4-D cardiac MR image analysis: left and right ventricular morphology and function.

Area of Science:

• Medical Imaging
• Cardiovascular Research
• Computational Anatomy

Background:

• Accurate segmentation of cardiac structures is crucial for diagnosing and monitoring heart conditions.
• Tetralogy of Fallot (TOF) presents complex ventricular abnormalities that challenge standard imaging analysis.

Purpose of the Study:

• To develop and validate a robust method for segmenting left and right ventricles in 4-D cardiac MR images of normal and TOF hearts.
• To extract quantitative shape and motion features for automated classification of TOF and assessment of disease progression.

Main Methods:

• Combined Active Shape Model (ASM) and Active Appearance Model (AAM) for 4-D cardiac MRI segmentation.
• Utilized both 4-D and 3-D models for robust and accurate ventricular segmentation across all cardiac phases.
• Extracted quantitative features including shape, volume-time, and dV/dt curves for analysis.

Main Results:

• Achieved subvoxel segmentation accuracy, high overlap ratios, and strong ventricular volume correlations compared to expert standards.
• Demonstrated high sensitivity (90%-100%) and specificity in automated discrimination between normal and TOF subjects.
• Identified higher feature variability in TOF hearts, indicating potential as disease progression indicators.

Conclusions:

• The combined ASM-AAM approach provides accurate and robust 4-D cardiac segmentation for normal and TOF hearts.
• Extracted quantitative features effectively characterize TOF abnormalities and enable accurate disease classification.
• The method shows promise for monitoring disease progression and guiding clinical management in TOF patients.