2006 · Caregiving
• Neuropathologic basis of white matter hyperintensity accumulation with advanced age 2013 · Brain scans and imaging
• White matter hyperintensity severity modifies gut metabolite association with cognitive outcomes 2025 · Vascular contributions

More on Vascular contributions

• A review that covers this ground Anterior temporal involvement in semantic word retrieval: voxel-based lesion-symptom mapping evidence from aphasia Brain : a journal of neurology · 2009-12-01 · usefulness score 74/100
• Newest in this subject Precision Nanocrystal Theranostics for Neurodegenerative Proteopathies: Strategies to Cross the Blood-Brain Barrier and Target Protein Aggregation Preprints.org · 2026-10-26 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Depth-Dependent Alterations in Cerebral Blood and Tissue Oxygenation Associated with Tau in Alzheimer’s Disease Models bioRxiv · 2026-10-16 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Vascular contributions →