Predicting Acute Ischemic Stroke Core in Multiphase CT Angiography Using a CT Perfusion-Trained Neural Network
Bibliographic record
Abstract
Early and accurate identification of tissue infarction (stroke core) is crucial for treatment decisions in acute ischemic stroke. While CT angiography (CTA) is widely used and accessible, it lacks standardized methods for core prediction. CT perfusion (CTP) offers improved tissue classification but is limited by higher cost and complexity. This study aims to bridge this gap by developing a convolutional neural network (CNN) trained on CTP core maps to differentiate ischemic core from healthy tissue in multiphase CTA (mCTA) images. We retrospectively analyzed acute ischemic stroke patients who underwent both CTP and mCTA (n=99). Two additional datasets with increasing stroke severity - Severe15 and Severe20 - were derived from the base dataset to assess model performance across a range of clinical scenarios (n=58 and n=31, respectively). The CNN was trained using CTP-derived core maps as ground truth and applied to mCTA-based feature maps. Model performance was optimized using cross-entropy loss and evaluated with thresholding analysis. The primary metric was the Dice similarity coefficient (DSC), supplemented by various thresholded accuracy metrics and threshold-independent precision-recall curve analysis. The model trained on the base dataset demonstrated reasonable performance in distinguishing ischemic stroke core from healthy tissue, achieving a DSC of 0.43. As stroke severity increased, thresholded accuracy metrics improved or remained consistent, with the Severe20 model achieving a DSC of 0.60. Precision-recall curve analysis further highlighted the tradeoff between DSC and overall model precision. This study demonstrates the potential of leveraging machine learning to enhance stroke imaging accessibility by integrating the core detection capabilities of CTP with the speed and availability of CTA. Future work will focus on optimizing performance for clinical translation and expanding validation to larger datasets.Clinical Relevance-- The model developed in this study provides a starting point for a reliable, data-driven approach to stroke imaging that can enhance equitable access to advanced diagnostics and support more informed clinical decision-making, especially in severe stroke cases.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".