Abstract 13861: Risk Prediction in Adults Late After Tetralogy of Fallot Repair: Does Machine Learning Provide Incremental Value Above Expert Clinical Judgement?
Bibliographic record
Abstract
Introduction and Aim: Machine learning (ML) can be used to predict major adverse cardiovascular events (MACE) in adults with repaired tetralogy of Fallot (rTOF). We sought to determine the incremental value of ML above expert clinical judgement for rTOF risk prediction. Methods: Adult congenital heart disease (ACHD) experts (≥10 years of clinical experience) participated (1 congenital heart surgeon and 4 cardiologists [2 with pediatric and 2 with adult cardiology training] with expertise in heart failure [HF], electrophysiology, imaging and intervention). Clinicians were individually asked to identify 10 high-yield clinical variables for 5-year MACE prediction (composite of mortality, resuscitated sudden death, sustained ventricular tachycardia and/or HF). Clinicians were blinded to outcomes and were asked to assign 5-year MACE risk (low, moderate, high) using 10 pre-specified variables for 25 adults with rTOF identified from an institutional database (prevalence of 5-year MACE 12%). A validated ML model was also used to predict outcomes in the same rTOF population using 10 variables. Results: Variables selected for risk prediction are shown (Figure A). As compared with the individual expert, the aggregate of 5 experts resulted in enhanced predictive capacity (Figure B). Predictive capacity of the ML model was similar to the aggregate of experts (Figures C,D). Experts with ≥20 years experience had better discriminative capacity for MACE prediction compared with <20 years (AUC 0.98 [95%CI 0.86-0.99] versus 0.80 [95%CI 0.56-0.93], p=0.03). In those with <20 years experience, ML provided incremental value such that the combined AUC approached ≥20 years (AUC 0.85 [95%CI 0.61-0.95], p=0.06). Conclusions: Prediction of 5-year MACE in rTOF using ML was similar to a multi-disciplinary team of ACHD experts. Risk prediction of less experienced clinicians was enhanced by incorporation of ML suggesting that there may be incremental value in select clinical settings.
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 | 0.010 | 0.050 |
| 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 | 0.000 |
| 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 | 0.001 |
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".