Optimal Filter Design for Multiple Sclerosis Lesions Segmentation from Regions of Interest in Brain MRI
Bibliographic record
Abstract
In this paper, we propose an optimal filter design strategy for the purpose of detecting and segmenting MS lesions in prescribed regions of interest within brain MRI data. Reliable segmentation of multiple sclerosis lesions in magnetic resonance brain imaging is important for at least three types of practical applications: pharmaceutical trials, decision making for drug treatment or surgery, and patient follow-up. Manual segmentation of the MS lesions in brain MRI by well qualified experts is usually preferred. However, manual segmentation is hard to reproduce and can be time consuming in the presence of large volumes of MRI data. On the other hand, automated segmentation methods are significantly faster and yield reproducible results. However, these methods generally produce segmentation results that agree only partially with the ground truth segmentation provided by the expert. In this work, we propose a semi-automated MS lesion detection system that combines the knowledge of the expert with the computational capacity to produce faster and more reliable MS segmentation results. In particular, the user selects coarse regions of interest (ROI's) that enclose potential MS lesions and a sufficient background of healthy white matter tissues. Having this two-class classification problem, we propose a feature extraction method based on optimal filter design that aim for producing output texture features corresponding to the MS lesions and healthy tissues background which are maximally separable. If this is achieved, the two output features may be easily separated using simple thresholding operations. The method is applied on real MRI data and the results are qualitatively compared to a ground truth, which is manually segmented by a human expert
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".