TY - JOUR AU - Ahmadian, Ghazaleh AU - Bohun, C. Sean AU - Ebrahimi, Mehran PY - 2016/10/03 Y2 - 2024/03/29 TI - Evaluation of a Coherent Point Drift Algorithm for Breast Image Registration via Surface Markers JF - Journal of Computational Vision and Imaging Systems JA - J. Comp. Vis. Imag. Sys. VL - 2 IS - 1 SE - Articles DO - 10.15353/vsnl.v2i1.90 UR - https://openjournals.uwaterloo.ca/index.php/vsl/article/view/90 SP - AB - <p>Breast Magnetic Resonance Imaging (MRI) is a reliable imaging<br />tool for localization and evaluation of lesions prior to breast conserving<br />surgery (BCS). MR images typically will be used to determine<br />the size and location of the tumours before making the incision<br />in order to minimize the amount of tissue excised.<br />The arm position and configuration of the breast during and<br />prior to surgery are different and one question is whether it would<br />be possible to match the two configurations. This matching process<br />can potentially be used in development of tools to guide surgeons<br />in the incision process.<br />Recently, a Thin-Plate-Spline (TPS) algorithm has been proposed<br />to assess the feasibility of breast tissue matching using fiducial<br />surface markers in two different arm positions. The registration<br />algorithm uses the surface markers only and does not employ the<br />image intensities.<br />In this manuscript, we apply and evaluate a coherent point drift<br />(CPD) algorithm for registration of three-dimensional breast MR images<br />of six patient volunteers. In particular, we evaluate the results<br />of the previous TPS registration technique to the proposed rigid<br />CPD, affine CPD, and deformable CPD registration algorithms on<br />the same patient datasets.<br />The preliminary results suggest that the CPD deformable registration<br />algorithm is superior in correcting the motion of the breast<br />compared to CPD rigid, affine and TPS registration algorithms.</p> ER -