Predictive model of neurocognitive functioning after acute coronary syndrome. A machine learning approach

Area of Science:

• Cardiology
• Neuroscience
• Machine Learning

Background:

• The relationship between coronary artery disease and neurocognitive dysfunction is complex and not fully understood.
• Several factors may contribute to this interplay, necessitating further investigation.

Purpose of the Study:

• To develop a predictive model for neurocognitive functioning in patients with acute coronary syndrome (ACS).
• To utilize a machine learning approach for predicting neurocognitive outcomes in ACS patients.

Main Methods:

• A cross-validated random forest model (RF_cv) was employed for prediction.
• Sixty-three patients in a cardiac rehabilitation program underwent neurocognitive assessment.

Main Results:

• The RF_cv model demonstrated high accuracy with an r-squared of 0.978.
• Top predictors included HDL, depression, glucose, HbA1c, BNP, BMI, WHR, cholesterol, anxiety, and age.

Conclusions:

• Neurocognitive functioning variance is linked to biochemical markers and body composition, including cardiovascular risk factors and depression.
• The predictive model aids in early identification of patients for personalized interventions.