Vascular plaque detection from optical coherence tomography images
Bibliographic record
Abstract
It is difficult to detect atherosclerotic plaque from optical coherence tomography (OCT) images via visual inspection. In this work, we developed three algorithms to allow us to detect atherosclerotic plaque more effectively: (i) a statistical method that uses higher-order moments; (ii) a model-based method that enables vascular plaque to be automatically identified based on the textural features in OCT images; (iii) and a sparsity-based segmentation algorithm in the curvelet domain. All three algorithms do not rely on visual inspection at all. The statistical method consists of three main components: extracting statistical image textural features using the Spatial Gray Level Dependence Matrix (SGLDM) method; applying an unsupervised Fuzzy C-means clustering algorithm to these features; and, finally, mapping specific clustered regions—namely, background, plaque, vascular tissue, and the deep-depth degraded signal in feature-space—back to the actual image. Since the use of the full set of 26 textural features is computationally expensive and may not be practical for real-time implementation, we identified a reduced set of 6 textural features, which were used to characterize vascular plaque via sparse principal component analysis. However, our clustering-based algorithm results had some limitations, most notably non-smooth and coarse segmentation results. To overcome this low spatial resolution limitation, we developed a stochastic model to segment OCT images of vascular tissue into plaque and non-plaque (i.e., healthy tissue) regions, as well as background regions. Our stochastic model is based on a maximum a posteriori-Markov Random Field (MRF-MAP) framework wherein OCT images of vascular tissue were modeled as a Markov random field. This MRF-MAP-based algorithm yielded results with better spatial resolution, but it is not consistent and also computationally expensive, thereby impractical for real-time implementation. Our third approach, using a sparsity-based segmentation algorithm in the curvelet domain, overcame the two limitations above by generating both fast and high-resolution vascular plaque detection from OCT images. We verified the validity of the results of all three methods using both qualitative and quantitative methods. Specifically, all results were compared with 1) actual photographic images of vascular tissue samples, 2) histology results, and 3) ground truth obtained from manual segmentations performed by four cardiovascular surgeons from the Intervention Cardiology Group at St. Boniface Hospital, Winnipeg, Manitoba. These comparisons of results demonstrated that our three methods allow good plaque detection, thus making them potential clinical tools for the detection of vascular plaque from OCT images and for clinical studies involving OCT imaging of vascular plaque. .
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 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 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".