Cardiovascular diseases (CVDs) are circulatory system disorders caused by abnormal structures or functions of the heart or blood vessels, and imaging examinations are the core diagnostic methods for them. Traditional imaging techniques (such as echocardiography, angiography CT, magnetic resonance imaging, and positron emission tomography) play a key role in the diagnosis of CVDs, but they are limited by resolution, invasiveness or functional assessment capabilities. The emerging intravascular ultrasound imaging and optical coherence tomography technologies, by integrating the advantages of acoustic and optical imaging, have enable visualization of deep vascular structures and the precise detection of plaque microlesions, significantly enhancing the early diagnosis and the efficiency of treatment guidance of CVDs. In recent years, artificial intelligence (AI)-driven multimodal image fusion analysis has further broken through technical bottlenecks: by integrating multi-source data to build risk prediction models, it not only optimizes diagnostic accuracy but also achieves standardized and automated reporting, promoting the development of CVDs screening towards large-scale, high-precision, and high-efficiency directions. This article systematically reviews the research progress of traditional and AI-enabled imaging techniques in the diagnosis of CVDs, aiming to provide new ideas for clinical research and theoretical basis for the development of future precise diagnosis and treatment plans.