Evaluation of an artificial intelligence algorithm (HeartAssistᵀᴹ) in the assessment of fetal cardiothoracic ratio: a prospective study
Bibliographic record
Abstract
Introduction: To study intra-observer and inter-observer repeatability and inter-method agreement between manual and automatic methods in assessing fetal thoracic circumference (TC), cardiac circumference (CC), and cardiac/thoracic circumference ratio (C/T).Material and methods: In a prospective study on low-risk pregnant women undergoing second-trimester ultrasonographic examination, a frame of the thoracic circumference was obtained at the level of the 4-chamber view.For each frame the TC, CC, and C/T were measured offline by 2 examiners manually and by HeartAssist TM artificial intelligence software.Intra-and inter-observer repeatability and inter-method agreement between manual and automatic methods were analysed.Results: Fifty consecutive pregnant women were considered at a median gestational age of 20.9 weeks.All intraclass correlation coefficients (ICC) comparing manual with heart assist were > 0.929.Intra-and inter-observers ICC were respectively > 0.971 and > 0.931 for all the 3 variables, representing good agreement.The time necessary to obtain the measurements was significantly lower using heart assist than the manual method (82 s vs. 22.5 s; p < 0.0001).Conclusions: Heart assist allows automatic measurement of the C/T ratio.This technique was reproducible and reached the same accuracy as that of manual measurements.Heart assist is faster than manual and has the potential to become the preferred technique to obtain cardiac biometry.
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 | 0.022 | 0.049 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".