| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Division of Magnetic Resonance Imaging, Thomas
Jefferson University, 132 S 10th St., Suite 1096, Philadelphia, PA
19107.
2 Department of Radiology, Kobe University Graduate School of Medicine, Kobe,
Japan.
3 Department of Liver and Transplantation Surgery, Kobe University Hospital,
Kobe, Japan.
4 Department of Clinical Molecular Medicine, Division of Diabetes, Digestive and
Kidney Diseases, Kobe University Graduate School of Medicine, Kobe,
Japan.
Received May 10, 2005;
accepted after revision October 25, 2005.
Supported by the Smoking Research Foundation, Tokyo, Japan.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Single-shot spin-echo echo-planar diffusion-weighted MRI (TE = 66, b = 0, 600 s/mm2) was performed in phantom and clinical studies. The b value was set to minimize the effects of perfusion in tissue and to maintain signal-to-noise ratio. Bottle phantoms were scanned with and without parallel imaging and with various parallel imaging factors and at various positions to evaluate the effects of parallel imaging on ADCs. In 200 consecutive clinical patients (122 men and 78 women: mean age, 61.9 years), ADCs were calculated for liver (four segments), spleen, pancreas (head, body, tail), gallbladder, renal parenchyma, and back muscle, and then compared to evaluate the reliability of clinical ADC measurements with parallel imaging. ADCs were also calculated for diffuse diseases and focal lesions (94 malignant and 93 benign) of abdominal organs to evaluate the clinical usefulness of ADC.
RESULTS. Location-dependent changes in water ADCs were minimal with parallel imaging factors first of 3, then of 4, and were small except for measurements at the image periphery. Acetone ADCs were saturated at 4.00 x 10-3 mm2/s. Degraded image quality prevented ADC measurement of the left hepatic lobe and pancreas in 7-18 patients. There was no significant difference among ADCs of four liver segments (1.50 ± 0.24 [SD] x 10-3 mm2/s - 1.56 ± 0.31 x 10-3 mm2/s) and between ADCs of the right and left kidneys (2.65 ± 0.30 x 10-3 mm2/s, 2.59 ± 0.33 x 10-3 mm2/s). ADC of the pancreas tail (1.65 ± 0.37 x 10-3 mm2/s) was significantly lower than those of the head (1.81 ± 0.40 x 10-3 mm2/s) and body (1.81 ± 0.41 x 10-3 mm2/s) (p < 0.005). Renal ADCs were significantly lower in patients with renal failure (right: 2.15 ± 0.30 x 10-3 mm2/s; left: 2.11 ± 0.25 x 10-3 mm2/s) than in those without disease (right: 2.67 ± 0.29 x 10-3 mm2/s; left: 2.60 ± 0.32 x 10-3 mm2/s) (p < 0.005). ADC of pancreatic cancer was significantly higher than that of healthy pancreas (p < 0.05). ADC of renal angiomyolipoma was significantly lower than those of renal cell carcinoma and healthy renal parenchyma (p < 0.0005).
CONCLUSION. Clinical ADC measurements of abdominal organs and lesions using parallel imaging appear to be reliable and useful, and the effect of parallel imaging on calculated values is considered to be minimal.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Parallel imaging techniques, which use the spatial information from a phased-array multicoil to reduce the number of signals needed for a given spatial resolution, could be used to reduce the time of readout and scanning time, improving the quality of EPI [22, 23]. However, parallel imaging has been reported to have some problems. Signal intensity and signal-to-noise ratio (SNR) on the image can change, depending on the location in the section plane. We are aware of only one report that evaluated the reliability of ADC measurements using parallel imaging [24], and it was restricted to ADC of the right hepatic lobe in young, healthy volunteers. Furthermore, to our knowledge, no systematic study using phantoms to evaluate changes of ADCs related to parallel imaging and location relative to coil geometry has been described.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Liver cirrhosis was diagnosed in 57 patients (surgically, 14; biopsy, five; clinically, 38); chronic hepatitis in 29 patients (surgically, three; biopsy, nine; clinically, 17); chronic pancreatitis in three patients; acute pancreatitis in three patients; hydronephrosis or ureteral stone in seven patients; and chronic renal failure in eight patients. Clinical diagnosis of liver cirrhosis was based on the results of previous reports [25-27]. Fifty-two patients were diagnosed with cirrhosis associated with viral hepatitis (four with hepatitis B; 47 with hepatitis C; and one with both). Four patients had ethanol-induced cirrhosis and one had primary biliary cirrhosis.
Primary liver cancer was diagnosed in 40 patients (resection, 14; biopsy, two); secondary liver cancer in 19 patients (resection, 19); hepatic cavernous hemangiomas in 25 patients; hepatic cysts in 11 patients; pancreatic cancer in five patients (resection, five); renal cell carcinoma in 12 patients (resection, eight); renal angiomyolipoma in six patients (resection, two); renal cysts in 30 patients; gastrointestinal cancer in six patients (resection, five); and lymph node metastasis in eight patients. The lesions for which histologic proof were not available were mainly diagnosed on the basis of pathognomonic findings from sonography, CT, and MRI, other clinical findings, and follow-up examinations in 3-12 months.
Phantoms
To assess the reliability of ADCs and the location-in-plane-dependent ADC
changes with parallel imaging, bottle phantoms filled with water or acetone
(at 24°C) were scanned—applying a reduction factor for parallel
imaging (parallel imaging factor) of 1-4 or without parallel imaging—at
various positions, including at image center, 18 and 14 cm lateral from
center, and 1.5 cm from the receiver coils.
Figure 1 shows the details of
the phantom positioning.
Imaging Technique
All clinical and phantom MR studies were performed with a 1.5-T
superconducting imaging system (Gyroscan Intera, Philips Medical Systems). The
system provides a maximum gradient strength of 23 mT/m with a peak slew rate
of 105 mT/m/msec. A body coil was used for signal transmission and a
four-element phased-array multicoil for the body was used for reception.
In the clinical study, all patients underwent breath-hold single-shot echo-planar diffusion-weighted imaging (TR/TE = 1,500/66; matrix size = 96 x 96 [reconstructed to 256 x 256]; field of view = 400-450 mm; number of excitations = 2 [for each motion probing gradient direction]; EPI factor = 33 [i.e., the number of gradient-recalled echoes per spin echo]; slice thickness and gap = 6 and 2 mm, respectively; 30 transverse slices; parallel imaging factor = 3; bandwidth = 32.2 kHz; selective presaturation using inversion recovery [SPIR] for fat saturation) in addition to a routine abdominal imaging protocol that included T1-weighted dual fast gradient-recalled echo (TR/first-echo TE, second-echo TE = 126/2.3, 4.6; flip angle = 70°; matrix size = 256 x 196; field of view = 280-350 mm; number of excitations = 1; slice thickness and gap = 8 and 0 mm, respectively; 30 transverse slices; parallel imaging factor = 2), T2-weighted respiratory-triggered fast spin-echo (TR/TEeff = 2,500/80; echo-train length = 9; matrix size = 256 x 196; field of view = 280-350 mm; number of excitations = 2; slice thickness and gap = 8 and 0 mm, respectively; 30 transverse slices; SPIR for fat saturation; parallel imaging factor = 2), and T1-weighted 3D gradient-echo (TR/TE = 4.2/1.2; flip angle = 30°; matrix size = 224 x 168; field of view = 280-350 mm; number of excitations = 1; slice thickness = 8 mm; 30 transverse slices; SPIR was used for fat saturation; parallel imaging factor = 2). The single-shot technique was chosen to reduce motion-related artifacts arising with the EPI technique [1, 28]. The values of b factor were set at 0 and 600 s/mm2 to reduce the effect of perfusion in the tissues [2, 5, 12, 14, 15, 29]. The parallel imaging factor was set at 3 to keep the EPI factor small and to reduce image blurring. Values of b factor greater than 600 and parallel imaging factor of 4 were not used considering SNR on the diffusion-weighted images. The motion-probing gradient pulses were placed in the three orthogonal axes. Phase encoding direction was set anteroposteriorly in all sequences. For diffusion-weighted imaging, 15 sections were obtained during a breath-hold of 15 seconds, so two to three sequential acquisitions were required to encompass the entire upper abdomen. Saturation bands were not used for diffusion-weighted imaging.
In the phantom study, diffusion-weighted imaging as detailed was performed using the Gyroscan Intera MRI scanner three times for each position and for each parallel imaging factor or nonparallel imaging scans. Without using parallel imaging and with parallel imaging factors of 1, 2, 3, and 4, the shortest respective TEs were 68, 68, 62, 58, and 58 milliseconds. TE was set at 68 milliseconds for all conditions. A body coil was used for signal transmission and a four-element phased-array multicoil for the body was used for reception.
Image Analysis
All ADCs were calculated on a workstation with standard software (Diffusion
Calculation, Philips Medical Systems). The signal intensities for ADC
calculation were measured by using operator-defined region-of-interest (ROI)
measurements performed by the same radiologist.
All clinical MR images were analyzed retrospectively. ADCs were calculated for liver (four segments), spleen, pancreas (head, body, tail), gallbladder, renal parenchyma, back muscle, and focal lesions of liver, pancreas, kidney, gastrointestinal tract, and lymph nodes. The ROIs for liver (four segments), spleen, pancreas (head, body, tail), gallbladder, renal parenchyma, and back muscle were an oval of 100 mm2. The ROI of liver was located peripherally in each segment, and the ROI of the spleen was placed centrally. In the liver and spleen, vessels were avoided as much as possible. In the pancreas, vessels, pancreatic duct, and common bile duct were avoided as much as possible. The ROI within the gallbladder was placed in the largest possible area. The ROI of the renal parenchyma was placed peripherally in the posterior labrum. ADCs of cortex and medulla could not be separately measured because the matrix sizes used for the diffusion-weighted imaging were so small that it was difficult to distinguish these structures on the images [30].
At the time of ROI placement, the operator was aware of the location of the lesions and could refer to routine MR images but was blinded to the final diagnoses. ROI placements were checked by another experienced radiologist who was aware of the final diagnoses, and consensus opinion was then attained by the two operators for each ROI placement. The initial ROI was modified in one case of pancreatic cancer with secondary pancreatitis.
ADCs were compared among four liver segments, among three pancreatic parts, and between right and left renal parenchyma to evaluate anatomic changes in ADC. For liver ADC, mean values of three subgroups—with chronic hepatitis, with cirrhosis, and without liver disease—were compared in each segment. Patients with fatty infiltration, hemochromatosis, malignant liver tumors, or diffuse biliary disease, or after undergoing transarterial chemoembolization, systemic chemotherapy, or biliary intervention, or with a history of acute hepatitis (cured), were excluded from analysis of these subgroups.
For pancreatic ADC, the mean values of the three subgroups—with chronic pancreatitis, with acute pancreatitis, and without pancreatic diseases—were compared in each pancreatic part. Patients with diabetes, a history of acute pancreatitis (cured), pancreatic tumors, diffuse biliary disease, or duodenal or retroperitoneal tumors were excluded. For renal ADC, the mean values of the three subgroups—with chronic renal failure, with hydronephrosis or ureteral stone, and without renal or urinary tract disease—were compared in bilateral kidneys. Patients with abnormally increased serum creatinine (without a diagnosis of renal failure), malignant renal tumors, renal benign tumors larger than 5 cm, and retroperitoneal tumors were excluded. These pathologic conditions were chosen for analysis because previous reports using diffusion-weighted imaging without parallel imaging suggested their clinical usefulness for ADC measurements [2, 3, 5-9, 11-14, 16]. ADCs of patients without disease were used to establish the normal values for our system.
In the ADC measurement of lesions, only lesions larger than 10 mm in diameter were analyzed because measurements of smaller lesions would not be sufficiently reliable owing to the partial volume averaging effect. The ROI within each lesion was oval and was placed on the largest possible area, including necrotic or cystic parts. For measurement purposes in each patient, a maximum of three large lesions seen on diffusion-weighted images were chosen for analysis. Thus, the measurement was performed in 94 malignant lesions and 93 benign lesions.
The measured lesions included 28 primary liver cancers (mean diameter ± SD, 3.9 ± 3.1 cm), 26 secondary liver cancers (3.7 ± 2.5 cm), 19 hepatic hemangiomas (4.0 ± 1.3 cm), 13 hepatic cysts (4.1 ± 3.4 cm), five pancreatic cancers (5.3 ± 2.6 cm), three pancreatic simple cysts (2.4 ± 1.4 cm), three pancreatic pseudocysts (4.1 ± 1.8 cm), 12 renal cell carcinomas (5.7 ± 3.3 cm), eight renal angiomyolipomas (2.8 ± 1.0 cm), 42 renal cysts (2.6 ± 1.2 cm), five complicated renal cysts (2.4 ± 0.7 cm), six gastrointestinal cancers (4.1 ± 1.7 cm), and 17 lymph node metastases (3.5 ± 1.4 cm). Three follow-up cases with two intraductal papillary tumors and one serous cystadenoma of the pancreas were excluded from these measurements. Mean ADCs of the various lesion types and mean values of all hepatic segments (overall liver parenchyma), of all pancreatic parts (overall pancreatic parenchyma), and of bilateral renal parenchyma (overall renal parenchyma) were compared.
In the phantom study, ADCs were calculated on the sections for the center of phantom in the z-axis. The ROI was placed at the center of phantom on the image and was an oval of 600 mm2. Mean ADCs were compared per each parallel imaging factor and per each position. Image distortion on the phantom image was visually scored on a 3-point scale (1, severe; 2, moderate; 3, mild) by two experienced radiologists who recorded a consensus opinion. The average score was calculated for each parallel imaging factor and for each position.
Statistical Analysis
Statistical analysis of the mean ADCs of the liver segments, pancreas
parts, liver with and without diseases, focal liver lesions, and focal renal
lesions was done by one-way analysis of variance and using the Scheffé
test. Statistical analysis of the mean ADCs of pancreas with and without
disease, of renal parenchyma with and without disease, and of focal pancreatic
lesions was done using the Kruskal-Wallis and Scheffé tests. The
Student's t test was used for statistical analysis of the mean ADCs
of bilateral renal parenchyma.
Statistical analysis of the mean ADCs and of the image distortion scores of the phantoms was performed with the one-way analysis of variance and Scheffé test to determine the effects of parallel imaging technique and location-dependent changes.
All values were expressed as mean ± SD. For all tests used, a p value of less than 0.05 was considered statistically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
Tables 3 and 4 show the results of the patient studies. There was no significant difference among ADCs of the four liver segments or between ADCs of the right and left renal parenchyma. The ADC of the pancreas tail was significantly lower than those of the head and body (p < 0.005). ADCs of spleen, gallbladder, and back muscle were 1.26 ± 0.23, 3.50 ± 0.51, and 2.13 ± 0.25 x 10-3 mm2/s, respectively.
|
|
ADCs of the liver in patients with chronic hepatitis or cirrhosis had a nonsignificant trend toward being lower than those in patients without disease in each segment. ADC of the pancreas in patients with chronic pancreatitis had a nonsignificant trend toward being lower than those in patients without disease in each part. ADCs of renal parenchyma in patients with renal failure were significantly lower than those in patients without disease in both kidneys (p < 0.005). Mean ADCs of hydronephrosis and ureteral stones were not significantly different from those of normal renal parenchyma in both kidneys.
There was no significant difference among mean ADCs of primary liver cancer, secondary liver cancer, and the overall liver parenchyma (mean values of all segments). Mean ADCs of hepatic hemangioma and cyst were significantly higher than those of overall liver parenchyma and primary and secondary liver cancers (p < 0.0001). Mean ADC of hepatic cyst was significantly higher than that of hepatic hemangioma (p < 0.0001). Mean ADC of pancreatic cancer was significantly lower than those of pancreatic simple cyst (p < 0.0001) and pseudocyst (p < 0.0005), and significantly higher than that of overall pancreatic parenchyma (mean values of all parts) (p <0.05) (Figs. 3A, 3B, 3C, and 3D). Mean ADCs of pancreatic simple cyst and pseudocyst were significantly higher than those of overall pancreatic parenchyma (p < 0.0001). There was no significant difference between the mean ADCs of pancreatic simple cyst and pseudocyst. Mean ADC of renal angiomyolipoma was significantly lower than those of renal cell carcinoma (p < 0.0005), renal cyst (p < 0.0001), renal complicated cyst (p < 0.0001), and overall renal parenchyma (mean values of bilateral kidneys) (p < 0.0001) (Figs. 4A, 4B, 4C, and 4D). Mean ADC of renal cyst was significantly higher than those of renal cell carcinoma, complicated cyst, and overall renal parenchyma (p < 0.0001). There was no significant difference among mean ADCs of renal cell carcinoma, complicated cyst, and overall renal parenchyma.
|
|
|
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The measured ADCs in our phantom study were relatively higher than those of previous reports [5, 15, 16, 34]. This might be a result of differences in imaging sequence, workstation, and software, and suggests that ADCs should be considered as relative values, with calibration needed for each MR scanner. ADC measurements at the image periphery were not reliable, possibly because of the sensitive area of surface coils or of EPI. Our results suggest that the parallel imaging technique may help reduce artifacts and improve the reliability of the measured values. For the measurement of the ADC of acetone, the values were saturated, possibly caused by extremely low SNR in the ROIs on the diffusion-weighted images. When imaging sequences with many sections, a short acquisition time, and a single high b value are used, acetone might not be suitable as a standard substance.
Our hepatic ADC measurements were in the midrange of those in recent reports [24, 35]. Compared with our results, some investigators reported lower ADCs [16, 29], whereas others reported slightly higher ADCs [1, 7, 14]. These discrepancies may come from the differences in patient population, imaging sequence used, TE, b value, or special techniques that were applied, such as peripheral pulse gating, longer acquisition time, and diffusion tensor imaging (with increased acquired signals).
Previous studies did not address possible differences in ADC measurement throughout the liver [1-4, 9, 13, 14, 16, 29]; we found no differences among four liver segments. Also consistent with prior reports, we found that livers with chronic hepatitis and cirrhosis had lower a ADC than did healthy liver, and we found a similar ADC among primary and secondary liver cancer and healthy liver and a lower ADC for hemangioma versus cyst [2, 3, 5, 6, 8, 13, 14]. We are aware of only two publications discussing ADC of the pancreas [1, 3], each of which reported ADCs higher than those in our study.
The highest pancreatic ADCs were reported by Chow et al. [1], who evaluated young healthy volunteers, unlike Ichikawa et al. [3] and us, who evaluated clinical patients with a mean age of approximately 60 years. As a person ages, the pancreas shows several age-related changes such as atrophy, fatty infiltration, and fibrosis, and our results suggest that pancreatic ADC may also change with age. Moreover, age-related changes have been reported in the structure and functions of the liver and kidney. The volume of the liver decreases, despite an increase in the size of hepatocytes, suggesting loss of liver cells. In addition, the hepatic blood flow and the synthesis of urea and cholesterol are reduced. The aging kidney is characterized by a decrease in renal blood flow and glomerular filtration rate, mainly as a result of glomerulo-sclerosis. These age-related factors may affect ADCs of the organs.
We found ADC of the pancreatic tail to be lower than ADC of other parts of the pancreas, which is possibly caused by differences of surrounding tissues. ADC of pancreatic cancer was higher than ADC of a healthy pancreas and lower than simple and complicated cysts; however, the number of patients was small.
Parenchymal ADC in chronic renal failure was significantly lower than that of healthy parenchyma, which was consistent with the previous reports [8, 12]. We did not find any significant differences in ADC between healthy and hydronephrotic kidneys. In our study population, renal dysfunction was mild in the hydronephrotic patients. Toyoshima et al. [11] reported normally functioning hydronephrotic kidneys showed ADCs similar to those of healthy kidneys, whereas dysfunctioning kidneys showed lower ADCs. Although there are a few publications on ADC measurement of renal tumors, benign and malignant tumors were not evaluated separately [19-21]. We found that angiomyolipoma had significantly lower ADC than did renal cell carcinoma, cyst, complicated cyst, and overall healthy parenchyma. Renal cell carcinoma and complicated cyst had similar ADCs to overall healthy parenchyma. ADC of the gallbladder was slightly lower than ADCs of hepatic and renal cysts and was consistent with the results of Yamada et al. [2].
ADC of back muscle has not been reported previously. Our results showed the ADC measurement adjacent to the receiver coils is sufficiently reliable and might be usable for calibrating ADC and measurement of background noise for calculation of the SNR of the images.
Without parallel imaging, ADC measurements at the left hepatic lobe and pancreas are often difficult to perform, although unsuccessful measurements have not been reported. With parallel imaging, we can reduce the percentage of unsuccessful measurement to approximately 5% for the left medial segment of the hepatic lobe, 11% for the left medial segment of the liver, 9% for the pancreatic head, 10% for the pancreatic body, and 7% for the pancreatic tail. In addition, segmental ADC analysis of the liver and pancreas is possible and may be useful for the evaluation of various diffuse parenchymal diseases. In some patients, however, image degradation was still severe for the left hepatic lobe and pancreas. As for the left hepatic lobe, either cardiac or peripheral pulse triggering or reducing the number of motion-probing gradient directions might be helpful [24, 29]. However, pulse triggering requires more acquisition time, and reducing motion-probing gradient directions affects ADC of the kidney, which has anisotropic diffusion [1, 30, 36].
In the pancreas, the main causes of image degradation are thought to be susceptibility effect by gastrointestinal air and atrophy of the organ by aging. Some investigators have used the free-breathing technique for ADC measurements in the kidney to increase the number of signal acquisitions and to obtain multiple and high b values [12, 37]. This technique may be applicable to other retroperitoneal organs, such as the pancreas, large intestine, and duodenum, in which respiratory artifacts are relatively smaller than in the liver and spleen. The multishot (interleaved) EPI technique is reportedly effective for ADC measurement of the kidney [36]. This technique can reduce the EPI factor and susceptibility artifacts caused by gastrointestinal and lung air, although it requires longer acquisition time to encompass the abdomen.
The optimal b value and TE for abdominal diffusion-weighted imaging have not yet been determined. Some investigators recommended a b value larger than 400 s/mm2 because it can reduce "T2-shine-through" and intravoxel perfusion effects [2, 5, 12, 14, 15, 29]. However, a higher b value leads to a lower SNR. For our study, the b values were set at 0 and 600 s/mm2 to maintain SNR. A shorter TE has been recommended because it can increase SNR and reduce susceptibility-induced image distortions [1, 24]. A smaller EPI factor also has been recommended to reduce susceptibility-induced image distortions, and we accomplished this by using parallel imaging, filling 96 lines of k-space in the phase-encoding axis with a single-shot EPI factor of only 33. This required an increase in the parallel imaging factor up to 3. We are not aware of any previously published data of abdominal diffusion-weighted imaging using a parallel imaging factor of 3. Our results suggest that reducing the EPI factor is effective for improving image quality. Further studies are needed to determine the feasibility of ADC measurement with a parallel imaging factor of 4.
Our study has several limitations. First, only a small number of patients had histopathologic proof. However, our results show the average values in clinical populations. In addition, most cases of cirrhosis and chronic hepatitis in our population were diagnosed without histologic proof—the diagnoses were based on laboratory and imaging criteria. The sensitivity and specificity for the noninvasive distinction between cirrhosis and chronic hepatitis are reported to be 71.7-89% and 75-89.9%, respectively [25-27]. Second, direct comparison of diffusion-weighted images with and without parallel imaging was not performed in our retrospective clinical study and should be done in clinical cases to confirm our results in the phantom part of our study. Third, only the b value of 600 s/mm2 was evaluated. Further evaluation of higher b values is needed, because higher b values can reduce the effect of perfusion on the calculated ADC values. Fourth, clinical diffusion-weighted imaging without parallel imaging, or with parallel imaging factors less than 3, was not obtained. Fifth, we evaluated only ADC of renal parenchyma because of the difficulty in distinguishing cortex and medulla parenchyma [30]. In addition, our imaging sequence had a small matrix size and we had many elderly patients. Finally, posttreatment lesions were not evaluated, although Kamel et al. [38] report the usefulness of diffusion-weighted imaging in the posttreatment evaluation of malignant liver tumors.
In conclusion, clinical ADC measurements of abdominal organs and lesions using parallel imaging appear to be reliable and useful, and the effect of parallel imaging on calculated values is considered minimal. However, some problems still remain in measurements at the left lobe of the liver and pancreas. Measurement at the periphery of the image should be avoided.
Acknowledgments
We wish to express special thanks to Kazuhiro Kubo of Kobe University
Hospital for his outstanding technical assistance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. G. Sheiman and A. Sitek Feasibility of Measurement of Pancreatic Perfusion Parameters with Single-Compartment Kinetic Model Applied to Dynamic Contrast-enhanced CT Images Radiology, October 7, 2008; (2008) 2492080026. [Abstract] [Full Text]
|
|
|
|
|
|
J. Zhang, Y. Mazaheri Tehrani, L. Wang, N. M. Ishill, L. H. Schwartz, and H. Hricak Renal Masses: Characterization with Diffusion-weighted MR Imaging--A Preliminary Experience Radiology, May 1, 2008; 247(2): 458 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Parikh, S. J. Drew, V. S. Lee, S. Wong, E. M. Hecht, J. S. Babb, and B. Taouli Focal Liver Lesion Detection and Characterization with Diffusion-weighted MR Imaging: Comparison with Standard Breath-hold T2-weighted Imaging Radiology, March 1, 2008; 246(3): 812 - 822. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||
