Related Experiment Video
Updated: Jun 9, 2025
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Precision phenotyping from routine laboratory parameters for out of hospital survival prediction in an all comers
Paul-Adrian Călburean1,2, Marius Harpa3,4, Anda-Cristina Scurtu3,4
1George Emil Palade University of Medicine, Pharmacy, Science, and Technology of Targu Mures, Gheorghe Marinescu Str., no. 38, Târgu Mureş, Târgu Mureş, Romania. calbureanpaul@gmail.com.
Insights
Predicting out-of-hospital deaths in coronary artery disease (CAD) patients after percutaneous coronary intervention (PCI) is crucial. Machine learning models using routine lab tests accurately predict survival, outperforming traditional risk scores.
Area of Science:
- Cardiology
- Medical Informatics
- Biostatistics
Background:
- Out-of-hospital mortality is a significant concern in coronary artery disease (CAD).
- Existing prediction tools for adverse events in CAD patients undergoing percutaneous coronary intervention (PCI) are limited.
- Precision phenotyping could improve risk stratification.
Purpose of the Study:
- To evaluate the utility of routine laboratory parameters for predicting out-of-hospital survival in CAD patients treated with PCI.
- To develop and validate a machine learning (ML) model for this prediction task.
Main Methods:
- A prospective registry of 4027 patients undergoing PCI between 2016-2022 was analyzed.
- An extreme gradient-boosted decision tree ML algorithm was trained on 115 PCI registry and 266 routine laboratory parameters.
- The model's predictive performance was validated on an independent dataset of 1729 patients.
Main Results:
- The ML model achieved high accuracy in predicting all-cause (AUC 0.844) and cardiovascular (AUC 0.837) mortality.
- Key predictors included parameters of renal function, hematologic status, and inflammatory markers.
- ML models demonstrated superior performance compared to established clinical risk scores.
Conclusions:
- Routine laboratory parameters, when analyzed with ML, can accurately predict out-of-hospital survival in CAD patients post-PCI.
- This approach offers precise patient phenotyping and improved risk stratification.
- ML models have the potential to enhance clinical decision-making for CAD management.
Abstract:
Out-of-hospital mortality in coronary artery disease (CAD) is particularly high and established adverse event prediction tools are yet to be available. Our study aimed to investigate whether precision phenotyping can be performed using routine laboratory parameters for the prediction of out-of-hospital survival in a CAD population treated by percutaneous coronary intervention (PCI). All patients treated by PCI and discharged alive in a tertiary center between January 2016 - December 2022 that have been included prospectively in the local registry were analyzed. 115 parameters from the PCI registry and 266 parameters derived from routine laboratory testing were used. An extreme gradient-boosted decision tree machine learning (ML) algorithm was trained and used to predict all-cause and cardiovascular-cause survival. A total of 4027 patients with 4981 PCI hospitalizations were randomly included in the 70% training dataset and 1729 patients with 2160 PCI hospitalizations were randomly included in the 30% validation dataset. All-cause and cardiovascular cause mortality was 17.5% and 12.2%. The integrated area under the receiver operator characteristic curve for prediction of all-cause and cardiovascular cause mortality by the ML on the validation dataset was 0.844 and 0.837, respectively (all p < 0.001). Parameters reflecting renal function (first and maximum serum creatinine), hematologic function (mean corpuscular hemoglobin concentration, platelet distribution width), and inflammatory status (lymphocyte per monocyte ratio) were among the most important predictors. Accurate out-of-hospital survival prediction in CAD can be achieved using routine laboratory parameters. ML outperformed clinical risk scores in predicting out-of-hospital mortality in a prospective all-comers PCI population and has the potential to precisely phenotype patients.
