Subsequent projects
Friedrich-Alexander-Universität
Erlangen-Nürnberg
Lehrstuhl für Informatik 5 (Musterekennung)
Stanford University
Department of Radiology - Lucas Center
Weight Bearing C-Arm CT
Today computer tomography expects the patient to lie on the table. Weight bearing 3-D imaging, for instance of the spine or the knee joint, is impossible so far. In the funded BaCaTec project we want to do research on weight bearing 3-D imaging and in the long term 3-D reconstruction of a patient in motion using robot guided C-arm systems. The project’s challenge is the combination of 3-D sensors (like the Kinect) acquiring dynamic 3-D surface information with CT reconstruction algorithms. The preliminary research on 3-D/4-D cardiac imaging and the experience with 3-D surface data processing of the FAU-Stanford-team fulfills excellent prerequisites for pursuing the planned research. Both the algorithmic expertise as well as the required acquisition hardware is available for the project. The success of the planned studies might open up a new field in 3-D/4-D CT imaging.
Primary project: Cardiac C-arm CT: 4D Heart Motion Modeling and Evaluation Study
Final Report
The usual application area for C-arm computed tomography (CT) is in the interventional suite, where typically projection images of the thorax or head are acquired using a vertical trajectory around the patient table. In this subsequent project we made use of a robotic C-arm device that was able to perform CT scans using a horizontal trajectory. This new type of trajectory allowed us to acquire 3D volumetric images of knees while the patient is in a weight-bearing position, i.e., standing upright or even in a squatting position.
In the context of this project multiple problems that arise from the combination of the new scanning trajectory as well as the weight-bearing position have been investigated and solved. Besides a thorough validation of the accuracy and reproducibility of horizontal scanning trajectories [1], the main focus has been on the correction of involuntary patient motion. The standing position resulted in increased motion artifacts in the reconstructed volumes. To improve image quality, methods have been developed to estimate the patient motion. Then, the motion estimates have been incorporated into an improved image reconstruction which significantly reduced motion artifacts in the images [2][3].
Additional to the patient motion we proposed a novel method to dynamically increase the insufficient field-of-view. Using an improved non-circular trajectory design we are now able to acquire both knee joints simultaneously [4]. Further, a novel correction for overexposure artifacts has been presented that uses simple geometric shape fitting to estimate the missing data in the saturated regions of the projection images [5].
The profound advances during this project now allow for detailed clinical studies using weight-bearing trajectories. For example, we currently investigate knee cartilage deformations using realistic standing or squatting position. A subsequent funding by the National Institute of Health (NIH Grant-Nr: R01 AR065248) further underlines the success and importance of the weight-bearing project.
References
[1] A. Maier, J.-H. Choi, A. Keil, C. Niebler, M. Sarmiento, A. Fieselmann, G. Gold, S. Delp, and R. Fahrig. Analysis of vertical and horizontal circular C-arm trajectories. In Norbert J. Pelc, Ehsan Samei, and Robert M. Nishikawa, editors, SPIE Medical Imaging, pages 796123-796123-8. International Society for Optics and Photonics, March 2011.
[2] Jang-Hwan Choi, Rebecca Fahrig, Andreas Keil, Thor F Besier, Saikat Pal, Emily J McWalter, Gary S Beaupre, and Andreas Maier. Fiducial marker-based correction for involuntary motion in weight-bearing C-arm CT scanning of knees. Part I. Numerical model-based optimization. Medical Physics, 40(9):091905, September 2013.
[3] Jang-Hwan Choi, Andreas Maier, Andreas Keil, Saikat Pal, Emily J. McWalter, Gary S. Beaupre, Garry E. Gold, and Rebecca Fahrig. Fiducial markerbased correction for involuntary motion in weight-bearing C-arm CT scanning of knees. II. Experiment. Medical Physics, 41(6):061902, June 2014.
[4] Magdalena Herbst, Frank Schebesch, Martin Berger, Jang-Hwan Choi, Rebecca Fahrig, Joachim Hornegger, and Andreas Maier. Dynamic Detector Offsets for Field of View Extension in C-Arm Computed Tomography with Application to Weight-Bearing Imaging. Medical Physics, 2015 (in press).
[5] Alexander Preuhs, Martin Berger, Yan Xia, Andreas Maier, Joachim Hornegger, and Rebecca Fahrig. Over-Exposure Correction in CT Using Optimization-Based Multiple Cylinder Fitting. In H. Handels, editor, Bildverarbeitung fur die Medizin 2015, pages 35-40, 2015.