Key result
Extreme gradient boosting demonstrated the highest predictive ability for post-operative infections, whereas K-nearest neighbor had the lowest performance.
Why the study?
Machine learning models are increasingly used in surgical decision-making algorithms, but given their distinct functioning, not all models may be suitable for acutely ill surgical patients.
Which machine learning model best predicts post-operative infections in urgent and emergent surgical patients?
Observational (n=624,625)
Yes
Which machine learning model best predicts post-operative infections in urgent and emergent surgical patients?
Different machine learning models vary significantly in their ability to predict post-operative infections, with extreme gradient boosting outperforming others like K-nearest neighbor.
Answers are grounded in this paper and cite their sources.
Model choice affects SSI prediction accuracy in NSQIP data; leaves open which algorithms warrant prospective validation before clinical use.
Background:Surgical quality datasets are critical to decision-making tools including surgical infection (SI). Machine learning models (MLMs), a branch of artificial intelligence (AI), are increasingly being ingrained within surgical decision-making algorithms. However, given the unique and distinct functioning of individual models, not all models may be suitable for acutely ill surgical patients. Patients and Methods:This is a 5-year retrospective review of National Surgical Quality Improvement Program (NSQIP) patients who underwent an operation. The data were reviewed for demographics, medical comorbidities, rates, and sites of infection. To generate the MLMs, data were imported into Python, and four common MLMs, extreme gradient boosting, K-nearest neighbor (KNN), random forest, and logistic regression, as well as two novel models (flexible discriminant analysis and generalized additive model) and ensemble modeling, were generated to predict post-operative SIs. Outputs included area under the receiver-operating characteristic curve (AUC ROC) including recall curves. Results:Overall, 624,625 urgent and emergent NSQIP patients were included. The overall infection rate was 8.6%. Patients who sustained a post-operative infection were older, more likely geriatric, male, diabetic, had chronic obstructive pulmonary disease, were smokers, and were less likely White race. With respect to MLMs, all four MLMs had reasonable accuracy. However, a hierarchy of MLMs was noted with predictive abilities (XGB AUC = 0.85 and logistic regression = 0.82), wherein KNN has the lowest performance (AUC = 0.62). With respect to the ability to detect an infection, precision recall of XGB performed well (AUC = 0.73), whereas KNN performed poorly (AUC = 0.16). Conclusions:MLMs are not created nor function similarly. We identified differences with MLMs to predict post-operative infections in surgical patients. Before MLMs are incorporated into surgical decision making, it is critical that surgeons are at the fore of understanding the role and functioning of MLMs.
No takes yet. Share an insight, caveat, or question.
Ganguli et al. (2025) conducted an observational in Urgent and emergent surgical patients (n=624,625). Machine learning models (e.g., Extreme gradient boosting) vs. Other machine learning models (e.g., K-nearest neighbor) was evaluated on Prediction of post-operative surgical infections (AUC ROC) (AUC 0.85). Extreme gradient boosting demonstrated the highest predictive ability for post-operative infections, whereas K-nearest neighbor had the lowest performance.
Synapse has enriched 10 related papers. Consider them for comparative context: