TY - JOUR
T1 - Comparative imaging study in ultrasound, MRI, CT, and DSA using a multimodality renal artery phantom
AU - King, Deirdre M.
AU - Fagan, Andrew J.
AU - Moran, Carmel M.
AU - Browne, Jacinta E.
N1 - Funding Information:
This work was supported by the Technological Sector Research Strand 1 Scheme, HEA, and the TERs 2005 and CABS 2007 Schemes, Research Support Unit of the Dublin Institute of Technology. C.M. Moran would like to acknowledge funding from the British Heart Foundation Grant No. PG/07/107/23895. TABLE I. Weight composition of the TMM. Component Weight composition(%) Distilled water 82.97 Glycerol 11.21 Silicon carbide (400 grain) 0.53 Aluminum oxide ( 3 μ m ) 0.94 Aluminum oxide ( 0.3 μ m ) 0.88 Benzalkoniumchloride 0.46 Agar 3.0 TABLE II. Comparison of average in vivo soft tissue and measured TMM properties. Parameter Average soft tissue in vivo values TMM measured values Speed of sound ( m s − 1 ) 1540 1549 ± 5 Attenuation coefficient ( dB cm − 1 MHz − 1 ) 0.5–0.7 0.52 ± 0.03 T 1 (ms) 382–1545 1504 ± 10 T 2 (ms) 29–81 40.0 ± 0.4 Hounsfield number 20–70 58.6 ± 7.3 TABLE III. Summary of over/underestimation of stenosis (%) with US, MRI, CT, and DSA. 0% 30% 50% 70% 85% US 0 −7 a −2 −10 a −12 a MRI 2 −3 −3 −5 a −10 a CT 3 −6 a −2 −10 a −13 a DSA 6 a 3 −7 a −12 a −14 a P value 0.0045 < 0.0001 0.0110 < 0.0001 < 0.0001 a Denotes statistically significant error in the estimation. FIG. 1. Normal renal artery computer model. FIG. 2. Removal inserts used to fabricate the diseased models of the renal artery. Figure reprinted with permission from D. M. King et al ., Ultrasound Med. Biol. 36 (7), 1135–1144 (2010). Copyright October 25, 2010, by Elsevier. FIG. 3. (a) Cross-sectional schematic diagram of the fiducial markers in the phantoms. (b) Photograph showing the positioning of the glass bead markers before inclusion of the metal artery model and background TMM, where ∗ marks the location of the stenosis relative to the fiducial markers. FIG. 4. Location of the reference measurements on the metal model at position A, before stenosis: 10 mm from inlet; position B, at stenosis: 15 mm from inlet; and position C, after stenosis: 10 mm from bend. FIG. 5. Calculation of the vessel diameter in a B-mode ultrasound image. (a) Line profile marked across the lumen and (b) plot of image intensity versus distance along the chosen line profile. Figure reprinted with permission from D. M. King et al. , Ultrasound Med. Biol. 36 (7), 1135–1144 (2010). Copyright October 25, 2010, by Elsevier. FIG. 6. Fiducial markers are clearly identifiable in all the imaging modalities. (a) Ultrasound, transverse image showing cross section of the lumen and two beads at depth 49 and 57 mm on the right side of the phantom. (b) MRI showing a slice with two rows of beads. (c) Angiographic image and (d) CT showing a slice with two rows of beads. FIG. 7. B-mode images showing inlet of the renal artery flow phantom with varying degrees of stenosis (0%, 30%, 50%, 70%, and 85%). Figure reprinted with permission from D. M. King et al. , Ultrasound Med. Biol. 36 (7), 1135–1144 (2010). Copyright October 25, 2010, by Elsevier. FIG. 8. Transverse views of a phantom with a 50% stenosis at the position of (a) 10 mm after the inlet, (b) the stenosis, and (c) 10 mm before the bend. FIG. 9. MRI, CT, and unsubtracted angiography contrast images of the range of multimodality phantoms.
PY - 2011/2
Y1 - 2011/2
N2 - Purpose: A range of anatomically realistic multimodality renal artery phantoms consisting of vessels with varying degrees of stenosis was developed and evaluated using four imaging techniques currently used to detect renal artery stenosis (RAS). The spatial resolution required to visualize vascular geometry and the velocity detection performance required to adequately characterize blood flow in patients suffering from RAS are currently ill-defined, with the result that no one imaging modality has emerged as a gold standard technique for screening for this disease. Methods: The phantoms, which contained a range of stenosis values (0%, 30%, 50%, 70%, and 85%), were designed for use with ultrasound, magnetic resonance imaging, x-ray computed tomography, and x-ray digital subtraction angiography. The construction materials used were optimized with respect to their ultrasonic speed of sound and attenuation coefficient, MR relaxometry (T1, T2) properties, and Hounsfield number/x-ray attenuation coefficient, with a design capable of tolerating high-pressure pulsatile flow. Fiducial targets, incorporated into the phantoms to allow for registration of images among modalities, were chosen to minimize geometric distortions. Results: High quality distortion-free images of the phantoms with good contrast between vessel lumen, fiducial markers, and background tissue to visualize all stenoses were obtained with each modality. Quantitative assessments of the grade of stenosis revealed significant discrepancies between modalities, with each underestimating the stenosis severity for the higher-stenosed phantoms (70% and 85%) by up to 14%, with the greatest discrepancy attributable to DSA. Conclusions: The design and construction of a range of anatomically realistic renal artery phantoms containing varying degrees of stenosis is described. Images obtained using the main four diagnostic techniques used to detect RAS were free from artifacts and exhibited adequate contrast to allow for quantitative measurements of the degree of stenosis in each phantom. Such multimodality phantoms may prove useful in evaluating current and emerging US, MRI, CT, and DSA technology.
AB - Purpose: A range of anatomically realistic multimodality renal artery phantoms consisting of vessels with varying degrees of stenosis was developed and evaluated using four imaging techniques currently used to detect renal artery stenosis (RAS). The spatial resolution required to visualize vascular geometry and the velocity detection performance required to adequately characterize blood flow in patients suffering from RAS are currently ill-defined, with the result that no one imaging modality has emerged as a gold standard technique for screening for this disease. Methods: The phantoms, which contained a range of stenosis values (0%, 30%, 50%, 70%, and 85%), were designed for use with ultrasound, magnetic resonance imaging, x-ray computed tomography, and x-ray digital subtraction angiography. The construction materials used were optimized with respect to their ultrasonic speed of sound and attenuation coefficient, MR relaxometry (T1, T2) properties, and Hounsfield number/x-ray attenuation coefficient, with a design capable of tolerating high-pressure pulsatile flow. Fiducial targets, incorporated into the phantoms to allow for registration of images among modalities, were chosen to minimize geometric distortions. Results: High quality distortion-free images of the phantoms with good contrast between vessel lumen, fiducial markers, and background tissue to visualize all stenoses were obtained with each modality. Quantitative assessments of the grade of stenosis revealed significant discrepancies between modalities, with each underestimating the stenosis severity for the higher-stenosed phantoms (70% and 85%) by up to 14%, with the greatest discrepancy attributable to DSA. Conclusions: The design and construction of a range of anatomically realistic renal artery phantoms containing varying degrees of stenosis is described. Images obtained using the main four diagnostic techniques used to detect RAS were free from artifacts and exhibited adequate contrast to allow for quantitative measurements of the degree of stenosis in each phantom. Such multimodality phantoms may prove useful in evaluating current and emerging US, MRI, CT, and DSA technology.
KW - in vitro experimentation
KW - multimodality
KW - phantom
KW - ultrasound
UR - http://www.scopus.com/inward/record.url?scp=79551678784&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79551678784&partnerID=8YFLogxK
U2 - 10.1118/1.3533674
DO - 10.1118/1.3533674
M3 - Article
AN - SCOPUS:79551678784
SN - 0094-2405
VL - 38
SP - 565
EP - 573
JO - Medical physics
JF - Medical physics
IS - 2
ER -