SCIENCE

In the July issue of the Journal of American College of Cardiology (JACC), Cardiovascular Imaging, leading medical experts indicate true fidelity 4D and 5D functional analytics are now achievable. Using supercomputing technology with deformable registration from Ziosoft, Inc., non-rigid (deformable) registration was performed on a series of computed cardiac tomographic (CT) images from a patient that underwent coronary CT angiogram. The authors noted a reduction in noise which improves motion coherence and enables functional analytics.

According to the authors, using multi-phase data sets, the CT voxel-to-voxel , 4D motion was calculated for estimating velocity and displayed as a parametric map overlaid on the volume rendered surface. This provides 5D imaging of the heart, or unimodal-multidimensional processing.

"The authors indicate that the convergence of multi-modality imaging and Ziosoft's PhyZiodynamic supercomputing technology could finally enable this long sought-after Holy Grail," said Rob Royea, president and COO of Ziosoft. "We believe this article brings to light the power of supercomputing in enabling deformable registration for accurate depiction of multidimensional imaging."

Co-authored by Partho Sengupta, associate professor of medicine, Cardiology Division at Mt. Sinai School of Medicine, Thomas Marwick, MD, PhD, section head of Cardiovascular Imaging for the Cleveland Clinic, and Jagat Narula, MD, PhD, professor of medicine and director of Cardiovascular Imaging, Mt. Sinai in New York, the article, "Adding Dimensions to Unimodal Cardiac Images," provides a definition of the various dimensions. For volumetric data sets (pixels become voxels), the first three dimensions are spatial while time scale (motion of 3D) is referred to as the fourth dimension. Additional functional data, such as wall motion or deformation or molecular imaging data for assessing biological and metabolic pathways, can be integrated with the four dimensional images, evolving as the fifth dimension.

The promise of CT and other imaging modalities has been to provide accurate and realistic images in multiple dimensions. The article indicates that using current imaging modalities, images can be augmented into additional dimensions using supercomputing deformable registration and functional analytics. To achieve the fifth dimension, true fidelity 4D is essential and supercomputing provides this capability."

According to the authors, the four dimensional images obtained with a single imaging modality are first processed for improving motion coherence with noise reduction. The individual voxels are then tracked through the cardiac cycle for deriving functional parameters like velocity. Parametric images obtained with velocity-encoded sequences are then combined with images obtained with dynamic cine sequences to provide myocardial velocity information superimposed on anatomic information.

"Additional studies are and will be underway to further determine information parameters and equate threshold measurement ranges to varying organ functionality," said Royea. "This could lead to many patient benefits, such as replacing invasive electrophysiology procedures or the accurate placement of defibrillator leads in viable tissue. I believe we are just seeing the tip of the iceberg in how supercomputing functional analytics could significantly impact the way we practice medicine."