Classification Algorithms for Stenotic Lesion Detection in Coronary CT Angiograms

Bachelor Thesis

Pattern recognition is an important task for computer aided diagnosis(CAD) systems. A prototype application for the automatic detection of stenosis in CT datasets using a specialized feature extraction method for coronary vessels in conjunction with the AdaBoost algorithm for learning and classification has been implemented.
Based on this prototype software, different weak- and strong-learning algorithms for the detection of stenotic lesions in coronary CT angiograms shall be implemented and investigated.

Advisor(s)