PhD thesis defense to be held on April 17, 2024, at 17:00 (Teleteaching Room, NTUA Central Library)

Picture Credit: K. Trikouraki

Thesis title: Development of new combined biotechnological techniques and algorithms for the evaluation of cardiovascular disease biomarkers

Abstract: The objective was to assess biomarkers between symptomatic vs. asymptomatic patients, and to construct a classification and regression tree (CART) algorithm for their discrimination. For that reason 136 patients were enrolled. They were symptomatic (high risk) (N=82, stenosis degree ≥50%, proven to be responsible for ischemic stroke the last six months) and asymptomatic (low risk) (N=54, stenosis degree ≤50%). Levels of Fibrinogen, matrix metalloproteinase-1 (MMP-1), tissue inhibitor of metalloproteinase-1 (TIMP-1), soluble intercellular adhesion molecule (SiCAM), soluble vascular cell adhesion molecule (SvCAM), Adiponectin and Insulin were measured on a Luminex 3D platform and their differences were evaluated; subsequently a CART model was created and evaluated. As far as the results concerned all measured biomarkers, except Adiponectin, had significantly higher levels in symptomatic patients. The constructed CART prognostic model had 97.6% discrimination accuracy on symptomatic patients and 79.6% on asymptomatic, while the overall accuracy was 90.4%. Moreover the population was split into training and test sets for CART validation.
To conclude significant differences were found in the biomarkers between symptomatic and asymptomatic patients. The CART model proved to be a simple decision making algorithm linked with risk probabilities and provided evidence to identify and therefore treat patients being at high risk for cardiovascular disease.

Supervisor: Professor Emerita Dido Yova

PhD Student: K. Trikouraki