Development of Predictive Models for Continuous Flow Left Ventricular Assist Device Patients using Bayesian Networks
Bibliographic record
Abstract
Background: Existing prognostic tools for patient selection for ventricular assist devices (VADs) such as the Destination Therapy Risk Score (DTRS) and newly published HeartMate II Risk Score (HMRS) have limited predictive ability, especially with the current generation of continuous flow VADs (cfVADs). This study aims to use a modern machine learning approach, employing Bayesian Networks (BNs), which overcomes some of the limitations of traditional statistical methods. Methods: Retrospective data from 144 patients at Allegheny General Hospital and Integris Health System from 2007 to 2011 were analyzed. 43 data elements were grouped into four sets: demographics, laboratory tests, hemodynamics, and medications. Patients were stratified by survival at 90 days post LVAD. Results: The independent variables were ranked based on their predictive power and reduced to an optimal set of 10: hematocrit, aspartate aminotransferase, age, heart rate, transpulmonary gradient, mean pulmonary artery pressure, use of diuretics, platelet count, blood urea nitrogen and hemoglobin. Two BNs, Naïve Bayes (NB) and Tree-Augmented Naïve Bayes (TAN) outperformed the DTRS in identifying low risk patients (specificity: 91% and 93% vs. 78%) and outperformed HMRS predictions of high risk patients (sensitivity: 80% and 60% vs. 25%). Both models were more accurate than DTRS and HMRS (90% vs. 73% and 84%), Kappa (NB: 0.56 TAN: 0.48, DTRS: 0.14, HMRS: 0.22), and AUC (NB: 80%, TAN: 84%, DTRS: 59%, HMRS: 59%). Conclusion: The Bayesian Network models developed in this study consistently outperformed the DTRS and HMRS on all metrics. An added advantage is their intuitive graphical structure that closely mimics natural reasoning patterns. This warrants further investigation with an expanded patient cohort, and inclusion of adverse event outcomes.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".