An XGBoost machine learning model based on coronary angiographic characteristics accurately predicted immediate PCI success in patients with moderate-to-severe calcification (AUC 0.972).
Observational (n=3,000)
Yes
Can machine learning models based on coronary angiographic characteristics accurately predict immediate PCI success in patients with moderate-to-severe coronary artery calcification?
Machine learning models, particularly XGBoost, can highly accurately predict immediate PCI success in patients with moderate-to-severe coronary calcification using key angiographic features.
Effect estimate: AUC 0.972
Abstract Objective Given the challenges faced during percutaneous coronary intervention (PCI) for heavily calcified lesions, accurately predicting the PCI success is crucial for enhancing patient outcomes and optimizing procedural strategies. The aim of this study was to utilize machine learning to identify coronary angiographic vascular characteristics associated with the immediate procedural success rates of PCI in patients exhibiting moderate-to-severe coronary artery calcification (MSCAC). Methods A retrospective analysis was performed using data from two prospective clinical studies. Six machine learning models were developed and validated, with SMOTE used for dealing with imbalanced data. The study screened for the top five features associated with PCI success, validated in the external testing cohort. Results More than three thousand patients with MSCAC met the inclusion criteria, with an overall immediate PCI success rate of 89.4%. The XGBoost model emerged as the most predictive, achieving an AUC of 0.946 in the development set and validating with an AUC of 0.972 in the external testing dataset. The key predictive factors identified for immediate PCI success included minimum lumen diameter, lesion length, chronic total occlusion, degree of calcification, and angulation. Conclusions This study successfully revealed the important PCI failure risk factors utilizing machine learning models to help clinicians manage PCI strategies in patients with challenging coronary artery anatomies such as MSCAC.
Ye et al. (Sat,) conducted a observational in Moderate-to-severe coronary artery calcification (MSCAC) (n=3,000). Coronary angiographic vascular characteristics (minimum lumen diameter, lesion length, chronic total occlusion, degree of calcification, and angulation) was evaluated on Immediate procedural success rates of PCI (AUC 0.972). An XGBoost machine learning model based on coronary angiographic characteristics accurately predicted immediate PCI success in patients with moderate-to-severe calcification (AUC 0.972).