Clinical Advancements in Angiographic Imaging

Figure 1: A simulation of the way image subtraction can be used to quantify motion. In (a-c), the high motion phase in mid-systole results in substantial movement between the objects in the image. Since any object that is in the same place will subtract away and any object that moves will not, the resulting difference image in (c) shows far more bright areas than the corresponding difference image in (f) which corresponds to a relatively low motion phase in late diastole. Figure 2: A motion map is created by adding up the intensity of all the difference images for all phases and longitudinal locations (top image). The graph shows the amount of motion vs cardiac phase. phaseXact reconstructs at the point of least cardiac motion. Since the phase with minimal cardiac motion is a function of the patient’s cardiac morphology, physiology, and heart rate at the time of the scan, the optimal reconstruction phase must be determined empirically on a patientby-patient, scan-by-scan basis. Patients with lower heart rates typically have their optimal phase in late diastole while in patients with higher heart rates, the phase that minimizes motion is often in systole1,2 . While these phases are always patient dependent, the best diastolic phase is usually found somewhere between 60% and 80% of the R-R interval and the best systolic phase is found between 25% and 50% R-R3,4. For the most part, the scientiﬁc studies have analyzed patients with steady heart rates. Unfortunately, this does not always reﬂect real-world scenarios. There is little published data on how heart rate variations affect optimal phase selection and choosing the best phase to analyze in these patients can be challenging4. Ultimately, regardless of whether the heart rate is steady or variable, choosing the optimal phase from the broad range of choices is a critical step towards reaching a diagnosis quickly and accurately. With the potentially severe consequences of interpreting images reconstructed at a suboptimal phase, a considerable amount of clinical time and effort is devoted to ﬁ nding the optimal phase. The conventional method of determining this phase is to assume that the best phase is located between 60-80%, and span that potion of the R-R interval with 5 reconstructions spaced at 5% intervals. The practical implication of this approach is that multiple phases are reconstructed and archived, even if the majority of the reconstructed images will never be used for postprocessing or image interpretation. For example, in a case with 300 images per phase, ﬁve sets of 300 image volumes are reconstructed, transferred, and archived for interpretation. In the most inefﬁcient system, these 1500 images would then be sent to PACS, monopolizing bandwidth and the storage capacity of the archival system. Given the growing concern regarding the ability to manipulate and store large data sets, this inefﬁcient use of data presents an even larger problem as the volume of cardiac CT angiography studies continues to grow. This workﬂow challenge is further compounded by the fact that the imaging professional must search these data sets to determine the one phase from which analyses will begin. If none of the original reconstructed phases proves adequate, additional reconstructions are required which takes up more time and bandwidth. This iterative combination of a “brute force” and “hit-or-miss” approach will ultimately yield adequate diagnostic data sets, but at the cost of physician interpretation time, network bandwidth and image storage. Oc tober 2 0 0 7 A public ation produced by Toshiba Medic al Sys t ems 5

Table of Contents Feed for the Digital Edition of Clinical Advancements in Angiographic Imaging

Contents PhaseXact™: Automatic Phase Cardiac Reconstruction Giant Internal Carotid Aneurysm Non-Contrast SSFP Time-SLIP Imaging: Renal Artery Stenosis Springhill Medical Center: A Partnership to Improve Patient Safety Anterior Cerebral Artery Aneurysm Right Coronary Artery (RCA) lesion with SUREPlaque analysis Neurovascular CTA: When every second counts

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.