Cardiac health diagnosis using data fusion of cardiovascular and haemodynamic signals

Area of Science:

• Biomedical Engineering
• Cardiology
• Artificial Intelligence in Medicine

Background:

• Electrocardiograms (ECG) provide vital cardiac information but are difficult to interpret alone.
• Subtle ECG details like P-QRS-T waves and r-r intervals are crucial for diagnosing heart conditions.
• Current methods struggle to fully evaluate cardiac health solely through ECG analysis.

Purpose of the Study:

• To develop a computer-based system for analyzing heterogeneous physiological signals.
• To improve clinical diagnosis of patients in cardiac care units through data fusion.
• To introduce a novel 'deterioration index' for quantifying cardiac disease severity.

Main Methods:

• Utilized fuzzy logic for data fusion of multiple physiological signals, including ECG, blood pressure, oxygen saturation, and respiratory data.
• Developed a computer-based analysis and display system for real-time monitoring.
• Proposed and calculated a new parameter, the 'deterioration index', to assess disease severity.

Main Results:

• Demonstrated the effectiveness of fuzzy logic-based data fusion for detecting life-threatening cardiac states.
• Successfully implemented a computer-based system for analyzing combined physiological data.
• Tabulated results for the 'deterioration index' across various patient cases, showing its utility in severity evaluation.

Conclusions:

• Data fusion of ECG with other vital signs significantly enhances the accuracy of cardiac patient diagnosis.
• The proposed 'deterioration index' offers a valuable tool for objectively evaluating the severity of cardiac disease.
• Fuzzy logic provides a robust framework for integrating complex physiological data in critical care settings.