Published online October 14, 2002, 10.1148/rg.e8
(Radiographics. 2003;23:e8-e8.)
© RSNA, 2003
Real-Time Four-dimensional Imaging of the Heart with Multi–Detector Row CT1
Kimiaki Saito, MD,
Motoaki Saito, MD,
Shuhei Komatu, MD and
Kuni Ohtomo, MD
1 From the Department of Radiology, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. Presented as a scientific exhibit at the 2001 RSNA scientific assembly. Received March 31, 2002, revision requested May 23, revision received and accepted August 30. Address correspondence to K.S. (e-mail: kimiakis{at}horae.dti.ne.jp).
 |
Abstract
|
|---|
An interactive four-dimensional (4D) visualizing system for the heart was developed by the authors. The system realizes high-resolution three-dimensional (3D) imaging with temporal resolution in a beating heart by using eight or more data sets reconstructed from multi–detector row computed tomography (MDCT) with a retrospective electrocardiograph-gated reconstruction algorithm. The motion of heart walls, papillary muscles, septa, and valves can now be observed in 4D multiplanar reformations (MPRs), as with sonography, while coronary arteries, coronary sinuses, and cardiac veins can be analyzed during the optimal phase in 4D volume-rendering images, as with angiography. All parameters such as window width, window level, field of view, panning, tilt, thresholds, opacity, color, and segmentation function are completely interactive in 4D imaging. Two longitudinal views and one latitudinal view of a heart can be simultaneously visualized in the three relative 4D MPR views. These newly developed capabilities in viewing both 3D volume and temporal resolution data, functional data, and even multiphase data with registration add considerable diagnostic potential. The advent of this real-time 4D visualizing system has enhanced the capabilities of MDCT.
© RSNA, 2003
Index Terms: Heart, CT, 51.12113, 51.12115, 51.12116, 51.12117, 51.12118, 54.12113, 54.12115, 54.12116, 54.12117, 54.12118
|
Introduction
|
|---|
The latest implementations of multi–detector row computed tomography (MDCT) can acquire data in moving organs such as a heart and generate three-dimensional (3D) data sets in each phase with a retrospective electrocardiograph (ECG)-gated reconstruction algorithm (1–9). Noninvasive coronary CT angiography has been performed and assessed with 3D data sets, but we reviewed one 3D data set at each phase or generated the 3D data sets as a movie to show movement; the movie, however, was not interactive (3–8). To our knowledge, a system in which the viewer could interactively handle 3D data sets with temporal resolution has not been reported. We defined that as four-dimensional (4D) imaging. To realize 4D imaging, we developed a 4D viewer with a very fast image processor engine. The system can realize real-time and interactive 4D imaging as 4D volume rendering, 4D maximum-intensity projection (MIP), and 4D multiplanar reformation (MPR). We report here the advantages and applications of 4D imaging.
|
Materials and Methods
|
|---|
MDCT scans of an entire heart were obtained with a 2-mm section thickness and a 1-mm interval with an Aquilion four–detector row CT scanner (Toshiba Medical, Tokyo, Japan). Images were acquired within a single breath hold after injection of a contrast medium (Omnipaque350; Daiichi Pharmaceutical, Tokyo, Japan; injection rate, 2.5 mL/sec; volume, 120 mL). We used the bolus tracking system Real-Prep software included in the CT scanner to start scanning at the optimal time. By simultaneous recording of the ECG, images are reconstructed with a retrospective ECG-gated reconstruction algorithm to generate eight or more 3D data sets from raw data at each phase of the cardiac cycle (Figure). The number of total sections is more than 1,000. To view 4D images interactively, we needed a powerful viewer with very fast image processing capability and more than 1 GByte of memory. Therefore, we selected the Aquarius Workstation featuring VolumePro (TeraRecon, San Mateo, Calif). We collaborated with the processor design group and the application development team of TeraRecon and developed a real-time 4D viewer with applications. The viewer can handle multiphase data sets and show multiple images with multiple protocols. We interactively viewed 4D images of a heart while changing all parameters, including phase parameters.
View larger version (46K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure. Diagram shows retrospective ECG-gated reconstruction. One image is generated from the raw data from three heartbeats. The data acquisition window is about 100 msec in the ideal case. We can get 3D data sets from any phase by selecting part of the raw data sets positioned at any R-R interval.
|
|
|
Results
|
|---|
Preview of 4D Imaging
The heart data sets were large, and it took about a minute to load the 3D data sets and convert them into isotropic data sets with z-axis interpolation. We viewed the original axial images, changing window width and level, location, and phase in a preview window before loading the data sets (Movie 1). The viewer showed 4D axial images at 10 or more (up to about 20) frames per second, depending on parameters used. We checked the order of cardiac cycle phases and looked for any errors due to importing of images.
4D Volume Rendering
After all data sets selected in the preview window were loaded, the 4D volume-rendering image appeared (Movies 2, 3). We interactively changed thresholds, opacity, color, tilt, panning, and magnification. We observed coronary arteries, calcium plaques, coronary sinuses, and surfaces of moving cardiac walls. We could adjust the viewing speed, and the maximum frame rate was 10 or more (up to about 20) frames per second. For better observation of the heart, we eliminated the ribs and spine with the segmentation function.
4D MPR Viewing
We changed the protocol to 4D MPR viewing and easily observed motion of the heart walls, papillary muscles, ventricular septum, valves, and masses (Movies 4–6). We interactively changed window width and level, rotation, tilt for oblique views, depth, panning, and magnification. We could adjust the viewing speed, and the maximum frame rate was was 10 or more (up to about 20) frames per second. Rotation and tilt parameters were changed at the viewer. By panning, we also established the area of interest at the viewer, and by using the rotation and tilt functions, we could observe any 4D MPR view that included the area of interest.
4D MIP Viewing
We sometimes lost coronary arteries on the 4D MPR views because they were curved and moving. Therefore, we observed them on the 4D slab MIP (Movie 7). We interactively changed window width and level, rotation, tilt, slab thickness, depth, panning, and magnification. We could adjust the viewing speed, and the maximum frame rate was was 10 or more (up to about 20) frames per second. We selected the optimal phase and angle for vessel analysis.
Synchronized 4D Imaging
We observed movement of objects on 4D images of a heart with temporal resolution. Four-dimensional imaging proved useful for heart examinations; however, it was difficult to show the objects we were interested in at any time on one view because these objects were moving. Therefore, we developed synchronized real-time 4D imaging that shows four or more views (eg, axial, coronal, sagittal 4D MPR and 4D volume rendering) simultaneously (Movies 8, 9). With this application, we could observe the relevant objects more easily on multiple 4D views.
|
Discussion
|
|---|
The number of detector rows and the scan speed available in MDCT are improving. We can easily obtain multiphase data sets such as liver dynamic data sets, volume-perfusion data sets, and multiphase heart data sets. Each area of interest is viewed in the optimal phase and protocol. It was reported that for the left main and left anterior descending coronary arteries, optimal image quality was usually achieved if the data aquisition was centered at 70% or 80% of the R-R interval; whereas for the left circumflex and right coronary arteries, the optimal position of the data aquisition was at 50% of the R-R interval (8). It is possible that motion artifacts can hide small objects in some phases. Therefore, we must review all data sets with a viewer for diagnosis. The typical viewer can handle only one or at most a few data sets and cannot show temporal resolution. It is not easy to understand the movement or observe the area of interest clearly on a few static images.
The real-time 4D image viewer can interactively show the areas of interest with temporal resolution. We used eight or more data sets to view 4D images of the heart. The 4D views showed motion of the heart walls, papillary muscles, ventricular septum, valves, and masses and provided the optimal images of coronary arteries, calcium plaques, and coronary sinuses for diagnosis. We analyzed the areas of interest at the optimal phase. This noninvasive coronary CT angiography is useful for diagnosis.
The data acquisition window is more than 100 msec, and motion artifact increases if a patient has arrhythmia during scanning. In the worst case, we cannot acquire the images with the retrospective ECG-gated reconstruction algorithm. The exposure dose is about three times greater than that of lung scanning. Therefore, the indications for noninvasive coronary CT angiography require additional study and discussion.
The performance of 256-row-detector CT and panel-detector CT allows acquisition of multiphase data sets in a target that does not have a cycle and moves irregularly. It was stated at a 3D CT seminar in Japan that the total number of images generated will be more than 5,000. Our viewer cannot handle 5,000 section images. We are, therefore, developing a new viewer that can handle 16,000 section images to show their temporal resolution. We believe that viewers that can easily handle multiphase data sets and perform 4D imaging will become standard tools for diagnosis.
In the present study of the heart, the temporal resolution of multiphase data sets generated by MDCT were interactively viewed with a 4D image viewer. We could clearly observe the motion of heart walls, papillary muscles, the ventricular septum, valves, and masses and the optimal images of coronary arteries, calcium plaques, and coronary sinuses for diagnosis.
|
Acknowledgments:
|
|---|
We thank Robert Taylor, PhD, Akio Iwase, PhD, and Keiji Ito, BE, for their valuable contributions to the preparation of this manuscript.
|
Footnotes
|
|---|
Abbreviations: ECG = electrocardiograph, MDCT = multi–detector row computed tomography, MIP = maximum-intensity projection, MPR = multiplanar reformation, 3D = three-dimensional, 4D = four-dimensional.
|
References
|
|---|
- Ohnesorge B, Flohr T, Becker C, et al. Cardiac imaging by means of electrocardiographically gated multisection spiral CT: initial experience. Radiology 2000; 217:564-571.[Abstract/Free Full Text]
- Kachelriess M, Ulzheimer S, Kalender WA. ECG-correlated image reconstruction from subsecond multi-slice spiral CT scans of the heart. Med Phys 2000; 27:1881-1902.[CrossRef][Medline]
- Schroeder S, Kopp AF, Kuettner A, et al. Influence of heart rate on vessel visibility in noninvasive coronary angiography using new multislice computed tomography: experience in 94 patients. Clin Imaging 2002; 26:106-111.[CrossRef][Medline]
- Kopp AF, Schroeder S, Kuettner A, et al. Coronary arteries: retrospectively ECG-gated multi-detector row CT angiography with selective optimization of the image reconstruction window. Radiology 2001; 221:683-688.[Abstract/Free Full Text]
- Hong C, Becker CR, Huber A, et al. ECG-gated reconstructed multi-detector row CT coronary angiography: effect of varying trigger delay on image quality. Radiology 2001; 220:712-717.[Abstract/Free Full Text]
- Nieman K, van Ooijen P, Rensing B, Oudkerk M, de Feyter PJ. Four-dimensional cardiac imaging with multislice computed tomography. Circulation 2001; 103:E62.
- Becker CR, Ohnesorge BM, Schoepf UJ, Reiser MF. Current development of cardiac imaging with multidetector-row CT. Eur J Radiol 2000; 36:97-103.[CrossRef][Medline]
- Kachelriess M, Ulzheimer S, Kalender WA. Noninvasive coronary angiography by retrospectively ECG-gated multislice spiral CT. Circulation 2000; 102:2823-2828.[Abstract/Free Full Text]
- Horiguchi J, Nakanishi T, Ito K. Quantification of coronary artery calcium using multidetector CT and a retrospective ECG-gating reconstruction algorithm. AJR Am J Roentgenol 2001; 177:1429-1435.
This article has been cited by other articles:
|
|
|
|
 
K. Sakakura, T. Yasu, Y. Kobayashi, T. Katayama, Y. Sugawara, H. Funayama, Y. Takagi, N. Ikeda, T. Ishida, Y. Tsuruya, et al.
Noninvasive Tissue Characterization of Coronary Arterial Plaque by 16-Slice Computed Tomography in Acute Coronary Syndrome
Angiology,
March 1, 2006;
57(2):
155 - 160.
[Abstract]
[PDF]
|
|
|
|
|
|
|
J. F. Bruzzi, M. Remy-Jardin, D. Delhaye, A. Teisseire, C. Khalil, and J. Remy
When, Why, and How to Examine the Heart During Thoracic CT: Part 1, Basic Principles
Am. J. Roentgenol.,
February 1, 2006;
186(2):
324 - 332.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Rosset, L. Spadola, L. Pysher, and O. Ratib
Informatics in Radiology (infoRAD): Navigating the Fifth Dimension: Innovative Interface for Multidimensional Multimodality Image Navigation
RadioGraphics,
January 1, 2006;
26(1):
299 - 308.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
