Key result
Machine learning models (AUC 0.709-0.712) showed negligible difference compared to traditional logistic regression (AUC 0.712) in predicting 30-day post-operative morbidity.
Why the study?
Prediction of morbidity by machine learning for laparoscopic colectomy had not been adequately analyzed or compared with traditional logistic regression models.
Do machine learning models improve the prediction of 30-day post-operative morbidity in laparoscopic colectomy patients compared to logistic regression?
Observational
Yes
Do machine learning models improve the prediction of 30-day post-operative morbidity in laparoscopic colectomy patients compared to logistic regression?
Effect estimate: AUC 0.712
Machine learning models did not outperform traditional logistic regression in predicting 30-day post-operative morbidity after laparoscopic colectomy.
BACKGROUND: Complex machine learning (ML) models have revolutionized predictions in clinical care. However, for laparoscopic colectomy (LC), prediction of morbidity by ML has not been adequately analyzed nor compared against traditional logistic regression (LR) models. METHODS: All LC patients, between 2017 and 2019, in the National Surgical Quality Improvement Program (NSQIP) were identified. A composite outcome of 17 variables defined any post-operative morbidity. Seven of the most common complications were additionally analyzed. Three ML models (Random Forests, XGBoost, and L1-L2-RFE) were compared with LR. RESULTS: Random Forests, XGBoost, and L1-L2-RFE predicted 30-day post-operative morbidity with average area under the curve (AUC): .709, .712, and .712, respectively. LR predicted morbidity with AUC = .712. Septic shock was predicted with AUC ≤ .9, by ML and LR. CONCLUSION: There was negligible difference in the predictive ability of ML and LR in post-LC morbidity prediction. Possibly, the computational power of ML cannot be realized in limited datasets.
No takes yet. Share an insight, caveat, or question.
Velmahos et al. (2023) conducted an observational in Laparoscopic colectomy. Machine learning models (Random Forests, XGBoost, L1-L2-RFE) vs. Logistic regression was evaluated on 30-day post-operative morbidity (composite outcome of 17 variables) (AUC 0.712). Machine learning models (AUC 0.709-0.712) showed negligible difference compared to traditional logistic regression (AUC 0.712) in predicting 30-day post-operative morbidity.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: