Attention Blocks Improve White Matter Hyperintensity Semantic Segmentation using U-Nets
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
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.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it