<?xml version="1.0" encoding="UTF-8"?><xml><records><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>47</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Jhimli Mitra</style></author><author><style face="normal" font="default" size="100%">Zoltan Kato</style></author><author><style face="normal" font="default" size="100%">Soumya Ghose</style></author><author><style face="normal" font="default" size="100%">Desire Sidibe</style></author><author><style face="normal" font="default" size="100%">Robert Martí</style></author><author><style face="normal" font="default" size="100%">Xavier Lladó</style></author><author><style face="normal" font="default" size="100%">Oliver Arnau</style></author><author><style face="normal" font="default" size="100%">Joan C Vilanova</style></author><author><style face="normal" font="default" size="100%">Fabrice Meriaudeau</style></author></authors><secondary-authors><author><style face="normal" font="default" size="100%">Jan-Olof Eklundh</style></author><author><style face="normal" font="default" size="100%">Yuichi Ohta</style></author><author><style face="normal" font="default" size="100%">Steven Tanimoto</style></author></secondary-authors></contributors><titles><title><style face="normal" font="default" size="100%">Spectral clustering to model deformations for fast multimodal prostate registration</style></title><secondary-title><style face="normal" font="default" size="100%">International Conference on Pattern Recognition (ICPR)</style></secondary-title></titles><dates><year><style  face="normal" font="default" size="100%">2012</style></year><pub-dates><date><style  face="normal" font="default" size="100%">Nov 2012</style></date></pub-dates></dates><urls><web-urls><url><style face="normal" font="default" size="100%">http://hal.archives-ouvertes.fr/docs/00/71/09/43/PDF/ICPR_Jhimli.pdf</style></url></web-urls></urls><publisher><style face="normal" font="default" size="100%">IEEE</style></publisher><pub-location><style face="normal" font="default" size="100%">Tsukuba, Japan</style></pub-location><pages><style face="normal" font="default" size="100%">2622 - 2625</style></pages><isbn><style face="normal" font="default" size="100%">978-1-4673-2216-4 </style></isbn><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">&lt;div class=&quot;article&quot;&gt;&lt;p&gt;This paper proposes a method &lt;span class=&quot;snippet&quot;&gt;to&lt;/span&gt; learn &lt;span class=&quot;snippet&quot;&gt;deformation&lt;/span&gt; parameters off-line for &lt;span class=&quot;snippet&quot;&gt;fast&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;multimodal&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; of ultrasound and magnetic resonance &lt;span class=&quot;snippet&quot;&gt;prostate&lt;/span&gt; images during ultrasound guided needle biopsy. The &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; method involves &lt;span class=&quot;snippet&quot;&gt;spectral&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;clustering&lt;/span&gt; of the &lt;span class=&quot;snippet&quot;&gt;deformation&lt;/span&gt; parameters obtained from a spline-based nonlinear diffeomorphism between training magnetic resonance and ultrasound &lt;span class=&quot;snippet&quot;&gt;prostate&lt;/span&gt; images. The &lt;span class=&quot;snippet&quot;&gt;deformation&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;models&lt;/span&gt; built from the principal eigen-modes of the &lt;span class=&quot;snippet&quot;&gt;clusters&lt;/span&gt; are then applied on a test magnetic resonance image &lt;span class=&quot;snippet&quot;&gt;to&lt;/span&gt; register with the test ultrasound &lt;span class=&quot;snippet&quot;&gt;prostate&lt;/span&gt; image. The &lt;span class=&quot;snippet&quot;&gt;deformation&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;model&lt;/span&gt; with the least &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; error is finally chosen as the optimal &lt;span class=&quot;snippet&quot;&gt;model&lt;/span&gt; for deformable &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt;. The rationale behind &lt;span class=&quot;snippet&quot;&gt;modeling&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;deformations&lt;/span&gt; is &lt;span class=&quot;snippet&quot;&gt;to&lt;/span&gt; achieve &lt;span class=&quot;snippet&quot;&gt;fast&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;multimodal&lt;/span&gt; &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; of &lt;span class=&quot;snippet&quot;&gt;prostate&lt;/span&gt; images while maintaining &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; accuracies which is otherwise computationally expensive. The method is validated for 25 patients each with a pair of corresponding magnetic resonance and ultrasound images in a leave-one-out validation framework. The average &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; accuracies i.e. Dice similarity coefficient of 0.927 ± 0.025, 95% Hausdorff distance of 5.14 ± 3.67 mm and target &lt;span class=&quot;snippet&quot;&gt;registration&lt;/span&gt; error of 2.44 ± 1.17 mm are obtained by our method with a speed-up in computation time by 98% when compared &lt;span class=&quot;snippet&quot;&gt;to&lt;/span&gt; Mitra et al. [7].&lt;/p&gt;&lt;/div&gt;&lt;p&gt;&amp;nbsp;&lt;/p&gt;</style></abstract><work-type><style face="normal" font="default" size="100%">Conference paper</style></work-type><accession-num><style face="normal" font="default" size="100%">13325059</style></accession-num></record></records></xml>