Cardiovascular Diseases

It is well established that carotid atherosclerotic plaques highly predispose to cerebral ischemic events, with the majority of stroke events being provoked due to the disease. As early and valid discrimination between vulnerable and stable carotid plaques is critical for optimal management of carotid atherosclerosis, our research heads towards novel noninvasive and low-cost markers for predicting strokes in asymptomatic patients using ultrasound-image-based features. In this context, echogenicity, texture and motion characteristics are combined to assist clinical decision towards improved prediction, assessment and management of cardiovascular risk. Our research in this field focuses on establishing novel markers for reliable early-detection and risk-stratification of carotid atherosclerosis based on ultrasound image analysis and can be summarized as follows.

  • Segmentation of the arterial wall in order to identify automatically the areas of interest, i.e. the healthy arterial wall, the atherosclerotic lesions and intra-plaque structures revealing their composition.
  • Implementation and in silico evaluation of novel motion estimators with enhanced accuracy and efficiency in motion tracking, as there is evidence that current clinical practice -which decides only upon the degree of lumen stenosis, a criterion that does not ensure valid identification of vulnerable atherosclerotic lesions and assessment of these patients- in not sufficient for the evaluation of these patients.
  • Development of a standardized framework for arterial wall characterization in terms of texture and motion/strain features in vivo in order to study different phenotypes of the disease and quantify plaque response to treatment by using imaging and biochemical biomarkers. The aforementioned framework has been combined with sophisticated classification tools aiming at personalized risk assessment of patients with carotid atherosclerosis.
  • Incorporation of the developed computational tools, together with effective data organization technologies (semantic web), in a web-based platform for personalized computer-assisted diagnosis and treatment. This work has been carried out within the framework of a national grant on “The carotid atheromatous plaque: a multidisciplinary approach for optimal management of symptomatic and asymptomatic patients”.
  • Investigation of the role of Elastography Imaging, a novel promising ultrasound technique yet not fully explored and established. Elasticity of the carotid artery is thought to be an important characteristic of the arterial wall and atherosclerotic lesions, determining its motion behavior, thus its instability. Elastography Imaging is promising for optimizing diagnosis of carotid disease and our research aims to investigate the association between elasticity -calculated from elastographic images- and texture and motion indices -calculated from duplex ultrasound image analysis.
  • Combination of motion and texture analysis of, and within, the atherosclerotic plaque, towards the development of a non-invasive biopsy tool of atherosclerotic lesions based on ultrasound imaging.