My research focuses on deep learning within healthcare informatics.
I am currently concentrating on the generative adversarial networks for domain adaptation, noise reduction, and disease interpolation in ultrasound images.
Project: Evaluation of New AI Methods for Ultrasound Imaging Based Subclinical Atherosclerosis Risk Assessment.
Ultrasound imaging of the carotid arteries is an important method for the assessment of subclinical atherosclerosis. In atherosclerotic disease, fatty streaks start infiltrating the arterial walls and may eventually build up to localized plaques, which may rupture and cause stroke. However, ultrasound images are difficult to read because of operator dependence and noise contamination. In this study, artificial intelligence will be used to develop new decision support systems aiming to overcome some of these difficulties.