Cardiovascular Diseases (CVDs), predominantly caused by atherosclerosis, necessitate precise diagnostic tools for effective treatment. Intravascular ultrasound (IVUS) imaging is widely used for assessing arterial structures and identifying cholesterol plaques. This study introduces an advanced image-processing framework leveraging the DeepLab model, a state-of-the-art deep learning architecture for semantic segmentation, to analyze IVUS images. The DeepLab model is enhanced with atrous spatial pyramid pooling (ASPP) and fine-tuned for segmenting critical arterial features, including media-adventitia borders, luminal regions, and calcified plaques. The system incorporates advanced loss functions such as Dice, Tversky, and focal loss to address class imbalance and improve segmentation accuracy. Comparisons with commercial software, such as VH-IVUS, highlight the superiority of the proposed method in scenarios involving shadow artifacts or side vessels. This DeepLab-based approach offers a robust and efficient solution for IVUS image analysis, paving the way for improved diagnostic accuracy and better clinical outcomes in the management of CVDs.