The complexity involved in the visualization of digital cardiac images has prompted some cardiologists to explore consumer 3D graphics cards for use in processing high-resolution, multimodality cardiac imaging data.
The complexity involved in the visualization of digital cardiac images has prompted some cardiologists to explore consumer 3D graphics cards for use in processing high-resolution, multimodality cardiac imaging data.
A new study documents how researchers at the University of California, Los Angeles and the University of Western Ontario implement a real-time processing platform, using hardware-accelerated trilinear filtering, 2D and 3D texturing, and the programmable texture and vertex shader capabilities of 3D graphics cards (Comput Med Imaging Graph. 2005 Sep;29(6):463 75).
"We're exploiting the latest advances in video/gaming technology in medical imaging, allowing dramatic increases in speed and quality for 3D visualizations of multimodal cardiac images," said Piotr Slomka, Ph.D., an associate professor of medicine at the David Geffen School of Medicine, UCLA.
Detection of coronary heart disease often involves multiple modality examinations: SPECT, PET, cardiac CT, cardiac MRI, and echocardiography.
Efficient, interactive visualization of cardiac imaging in a uniform fashion has been a challenge due to the dynamic, multiframe character of the data, distinct character of the images obtained by different modalities, presence of extracardiac structures, and data sets often exceeding 100 MB.
Previously, such dynamic volume rendering abilities were limited to high-end, expensive 3D graphics workstations. But new consumer graphics processors such as the Radeon 9700 Pro (from ATI, Ontario, Canada) and the Geforce FX (from Nvidia, Santa Clara, CA) provide similar rendering power to desktop PCs, Slomka said.
Using the new video components, the researchers demonstrated the implementation of accelerated, high-quality techniques for interactive 4D volume visualization of 16-bit data, coronary angiography, MRI cine data, fused multimodality data, and functional SPECT and PET data.
The research showed, for example, that the Radeon 9700 Pro video card with 128 MB of texture memory can render a 512 x 512 x 128 cardiac CT scan at 0.9 to 60 frames per second, depending on the size of the volume-of-interest displayed and the detail at which the volume is rendered.
"These techniques allow physicians to efficiently visualize multimodality 4D cardiac scans in real-time using inexpensive PC platforms with the latest 3D graphics cards," Slomka said.
MRI-Based AI Radiomics Model Offers 'Robust' Prediction of Perineural Invasion in Prostate Cancer
July 26th 2024A model that combines MRI-based deep learning radiomics and clinical factors demonstrated an 84.8 percent ROC AUC and a 92.6 percent precision-recall AUC for predicting perineural invasion in prostate cancer cases.
Breast MRI Study Examines Common Factors with False Negatives and False Positives
July 24th 2024The absence of ipsilateral breast hypervascularity is three times more likely to be associated with false-negative findings on breast MRI and non-mass enhancement lesions have a 4.5-fold likelihood of being linked to false-positive results, according to new research.
Can Polyenergetic Reconstruction Help Resolve Streak Artifacts in Photon Counting CT?
July 22nd 2024New research looking at photon-counting computed tomography (PCCT) demonstrated significantly reduced variation and tracheal air density attenuation with polyenergetic reconstruction in contrast to monoenergetic reconstruction on chest CT.