A wall tracking method to estimate ejection fraction from the parasternal long axis view in point of care ultrasound
Bibliographic record
Abstract
The left ventricular ejection fraction is a key metric for evaluating the systolic function in critically ill patients. Traditionally, it is computed using apical 2- and 4-chamber views in an echocardiogram; however, obtaining these views in an acutely ill patient in the emergency department is often difficult. A parasternal long-axis view, acquired with point of care ultrasound, is a faster and easier alternative. Unfortunately, the methods for estimating the ejection fraction from this view underperform when the left ventricular wall movement is not uniform or when its shape is not properly modeled as an ellipsoid. We propose a novel method that tracks the movement of the visible portions of the walls during the full cardiac cycle, and then estimates the ejection fraction based on that movement. We compared the performance of this method with the ejection fraction from the cardiology report on a dataset of 613 patients. Our experiments showed an accuracy of 85% for identifying critically low values for ejection fraction (EF < 30%) and 87% for abnormal ones (EF < 50%). These values are comparable with the results obtained from the apical views and superior to current methods for the parasternal long-axis view. Since our method is fully automated, we expect that it can be adopted at scale in real-world clinical scenarios, giving practitioners a new tool to properly estimate the ejection fraction in clinically challenging scenarios. • We propose a novel method to estimate the LVEF from the PLAX view. • This method provides answers in real time, even if the wall movement is not uniform. • Its performance is comparable with the one obtained using apical views.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 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 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".