Abstract TP65: An Automated Algorithm For The Nih Stroke Scale Assessment
Bibliographic record
Abstract
The NIH Stroke Scale (NIHSS) is widely adopted in clinical practice. Despite being originally designed for research use, the NIHSS is a valued resource for communication and prognostication, and it is useful for the decision-making process regarding reperfusion therapies and prophylaxis. However, its assessment can be laborious and complex among even certified healthcare providers. In the context of increasing telemedicine use, an accurate assessment of the NIHSS may be crucial in acute stroke management We aimed to create and validate an automated tool for the NIHSS (SPOKES) in a national telemedicine service. A board of five certified vascular neurologists created an NIHSS algorithm based on a tree decision, including tips and hints in the main questions and auxiliary boxes. We randomized 22 spoke hospitals using an automated tool to invite emergency physicians not certified in the NIHSS to use or not the SPOKES. NIHSS-certified and blinded neurologists from a hub hospital performed a double-check of each item of the NIHSS. From June to August 2022, we included 144 cases from 10 spoke hospitals. Our algorithm was fully adopted in 27 cases (19%). The median of reported NIHSS was 3 [1, 5] and 3 [2, 7] points among users and non-users, p=0.38. The general difference between the reported and the double-checked score was 0 [0, 1] points – there was no difference between those who used or did not the SPOKES (p=0.12). A complete concordant score was achieved at 66.7% (n=18/27) and 45.3% (n=53/117), χ 2 =0.036, among users and non-users, respectively. In a bivariate regression analysis, the SPOKES increased the chance of complete agreement [OR 2.4, 95%CI 1-5.8, p=0.049]. There was no difference regarding discrepant scores (≥4 points), χ 2 =0.46. Among SPOKES cases, treatment with tPA was indicated in 11.1%, versus 12.7% among non-users (p=0.59). Despite the small number of included cases, our algorithm seems to be a promising tool for the NIHSS assessment in a national telemedicine service, increasing the chance of a complete agreement with certified neurologists. The tool is free and available at www.spokes-nihss.com
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 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| 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) | 0.008 | 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".