Key result
Machine learning with self-organizing maps fails to improve in-stent restenosis prediction over conventional analysis.
Why the study?
Machine learning approaches have the potential to uncover regular patterns in multi-layered data to better predict in-stent restenosis after percutaneous coronary intervention with stenting.
Does machine learning using self-organizing maps improve the prediction of angiographic in-stent restenosis in patients following PCI compared to conventional multivariable analysis?
Cohort (n=10,004)
Does machine learning using self-organizing maps improve the prediction of angiographic in-stent restenosis in patients following PCI compared to conventional multivariable analysis?
Absolute Event Rate: 0.728% vs 0.726%
p-value: p=0.3
Although unsupervised machine learning identified additional predictive variables for in-stent restenosis, it failed to improve overall predictive accuracy compared to established regression models, underscoring that novel algorithms do not automatically yield clinically superior risk prediction.
Machine learning via self-organizing maps does not improve in-stent restenosis prediction over conventional models; leaves open the value of newly identified predictors.
Objective: Machine learning (ML) approaches have the potential to uncover regular patterns in multi-layered data. Here we applied self-organizing maps (SOMs) to detect such patterns with the aim to better predict in-stent restenosis (ISR) at surveillance angiography 6 to 8 months after percutaneous coronary intervention with stenting. Methods: In prospectively collected data from 10,004 patients receiving percutaneous coronary intervention (PCI) for 15,004 lesions, we applied SOMs to predict ISR angiographically 6–8 months after index procedure. SOM findings were compared with results of conventional uni- and multivariate analyses. The predictive value of both approaches was assessed after random splitting of patients into training and test sets (50:50). Results: Conventional multivariate analyses revealed 10, mostly known, predictors for restenosis after coronary stenting: balloon-to-vessel ratio, complex lesion morphology, diabetes mellitus, left main stenting, stent type (bare metal vs. first vs. second generation drug eluting stent), stent length, stenosis severity, vessel size reduction, and prior bypass surgery. The SOM approach identified all these and nine further predictors, including chronic vessel occlusion, lesion length, and prior PCI. Moreover, the SOM-based model performed well in predicting ISR (AUC under ROC: 0.728); however, there was no meaningful advantage in predicting ISR at surveillance angiography in comparison with the conventional multivariable model (0.726, p = 0.3). Conclusions: The agnostic SOM-based approach identified—without clinical knowledge—even more contributors to restenosis risk. In fact, SOMs applied to a large prospectively sampled cohort identified several novel predictors of restenosis after PCI. However, as compared with established covariates, ML technologies did not improve identification of patients at high risk for restenosis after PCI in a clinically relevant fashion.
No takes yet. Share an insight, caveat, or question.
Güldener et al. (2023) conducted a cohort in Coronary artery disease post-PCI (n=10,004). Machine learning (self-organizing maps) vs. Conventional multivariable model was evaluated on Prediction of in-stent restenosis at surveillance angiography (AUC under ROC) (p=0.3). Machine learning using self-organizing maps identified new predictors but did not significantly improve the prediction of in-stent restenosis compared to conventional multivariable analysis (AUC 0.728 vs 0.726).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: