Related Experiment Video
Updated: Jul 6, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
AI-Enhanced Predictive Modeling for Identifying Depression and Delirium in Cardiovascular Patients Scheduled for
Karina Nowakowska1, Antonis Sakellarios2,3, Jakub Kaźmierski1
1Department of Old Age Psychiatry and Psychotic Disorders, Medical University of Lodz, 90-419 Lodz, Poland.
Insights
Cardiovascular disease patients with depression can be identified with 62% accuracy using an AI pipeline that analyzes biomarkers like sRAGE. This approach aids early intervention for improved cardiac patient care.
Area of Science:
- Cardiology
- Psychiatry
- Artificial Intelligence
Background:
- A significant association exists between cardiovascular disease (CVD) and mental health, with approximately one-third of CVD patients experiencing depression.
- This comorbidity elevates the risk of cardiac complications and mortality, independent of traditional risk factors.
- Early identification and intervention for depression in CVD patients are crucial for improving outcomes.
Purpose of the Study:
- To develop and validate a straightforward, explainable, and data-driven pipeline for predicting depression in patients with cardiovascular disease.
- To investigate the correlation between specific biomarkers and depression in the context of CVD.
- To leverage machine learning for enhanced diagnostic capabilities in cardiac care.
Main Methods:
- A cohort of 224 patients scheduled for elective coronary artery bypass graft (CABG) surgery was evaluated.
- Psychiatric evaluations were conducted pre-surgery to diagnose major depressive disorder (MDD) using DSM-5 criteria.
- An explainable AI pipeline, incorporating AdaBoost, random forest, and XGBoost algorithms, was trained and tested on curated patient data using stratified cross-validation.
Main Results:
- A significant correlation was found between the soluble form of receptor for advanced glycation end products (sRAGE) biomarker and depression (r = 0.32, p = 0.038).
- The random forest classifier achieved the highest accuracy (0.62), sensitivity (0.71), specificity (0.53), and area under the curve (0.67) in predicting depression.
- The AI pipeline demonstrated a 62% accuracy rate in predicting depression among CVD patients.
Conclusions:
- Depression in cardiovascular disease patients, especially those with elevated sRAGE levels, can be predicted with significant accuracy using an AI-driven approach.
- This study validates the utility of explainable AI in identifying mental health comorbidities in cardiac patients.
- The developed pipeline offers a promising tool for early detection and intervention, potentially transforming care strategies for this vulnerable population.
Abstract:
Several studies have demonstrated a critical association between cardiovascular disease (CVD) and mental health, revealing that approximately one-third of individuals with CVD also experience depression. This comorbidity significantly increases the risk of cardiac complications and mortality, a risk that persists regardless of traditional factors. Addressing this issue, our study pioneers a straightforward, explainable, and data-driven pipeline for predicting depression in CVD patients.
Methods:
Our study was conducted at a cardiac surgical intensive care unit. A total of 224 participants who were scheduled for elective coronary artery bypass graft surgery (CABG) were enrolled in the study. Prior to surgery, each patient underwent psychiatric evaluation to identify major depressive disorder (MDD) based on the DSM-5 criteria. An advanced data curation workflow was applied to eliminate outliers and inconsistencies and improve data quality. An explainable AI-empowered pipeline was developed, where sophisticated machine learning techniques, including the AdaBoost, random forest, and XGBoost algorithms, were trained and tested on the curated data based on a stratified cross-validation approach.
Results:
Our findings identified a significant correlation between the biomarker "sRAGE" and depression (r = 0.32, p = 0.038). Among the applied models, the random forest classifier demonstrated superior accuracy in predicting depression, with notable scores in accuracy (0.62), sensitivity (0.71), specificity (0.53), and area under the curve (0.67).
Conclusions:
This study provides compelling evidence that depression in CVD patients, particularly those with elevated "sRAGE" levels, can be predicted with a 62% accuracy rate. Our AI-driven approach offers a promising way for early identification and intervention, potentially revolutionizing care strategies in this vulnerable population.
