Automated characterisation of cerebral microbleeds using their size and spatial distribution on brain MRI
Bibliographic record
Abstract
Cerebral microbleeds (CMBs) are small, hypointense hemosiderin deposits in the brain measuring 2-10 mm in diameter. As one of the important biomarkers of small vessel disease, they have been associated with various neurodegenerative and cerebrovascular diseases. Hence, automated detection, and subsequent extraction of clinically useful metrics (e.g., size and spatial distribution) from CMBs are essential for investigating their clinical impact, especially in large-scale studies. While some work has been done for CMB segmentation, extraction of clinically relevant information is not yet explored. Herein, we propose the first automated method to characterise CMBs using their size and spatial distribution, i.e., CMB count in three regions (and their substructures) used in Microbleed Anatomical Rating Scale (MARS): infratentorial, deep, and lobar. Our method uses structural atlases of the brain for determining individual regions. On an intracerebral haemorrhage study dataset, we achieved a mean absolute error of 2.5 mm for size estimation and an overall accuracy > 90% for automated rating. The code and the atlas of MARS regions in Montreal Neurological Institute-MNI space are publicly available. RELEVANCE STATEMENT: Our method to automatically characterise cerebral microbleeds (size and location) showed a mean absolute error of 2.5 mm for size estimation and an over 90% accuracy for rating of infratentorial, deep and lobar regions. This is a promising approach to automatically provide clinically relevant cerebral microbleeds metrics. KEY POINTS: We present a method to automatically characterise cerebral microbleeds according to size and location. The method achieved a mean absolute error of 2.5 mm for size estimation. Automated rating for infratentorial, deep, and lobar regions achieved an over 90% overall accuracy. We made the code and atlas of Microbleed Anatomical Rating Scale regions publicly available.
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.
How this classification was reachedexpand
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 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".