IC‐P‐168: Lddmm‐Based Robust Multi‐Template Subcortical Segmentation Using an ADNI Template Library
Bibliographic record
Abstract
Morphological studies analyzing subcortical brain changes caused by neurodegenerative dementias or aging require the delineation or segmentation of subcortical structure boundaries from the structural Magnetic Resonance Imaging (MRI) images. As manual segmentation becomes tedious and extremely time consuming beyond small image datasets, automated techniques are needed in practice for subcortical MRI segmentation. In this work, we propose a novel method for subcortical segmentation following the well established multi-template fusion framework. Our proposed method employs the popular Large Deformation Diffeomorphic Metric Mapping (LDDMM) algorithm for propagating the manual labels from the template image onto the target image. The accuracy and efficiency of this label propagation is improved by considering a tight bounding boxes around the subcortical structures based on the initial segmentations obtained using Freesurfer (FS). The most important aspect of the proposed method is the use of a robust template selection strategy to choose templates that are anatomically similar to the target image while discarding the ones that are too dissimilar before the fusion step. The robust template selection involves ranking the propagated manual template segmentations based on their Hausdorff surface distance to the initial FS segmentation. Only the top ranked propagated template segmentations are fused to obtain the desired target segmentation. We developed a manual template library consisting 74 MRI images taken from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. In each of the template images, six subcortical structures, amygdala, caudate, hippocampus, putamen, thalamus and ventricles were delineated by a manual operator. The proposed method was validated using the ADNI template library in a leave-one-out cross-validation framework. The segmentation accuracy was measured using the dice overlap measure between the manual and automated segmentations. Our method obtained excellent (80-90%) dice scores for all the six subcortical structures on an average. Further, independent validation on the benchmark Harmonized Protocol (HarP) database with “ground truth” hippocampus labels also yielded (> 80%) mean dice scores. An accurate and robust automated subcortical segmentation method has been developed. The proposed method can be used to facilitate large scale studies on analyzing morphological changes of the subcortical structures in the brain using large image databases.
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) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.005 |
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".