Medical Image Analysis and Processing

(a) Analysis of Ultrasound images of the carotid arteries in order to predict the risk of stroke

Work in this field is about the development of a computer based system which will be able to help clinicians in the identification of the risk of stroke, using ultrasound images from the carotid arteries. There are indications that the morphology of atherosclerotic carotid plaques, obtained by high resolution ultrasound imaging, has prognostic value. In order to use these indications, texture and morphological features were extracted from the ultrasound plaque images so as to be used for the classification of the carotid plaques. The plaques were classified into two types: symptomatic because of ipsilateral hemispheric symptoms, or asymptomatic because they were not connected with ipsilateral hemispheric events.

Based on these indications a systems that helps clinicians work on ultrasound image analysis has been created and currently in use by several clinicians located at worldwide : Norway (Tromsoe University), UK (Imperial College, Addenbrooke's Hospital Cambridge University Hospitals NHS Foundation Trust, Vascular Studies Unit University Hospitals of Leicester), USA (Columbia University - NY, Jobst Vascular Center, Promedica health system), Cyprus (CDER trust), Greece (University of Athens, University of Crete), Slovenia (University of Ljubljana), Poland (Krakow University School of Medicine), Spain (Madrid), India (Boston Scientific Corporation), Germany (Universitätsmedizin der Johannes Gutenberg-Universität Mainz), Nederlands (Dept. of Vascular Medicine AMC Universiteit Van Amsterdam). Several other groups have expressed interest in using the system.

This research is been performed in collaboration with the medical team of Prof. Andrew Nicolaides recognised as a leader in the area of cardiovascular research for the evaluation of the risk of stroke.