My research interests center on developing medical imaging systems, particularly optical imaging techniques. I am passionate about leading innovations in optical imaging by combining my physics knowledge with computer science, deep learning, and medical image analysis. 

Having lost young family members to sudden heart attacks and having lived through the fear of having symptoms of cardiac arrhythmia and vascular obstruction, I take the importance of advancing the state-of-the-art in clinical cardiac interventions as a personal task. My background in physics enabled me to understand the nuances of cardiac imaging systems and their underlying principles. Cardiac imaging consists of an aggregation of radiological modalities and a full range of cardiac analyses. As a result, I realized that optical imaging systems are feasible cardiac diagnostic tools since (1) there is no harmful radiation involved, and (2) image acquisition is fast with minimum risk for patients. Most importantly, the state-of-the-art light-based intravascular imaging modality, Optical Coherence Tomography (OCT), provides intracoronary cross-sectional images with the highest resolution. However, despite its many advantages, OCT should be used more. The need for more use of OCT is related to time-consuming operator-dependent OCT image interpretation. I identified the gap as lacking a precise and accurate automated technology for analyzing OCT images of intracoronary arteries, which would enable widespread use of this technology for diagnosis and intervention purposes. Thus, I structured my Ph.D. and the two years of my Postdoctoral fellowship to conduct research addressing this clinical need.

Following the success story of Marvin Woodall –the retired president of Johnson & Johnson interventional systems– inspired me to investigate and define another clear and critical unmet clinical need to be investigated during my postdoctoral research. While Woodall was re-evaluating the success of his company's coronary stent, he studied the data accumulated through the early 2000s. He realized that coronary stents could not significantly reduce the cardiac mortality that many had hoped for. The reason was that stenting could treat stable plaques, which are not the main reason for death in populations with coronary events. Although interventional cardiology created a new platform to help millions of patients avoid bypass surgery, the current upper edge need is to detect the vulnerable plaque, those types of coronary plaques that rupture suddenly and result in sudden cardiac death. The rupture of vulnerable plaque causes seventy-five percent of deaths due to coronary events. Unfortunately, the current cardiac imaging systems have many limitations when detecting these plaques. Considering the advantages and incredible functionality of intravascular OCT imaging, my research focused on evaluating and automatically indicating vulnerable plaques using intracoronary OCT images.

Joining ViTAA medical solutions as Head Advanced Technology, I realized another crucial clinical need regarding the abdominal aortic aneurysm (AAA). Since AAA is highly asymptomatic, the symptoms will be shown when the aneurysm is ruptured or prone to rupture. Therefore, evaluating the risk progression and rupture is significant to be addressed. Although some imaging systems are used to assess AAA tissues, computed tomography (CT) imaging is required for collecting detailed information regarding the disease progression. The large variety in decision-making and planning the repair of an aneurysm from one patient to another stresses the need for personalized decision-making approaches. Other than that, the lack of accurate technique to automatically characterize various aortic tissues results in applying a high dose of contrast agent to make the visual representation of the multiple tissues possible. Visual detection of the tissues using a high dose of contrast agent is still time-consuming and error-prone from one observer to another. Therefore, my main focus was developing a model composed of different steps starting with the extraction of the aorta and iliac artery, followed by detecting the lumen and intraluminal thrombus.