AI-Driven Decision Support for Early Detection of Cardiac Events: Unveiling Patterns and Predicting Myocardial...

Area of Science:

• Cardiology
• Medical Informatics
• Data Science

Background:

• Cardiovascular diseases (CVDs) represent a major global health challenge, contributing significantly to mortality rates worldwide.
• Effective strategies for managing critical CVD events such as myocardial infarction, pulmonary thromboembolism, and aortic stenosis are essential for improving patient outcomes.
• The integration of advanced data analysis techniques into clinical practice holds potential for enhancing diagnostic and therapeutic decision-making.

Purpose of the Study:

• To develop and validate predictive machine learning (ML) models for early detection of cardiovascular events.
• To leverage exploratory data analysis (EDA) to uncover complex patterns within cardiovascular disease data.
• To provide medical practitioners with enhanced tools for informed decision-making and timely interventions in critical cardiac care.

Main Methods:

• Utilized the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology for structured data analysis.
• Performed extensive Exploratory Data Analysis (EDA) on patient data from Hospital Santa Maria to identify disease-specific patterns.
• Developed and trained predictive Machine Learning (ML) models to forecast the occurrence of cardiovascular events.

Main Results:

• Exploratory Data Analysis revealed significant, intricate patterns and relationships pertinent to cardiovascular diseases.
• Machine Learning models demonstrated predictive accuracies exceeding 80% for identified cardiovascular events.
• A predictive window of 13 minutes was established for forecasting myocardial ischemia incidents, facilitating proactive intervention.

Conclusions:

• The study successfully demonstrates a Proof of Concept for real-time data utilization and predictive analytics in cardiovascular medicine.
• The developed ML models offer a valuable tool for enhancing medical strategies and improving patient care through early detection and intervention.
• This approach has the potential to significantly impact the management of cardiovascular diseases by enabling proactive and informed clinical decision-making.