Elevation direction deconvolution in three-dimensional ultrasound imaging.

One-dimensional (1-D) linear transducer arrays can be used for three-dimensional (3-D) ultrasound image acquisition. However, the relatively low spatial resolution of these arrays in the elevation direction results in blurry 3-D images. Here, the authors introduce an elevation direction deconvolution (EDD) method that increases the spatial resolution of 3-D ultrasound images in the elevation direction. EDD is based on a deconvolution technique called power spectrum equalization. To evaluate the authors' method, Cartesian volumes were reconstructed with and without EDD from a series of two-dimensional (2-D) images of phantoms. Using these reconstructed volumes, the authors first evaluated the effect of EDD on elevation resolution by computing the full-width-at-quarter-maximum (FWQM) of peaks along lines of constant depth. They then evaluated the effect of EDD on the accuracy of volume calculation by computing the phantom's volumes. EDD decreased the FWQM of the peaks on elevation lines by an average of 17%; however, EDD did not significantly alter the accuracy of volume calculation. It is concluded that EDD can increase the spatial resolution in the elevation direction in 3-D ultrasound images and that EDD may improve the accuracy of volume calculation if a more consistent edge detection method is used.

Respiratory compensation in projection imaging using a magnification and displacement model.

Respiratory motion during the collection of computed tomography (CT) projections generates structured artifacts and a loss of resolution that can render the scans unusable. This motion is problematic in scans of those patients who cannot suspend respiration, such as the very young or intubated patients. Here, the authors present an algorithm that can be used to reduce motion artifacts in CT scans caused by respiration. An approximate model for the effect of respiration is that the object cross section under interrogation experiences time-varying magnification and displacement along two axes. Using this model an exact filtered backprojection algorithm is derived for the case of parallel projections. The result is extended to generate an approximate reconstruction formula for fan-beam projections. Computer simulations and scans of phantoms on a commercial CT scanner validate the new reconstruction algorithms for parallel and fan-beam projections. Significant reduction in respiratory artifacts is demonstrated clinically when the motion model is satisfied. The method can be applied to projection data used in CT, single photon emission computed tomography (SPECT), positron emission tomography (PET), and magnetic resonance imaging (MRI).

Correction of computed tomography motion artifacts using pixel-specific back-projection.

Cardiac and respiratory motion can cause artifacts in computed tomography scans of the chest. The authors describe a new method for reducing these artifacts called pixel-specific back-projection (PSBP). PSBP reduces artifacts caused by in-plane motion by reconstructing each pixel in a frame of reference that moves with the in-plane motion in the volume being scanned. The motion of the frame of reference is specified by constructing maps that describe the motion of each pixel in the image at the time each projection was measured; these maps are based on measurements of the in-plane motion. PSBP has been tested in computer simulations and with volunteer data. In computer simulations, PSBP removed the structured artifacts caused by motion. In scans of two volunteers, PSBP reduced doubling and streaking in chest scans to a level that made the images clinically useful. PSBP corrections of liver scans were less satisfactory because the motion of the liver is predominantly superior-inferior (S-I). PSBP uses a unique set of motion parameters to describe the motion at each point in the chest as opposed to requiring that the motion be described by a single set of parameters. Therefore, PSBP may be more useful in correcting clinical scans than are other correction techniques previously described.

Three-dimensional ultrasonic angiography using power-mode Doppler.

To visualize the vascular anatomy of parenchymal organs, we have developed a system for producing three-dimensional ultrasonic angiograms (3D USA) from a series of two-dimensional power-mode Doppler ultrasound (PDU) scans. PDU scans were acquired using a commercial scanner and image-registration hardware. Two-dimensional images were digitized, and specially designed software reconstructed 3D volumes and displayed volume-rendered images. The geometric accuracy of our system was assessed by scanning a flow phantom constructed from tubing. The system was tested on patients by scanning native and transplanted kidneys, and placentas. Three-dimensional images of the phantoms depicted the spatial relationships between flow within the tubing segments and contained less than 1 mm of geometric distortion. Three-dimensional images of the kidney and placenta demonstrated that spatial relationships between vasculature structures could be visualized with 3D USA. Applications of this new technique include analysis of vascular anatomy and the potential assessment of organ perfusion.

Evaluation of electron beam CT coronary angiography in healthy subjects.

OBJECTIVE: This study evaluated the performance characteristics of electrocardiographically triggered, contrast-enhanced electron beam CT (EBCT) in defining the coronary artery lumen in healthy subjects. SUBJECTS AND METHODS: The coronary arteries of 11 healthy young men (mean age, 24 years old) were evaluated by contrast-enhanced EBCT. Measured parameters included degree of luminal enhancement, intravascular contrast-to-noise ratio, apparent luminal diameter, and length of continuously visualized lumen (100-H threshold for diameter and length measurements). RESULTS: Aortic blood pool attenuation was 44 +/- 5 H (mean +/- SD) before and 278 +/- 35 H after IV injection of contrast material. Contrast-to-noise ratios ranged from a high of 10.0 +/- 2.6 in the proximal right coronary artery to a low of 3.2 +/- 2.7 in the distal left circumflex artery, decreasing from proximal to distal within each vessel. Apparent luminal diameters were as follows: left main coronary artery, 4.5 +/- 0.6 mm; left anterior descending artery, 3.7 +/- 0.5 mm; left circumflex artery, 2.9 +/- 0.6 mm; and right coronary artery, 3.5 +/- 0.5 mm. The mean lengths of visualized lumina were as follows: left main coronary artery, 10 +/- 4 mm; left anterior descending artery, 65 +/- 26 mm; left circumflex artery, 45 +/- 20 mm; and right coronary artery, 58 +/- 24 mm. CONCLUSION: EBCT angiography can reveal the lumen of long segments of the major coronary arteries.

Predictive respiratory gating: a new method to reduce motion artifacts on CT scans.

PURPOSE: To evaluate a gating system, called predictive respiratory gating (PRG), that reduces motion-induced artifacts on computed tomographic (CT) scans of patients who cannot suspend respiration. MATERIALS AND METHODS: PRG uses a respiration monitor and a new algorithm to predict when a motionless period is about to occur. It automatically starts scanning so the scan is temporally centered around the motionless period at end inspiration or end expiration. To demonstrate PRG, CT was performed on a motion phantom and a quietly breathing volunteer with and without gating. RESULTS: Scans of the phantom obtained with PRG contained less motion-induced streaking and blurring than did scans acquired without PRG. Scans of the volunteer gated at end expiration contained significantly less artifact than nongated scans (P < .03). CONCLUSION: PRG reduced motion artifact on scans of a spontaneously breathing volunteer. PRG may be able to reduce motion artifacts on scans of patients unable to suspend respiration.

Minimum scan speeds for suppression of motion artifacts in CT.

Cardiac and ventilatory motions cause artifacts at chest computed tomography (CT). To determine how short the scan times on third-generation units must be to avoid such artifacts, motion was measured with fast and ultrafast CT scans. Minimum detectable motion was then determined. The longest scan time that avoided a barely perceptible artifact was calculated by dividing the minimum detectable motion by the peak physiologic velocity. The posterior left ventricular wall moved at a maximum velocity of 52.5 mm/sec, necessitating a scan time of 19.1 msec or less to avoid artifact. Lung vessels near the heart moved at 40.5 mm/sec for a scan time of 24.7 msec or less. During quiet breathing, pulmonary vessels moved at 10.7 mm/sec for a scan time of 93.5 msec or less. The authors conclude that the shortest scan time on third-generation units (0.6 second) cannot prevent all artifacts arising from motion in the chest. Even ultrafast scan times (50 msec) are not short enough to eliminate artifacts on these units. Thus, reduction of motion artifacts will require techniques other than fast scanning.