Automated recognition of the pericardium contour on processed CT images using genetic algorithms

Area of Science:

• Medical Imaging Analysis
• Computational Biology
• Cardiovascular Health

Background:

• Accurate pericardium contour identification is crucial for medical diagnosis and cardiovascular risk assessment.
• Manual tracing of the pericardium is time-consuming and often not performed, limiting its clinical utility.
• Automated methods for pericardium detection can improve the quantification of epicardial and mediastinal fats.

Purpose of the Study:

• To develop and evaluate an automated method for tracing the pericardium contour using Genetic Algorithms (GA).
• To model pericardium slices as ellipses and optimize their parameters for accurate contour detection.
• To enhance automated fat separation methodologies and improve cardiovascular risk marker information.

Main Methods:

• Application of Genetic Algorithms (GA) for optimizing ellipse parameters to represent the pericardium contour.
• Utilizing Computed Tomography (CT) images for pericardium tracing and analysis.
• Developing an automated approach to separate epicardial and mediastinal fats based on accurate pericardium detection.

Main Results:

• Genetic Algorithms (GA) provide a feasible and efficient solution for tracing the pericardium contour in CT images.
• The proposed method demonstrates satisfactory performance in identifying the pericardium, aiding in automated fat quantification.
• The optimized elliptical models closely follow the actual pericardium, enabling accurate separation of surrounding fats.

Conclusions:

• Genetic Algorithms (GA) are a viable tool for automated pericardium contour detection in medical imaging.
• This automated approach can significantly reduce manual effort and improve the accuracy of cardiovascular fat quantification.
• Accurate pericardium identification using GA contributes to better medical diagnosis and cardiovascular risk assessment.