Issues involved in automatic selection and intensity based matching of feature points for mls registration of medical images

Abstract

Most of the Registration algorithms require selection of corresponding control points from the source and the target images. The MLS being a point based method requires the selection of control points for registration. The accuracy of registration process depends greatly on the selected control points. Hence, feature detection and matching play an important role in the process of point based registration of medical images. Therefore an analysis on the consequence of automating the control point selection process using feature extraction algorithms like Harris Corner, Min-Eigen, Speeded Up Robust Features (SURF) and Canny Edge pixels is deemed to be essential. Since the end users are medical practitioners, who prefer to have an interactive system, where the control points need to be selected based on the diagnostic requirements analysis on manual selection of control points has also been included. The issues involved in automatic control point selection from MRI/CT images have been discussed in this work.