Abstract WMP23: Automated ASPECTS Scoring of CT Scans for Acute Ischemic Stroke Patients Using Machine Learning
Bibliographic record
Abstract
Objective: The Alberta Stroke Program Early CT Score (ASPECTS) method has been widely used to assess non-contrast CT scans from acute ischemic stroke (AIS) patients. Although the ASPECTS is a simple and systematic approach, ASPECTS scoring accuracy and reliability is still a challenge to clinicians, especially with limited experience. The objective of this study is to develop an automated ASPECTS scoring method, which could provide objective assessment and decision-making support. Methods: We collected 160 AIS patient NCCT images with thickness of 5mm (<8 hours from onset to scans) followed by DWI acquisition within 1 hour of NCCT. Expert ASPECTS readings on DWI with 20% thresholding (20% of a given ASPECTS region showed diffusion restriction to be scored as affected) were used as ground truth for evaluations. A NCCT template with ASPECTS regions manually contoured was non-linearly registered onto all NCCT images and ASPECTS regions were then automatically mapped onto subject NCCT images. Image features extracted from each ASPECTS region, such as regional intensity profile, neighbor context, and texture information, were used to train a random forest classifier to discriminate whether an ASPECTS region has ischemic changes. Leave-one-out validation was performed to evaluate the trained model against expert readings on DWI. Results: The proposed method generated an individual ASPECTS region level detection accuracy of 85.3% and only a 1-point discrepancy in total ASPECTS scores compared to expert reading on MRI. Bland-Altman plot of automated ASPECTS vs. expert MRI ASPECTS shows good agreement (Figure 1). The automated ASPECTS method has very high agreement (91.3%) and specificity (98.5%) when dichotomized (ASPECTS 0-4 vs. 5-10). Conclusions: The automated ASPECTS scoring approach is reliable and accurate and can potentially be used to make decisions in patients with acute ischemic stroke.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".