We present a novel and clinically useful method to automatically determine the regions that carry cardiac and respiratory motion information directly from standard mono-plane X-ray fluoroscopy images. We demonstrate the application of our method for the purposes of retrospective cardiac and respiratory gating of X-ray images. Validation is performed on five mono-plane imaging sequences comprising a total of 284 frames from five patients undergoing radiofrequency ablation for the treatment of atrial fibrillation. We established end-inspiration, end-expiration and systolic gating with success rates of 100%, 100% and 95.3%, respectively. This technique is useful for retrospective gating of X-ray images and, unlike many previously proposed techniques, does not require specific catheters to be visible and works without any knowledge of catheter geometry.
|Title of host publication||Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges|
|Subtitle of host publication||STACOM 2013|
|Editors||Oscar Camara, Tommaso Mansi, Mihaela Pop, Kawal Rhode, Maxime Sermesant, Alistair Young|
|Number of pages||9|
|Publication status||Published - 26 Sep 2014|
|Name||Lecture Notes in Computer Science|