Coronary arteries segmentation based on the 3D discrete wavelet transform and 3D neutrosophic transform

Area of Science:

• Medical Image Processing
• Cardiovascular Imaging
• Computational Pathology

Background:

• Precise estimation is crucial for medical image processing applications.
• Accurate segmentation of coronary arteries is essential for detecting pathologies.

Purpose of the Study:

• To propose a novel segmentation method for coronary arteries.
• To enable automatic and accurate detection of coronary pathologies.

Main Methods:

• A two-part segmentation approach was developed.
• Initial segmentation utilized 3D region growing.
• Vessel information was detected using a vessel-texture discrimination algorithm with 3D Discrete Wavelet Transform (DWT) HHH subband coefficients, integrated with 3D neutrosophic transformation for accurate detection.

Main Results:

• The proposed method accurately segmented each subbranch of the coronary arteries.
• Results demonstrated superior efficiency compared to commercial software (GE Healthcare) and the level-set method (Yang et al., 2007).

Conclusions:

• The integration of 3D region growing, one-level 3D DWT, and 3D neutrosophic transformation provides accurate detection of coronary pathologies.
• This method enhances the precision of coronary artery segmentation in medical imaging.