Automatic alignment of pre- and post-interventional liver CT images for assessment of radiofrequency ablation

Christian Rieder, Stefan Wirtz, Jan Strehlow, Stephan Zidowitz, Philipp Bruners, Peter Isfort, Andreas H. Mahnken, Heinz-Otto Peitgen, David R. Holmes III, Kenneth H. Wong
<span title="2012-02-23">2012</span> <i title="SPIE"> <a target="_blank" rel="noopener" href="" style="color: black;">Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling</a> </i> &nbsp;
Image-guided radiofrequency ablation (RFA) is becoming a standard procedure for minimally invasive tumor treatment in clinical practice. To verify the treatment success of the therapy, reliable post-interventional assessment of the ablation zone (coagulation) is essential. Typically, pre-and post-interventional CT images have to be aligned to compare the shape, size, and position of tumor and coagulation zone. In this work, we present an automatic workflow for masking liver tissue, enabling a
more &raquo; ... gid registration algorithm to perform at least as accurate as experienced medical experts. To minimize the effect of global liver deformations, the registration is computed in a local region of interest around the pre-interventional lesion and post-interventional coagulation necrosis. A registration mask excluding lesions and neighboring organs is calculated to prevent the registration algorithm from matching both lesion shapes instead of the surrounding liver anatomy. As an initial registration step, the centers of gravity from both lesions are aligned automatically. The subsequent rigid registration method is based on the Local Cross Correlation (LCC) similarity measure and Newton-type optimization. To assess the accuracy of our method, 41 RFA cases are registered and compared with the manually aligned cases from four medical experts. Furthermore, the registration results are compared with ground truth transformations based on averaged anatomical landmark pairs. In the evaluation, we show that our method allows to automatic alignment of the data sets with equal accuracy as medical experts, but requiring significancy less time consumption and variability.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="">doi:10.1117/12.911188</a> <a target="_blank" rel="external noopener" href="">dblp:conf/miigp/RiederWSZBIMP12</a> <a target="_blank" rel="external noopener" href="">fatcat:3zwhlmqv55fajfqph57y264b2m</a> </span>
<a target="_blank" rel="noopener" href="" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href=""> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / </button> </a>