Faculty Bibliography

PURPOSE: To compare power Doppler sonography with color Doppler sonography in the depiction of normal intrarenal vasculature. MATERIALS AND METHODS: Twenty kidneys in 10 healthy, nonobese subjects were studied with color Doppler and power Doppler sonography at 3.5 MHz and a pulse repetition frequency of 800 Hz. The color gain was increased until "noise" first became perceptible. To standardize comparison, an ultrasound phantom was scanned with both techniques at the mean depth of each kidney. RESULTS: Power Doppler sonography demonstrated a diffuse "blush" of either the entire or almost the entire cortex in 18 of 20 kidneys and in approximately half the cortex of the two other kidneys. Color Doppler sonography did not depict a cortical blush in any kidney. In addition, subjective comparison showed that power Doppler sonography demonstrated the intrarenal vasculature better than color Doppler sonography in all kidneys. CONCLUSION: Power Doppler sonography was superior to color Doppler sonography in the demonstration of normal intrarenal vasculature.

OBJECTIVE: Power Doppler sonography is a new technique that offers extended dynamic range over that provided by conventional color Doppler imaging, thereby facilitating measurement of tissue perfusion. We evaluated the efficacy of power Doppler sonography in depicting soft-tissue hyperemia in musculoskeletal inflammatory conditions. SUBJECTS AND METHODS: Twenty-three patients with predominantly unifocal musculoskeletal symptoms were evaluated with conventional gray-scale imaging and power Doppler sonography. The shoulder was evaluated in 14 patients, the elbow in three, and various other sites in six. For comparison purposes, the contralateral asymptomatic joint or site was examined in 17 cases. Fluid collections were aspirated in seven patients. RESULTS: Soft-tissue hyperemia was seen on power Doppler sonograms at the symptomatic site(s) in 22 of 23 cases studied, with an appearance ranging from frank tissue blush to large, isolated peritendinous/peribursal vessels. Findings associated with hyperemia included rotator cuff tendinitis and/or tear, bursitis, and symptomatic wrist ganglia. Findings on power Doppler sonograms were normal in one patient who had a small, sterile hip joint effusion, and findings on subsequent scintigrams, pelvic radiographs, and CT scans were normal. CONCLUSION: Power Doppler sonography consistently shows hyperperfusion associated with musculoskeletal inflammatory disease. As such, it represents a useful adjunct to gray-scale sonography in these settings, particularly as findings on conventional sonograms in inflammatory conditions can be nonspecific.

Most radiologists do not use texture information contained in the trabecular patterns of hand radiographs to diagnose erosive changes and demineralization due to systemic inflammatory diseases that affect the skeletal system. However, high-resolution digitization achievable by a laser digitizer now makes it possible to access texture information that may not be perceived visually. We are studying the feasibility of computer-assisted early detection of these processes with particular attention to patients with hyperparathyroidism. In this paper the methods used to extract a region of interest (ROI) for texture analysis are discussed. The techniques include multiresolution sensing, automatic adaptive thresholding, detection of orientation angle, and projection taken perpendicular to the line of least second moment. The methods were tested on a database of 50 pairs of hand radiographs. We segmented the middle and the index fingers with an average success rate of 83% per hand. For the segmented finger strips, we located ROIs on both the middle and the proximal phalanges correctly over 84% of the times. Texture information was collected in the form of a concurrence matrix within the ROI. This study is a prelude to evaluating the correlation between classification based on texture analysis and diagnosis made by experienced radiologists.

Recent studies have investigated the detection of significant arterial stenoses through identification of the "pulsus tardus" phenomenon in Doppler waveforms obtained distal to the stenosis. The etiology of this phenomenon, however, has not yet been determined. Using an in vitro model based upon an electrical circuit analogy, the authors had as their objective to determine whether the compliance of the vessel wall immediately distal to a stenosis, in conjunction with the stenosis, is the cause of pulsus tardus. For a constant stenosis, it was found that the degree of pulsus tardus, as quantitated by the acceleration index, increased as the compliance of the poststenotic segment increased. It is concluded that pulsus tardus distal to an arterial stenosis is likely due to the compliance of the normally distensible artery, in conjunction with the stenosis. Pathological conditions that alter the compliance of the poststenotic segment may affect the degree of pulsus tardus, perhaps limiting its usefulness for upstream stenosis detection.

OBJECTIVE: The purpose of this study was twofold: (1) to determine the characteristic MR appearance of the xanthomatous tendons of heterozygous familial hypercholesterolemia and (2) to determine which of two imaging techniques, high-frequency linear-array sonography or MR imaging, is better for detection of xanthomas. SUBJECTS AND METHODS: Sonography and MR imaging were performed to evaluate the Achilles tendons in 10 patients (20 tendons) with heterozygous familial hypercholesterolemia. For sonography, 7.5- (nine patients) and 10.0- (one patient) MHz transducers were used. T1-weighted, proton density-weighted, and T2-weighted MR images were obtained at 0.5 T (four patients) and 1.5 T (six patients), including fat-suppressed (six patients) and water-suppressed (one patient) T1-weighted images at 1.5 T. Tendon abnormalities detected with both techniques were noted, and the results were compared. RESULTS: In all tendons, sonograms showed focal hypoechoic lesions compatible with xanthomas. MR images did not show focal lesions. Instead, all pulse sequences showed a diffuse speckled or reticulated pattern or both on axial images. This speckled or reticulated appearance was more obvious on fat-suppressed T1-weighted images and much less evident on water-suppressed T1-weighted images. Contrast resolution was subjectively better on sonograms than on MR images in all cases. CONCLUSION: The speckled or reticulated appearance is a characteristic, if not pathognomonic, MR feature of xanthomatous tendons and probably is due to edema or inflammation, not intratendinous lipid. However, localized lipid deposits detected on sonograms are more readily quantified than are the lesions seen on MR images. Therefore, we think sonography, rather than MR imaging as performed in our study, is the technique of choice for detecting xanthomas.

PURPOSE: The authors present a preliminary report to demonstrate a new color Doppler (CD) ultrasonography (US) technique called power Doppler (PD), which displays the total integrated Doppler power in color, and to compare PD with CD imaging, which generally displays an estimate of the mean Doppler frequency shift. MATERIALS AND METHODS: Two standard commercial US scanners that encode the integrated power in the Doppler signal in color were used to demonstrate PD. A standard nonflow-containing US phantom, a normal right kidney, and a torsive and normal contralateral testis were scanned in the power mode. In the phantom and kidney, results with CD and PD were directly compared. RESULTS: PD does not alias, is relatively angle independent, and displays background noise in a way that increases the usable dynamic range of a US scanner. This extended dynamic range should increase machine sensitivity and may demonstrate increased flow in certain circumstances. CONCLUSION: PD is a new CD imaging imaging mode that might be superior to CD in some cases.

PURPOSE: To determine, with hydrodynamic experiments, the true cause of pulsus tardus, a Doppler waveform alteration that often occurs distal to an arterial stenosis. MATERIALS AND METHODS: A flow phantom was constructed with a pulsatile pump, interchangeable stenoses of varying degree, and interchangeable poststenotic segments of tubing with various degrees of compliance. With the transstenotic pressure drop held constant for each stenosis, Doppler waveforms were obtained before and after stenosis, while the degree of poststenotic vessel compliance was varied. RESULTS: The degree of pulsus tardus increased as the compliance of the poststenotic segment of vessel increased, independent of the transstenotic pressure drop. CONCLUSION: Poststenotic pulsus tardus is caused by the compliance of the poststenotic vessel wall in conjunction with the stenosis, which produces the tardus effect by damping the high-frequency components of the arterial waveform. This information will allow prediction of conditions that may produce false-positive or false-negative results when the tardus phenomenon is used to predict substantial upstream stenosis.

We have previously described a technique to quantify surface fibrillatory changes in osteoarthritic articular cartilage. In that study, the angular distribution of the scattered acoustic field from an insonifying source directly related to the distribution of surface fibrillatory changes. In the current study, we demonstrate a more sensitive method to quantify surface roughness, the effect of global surface curvature in estimating surface roughness and the utility of using focused transducers in circumventing this potential problem for in vivo work. Phantoms composed of acrylic rods with and without sandpaper grit (about 15 to 72 microns, mean particle size) applied to the surface were scanned. A more robust angular scattering technique to measure the angle dependent data was employed, in which the integrated squared pressure amplitude over a finite time window (mean power) was measured as a function of incident acoustic angle for varying surface roughnesses and radii of curvature. We show that the potential dynamic range for making roughness discriminations diminishes with decreasing radius of curvature of the acrylic rod phantoms using an unfocused transducer. This effect is minimized with use of a focused transducer. Roughness effects are most evident at sufficiently large angles where incoherent scattering dominates. We conclude that the roughness of cylindrically curved surfaces can be quantitatively assessed using a focused ultrasound beam at sufficiently large incident angles, given that the focal spot size is sufficiently smaller than the radius of curvature of the surface.