AJR Women's Imaging Online
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Uppot, R. N.
Right arrow Articles by Mueller, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Uppot, R. N.
Right arrow Articles by Mueller, P. R.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.06.0409
AJR 2007; 188:433-440
© American Roentgen Ray Society

Perspective

Impact of Obesity on Medical Imaging and Image-Guided Intervention

Raul N. Uppot1, Dushyant V. Sahani, Peter F. Hahn, Debra Gervais and Peter R. Mueller

1 All authors: Department of Radiology, Division of Abdominal Imaging and Interventional Radiology, Harvard Medical School and Massachusetts General Hospital, 55 Fruit St., White 270, Boston, MA 02114.

Received March 24, 2006; accepted after revision June 28, 2006.

Address correspondence to R. N. Uppot (ruppot{at}partners.org).

R. N. Uppot served as a consultant for Siemens Medical Solutions and was paid to write an internal Siemens MRI Hot Topics article, "Obesity and MRI Imaging."

CME This article is available for CME credit. See www.arrs.org for more information.

Abstract

OBJECTIVE. The purpose of this article is to discuss the impact of obesity on medical imaging and provide some solutions that are currently available to tackle the challenges of imaging obese patients.

CONCLUSION. Increasingly, radiologists are asked to image morbidly obese patients. The challenges facing radiology departments include difficulties in transporting patients to the department, inability to accommodate large patients on currently designed imaging equipment, and difficulties in acquiring desired image quality.

Keywords: abdominal imaging • body imaging • interventional radiology • obesity • radiology practice • technologists

The prevalence of obesity is increasing in the United States. According to the Centers for Disease Control and Prevention (CDC), approximately 64% of Americans are overweight, obese, or morbidly obese [1]. Obesity directly impacts health, with an increased incidence of diabetes, heart disease, and certain types of cancer [2].

Increasingly, however, hospitals are faced with the indirect impact of obesity. As the prevalence of obesity increases in the United States, there is a growing need for larger hospital beds, larger wheelchairs, and larger operating room tables [3].

Radiology departments are also facing the impact of obesity [4]. The increasing prevalence of obesity in the United States and the growing popularity of bariatric surgery have led to increased use of medical imaging of obese patients. Currently designed imaging technology has limited ability to accommodate and provide the desired quality of images in obese patients.

Obesity is defined by body mass index (BMI), which takes into account both height and weight (Table 1). Despite the importance of BMI in the clinical classification of obesity, for radiologists two more appropriate measurements are weight and body diameter.


View this table:
[in this window]
[in a new window]

  		TABLE 1: Clinical Weight Classification Based on Body Mass Index

 

In the appropriate management of obese patients, radiology departments often have to address several issues: What is the best technique to image the obese patient for a given indication? How should the obese patient be scheduled and transported to the radiology department? How will the patient fit on the imaging equipment available? How should equipment settings be modified to optimize image quality for the obese patient? This review will describe the difficulties in imaging obese patients and discuss imaging strategies to overcome these difficulties.

Imaging Techniques for Obese Patients

More than the clinical indications, a patient's weight and body diameter are important factors in deciding which imaging technique to use. Although a brain MRI may be the most appropriate technique for evaluating an acute stroke, if the patient cannot fit on the MR scanner, it is of no practical use. Therefore, before scheduling an obese patient for a diagnostic imaging procedure, it is important to know the patient's weight and diameter to assess whether the patient can fit on the imaging equipment. Table weight limits and aperture diameters for fluoroscopy differ from those for CT and for MRI.

Currently, industry standards exist for table weight limits and aperture diameters for each of the imaging techniques (Table 2). In increasing order of cross-sectional diameter, according to current industry standards, the imaging techniques are fluoroscopy, CT, cylindric bore MRI, and vertical bore open MRI.


View this table:
[in this window]
[in a new window]

  		TABLE 2: Industry Standard Weight Limit and Maximum Aperture Diameter by Imaging Technique

 


Figure 1
View larger version (154K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1 —Gantry view of CT scanner. Although gantry diameter is listed as 70 cm (black line), once table enters gantry, table thickness must be accounted for and subtracts 15-18 cm from vertical height (arrow).

 
Weight Limit
Patients who exceed the weight limit of the table as defined by the manufacturer can potentially damage the table or its motor mechanics. The table and table motor are insured by manufacturers up to a certain weight. The cost of damage will not be covered if a heavier patient is placed on the table for imaging.

Aperture Diameter
Patients may meet the weight limit of a table but may exceed the gantry or bore diameter because of their girth. Typically, the industry-standard aperture in fluoroscopy is 45 cm; the gantry diameter in CT, including MDCT, is 70 cm; and the bore aperture in MRI is 60 cm. Although the aperture diameters are accurate in the horizontal plane on CT and MRI, they do not account for the table thickness entering the gantry or bore and therefore overestimate the vertical distance (anteroposterior distance in a supine patient). Typically, in the vertical plane, 15-18 cm must be subtracted from the gantry or bore diameter to account for the table thickness (Fig. 1).

Radiologists and technologists should be aware of weight and aperture limits of all imaging equipment at their diagnostic imaging facility. Weight and aperture limit data should be posted and be made easily available within the department. Despite these industry standard limits, some imaging vendors are now recognizing the issue of obesity and have increased the table weights and aperture dimensions of their newest imaging equipment (Table 3).


View this table:
[in this window]
[in a new window]

  		TABLE 3: Advances in Weight Limit and Maximum Aperture Diameter by Imaging Technique

 

Transportation to the Radiology Department

Once the imaging technique is selected, the next step is appropriate scheduling and transporting of the patient. Obese patients who exceed the weight limits of the transportation equipment can complicate transportation within a hospital. Typically, in most institutions, the resources (larger wheelchairs and stretchers) suitable for transporting larger patients are limited. Delay in transporting scheduled patients to the department increases the downtime of imaging equipment, hampers throughput, and increases the wait time for other patients.

Proposed solutions to address this issue include coordinating with the transport department to schedule obese patients only when the larger wheelchairs and stretchers are available, purchasing more wheelchairs or stretchers that can accommodate larger patients if the institution is seeing a growing percentage of obese patients, and acquiring portable radiography or sonography equipment for patients who cannot be transported.

Can the Patient Fit?

In addition to the discussed limitations of table weight and aperture diameter for fluoroscopy, CT, and MRI, other techniques have constraint. For radiography, obese patients' surface areas may be too large to fit on a 14 x 17 inch (35.56 x 43.18 cm) cassette. Magnification may also cone off the area of interest even if it overlies the cassette. On sonography, patient obesity may limit the ability of technologists to appropriately position patients for quality images.


Figure 2
View larger version (101K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2 —Left upper quadrant radiograph in 31-year-old man who weighed 590 lb (268 kg) and exceeded film cassette dimensions of 14 x 17 inches (35.56 x 43.18 cm). Imaging each quadrant separately is important to obtain desired quality abdominal radiograph.

 
There are several solutions for fitting patients on imaging equipment. For radiographs, technologists must consider the routine use of multiple cassettes to image quadrants of the body in a large patient (Fig. 2) instead of using a single cassette, which may not cover the entire surface area and thereby render the images nondiagnostic. On sonography, imaging the kidneys via a flank approach can be particularly difficult for small technologists scanning large patients. Therefore, judicious use of pillows to help support the patient's body is recommended. Also, use of the spleen or liver window to scan the kidneys can be helpful. On CT and MRI, the patient should be positioned so that the area of interest is imaged. For scanning the abdomen, patients can be inserted feet first into the gantry to avoid the girth of the upper abdomen or chest. Another alternative is to use a different imaging technique that can accommodate the patient. In nuclear medicine, patients who exceed the weight limit of the tables may be imaged on their stretchers with the gamma camera.

Diagnostic Quality Images

After choosing the appropriate technique, transporting the patient, and fitting the patient, the final hurdle in imaging obese patients is the ability to obtain diagnostic quality images. The difficulties and solutions for imaging obese patients are specific for each imaging technique.

Radiography
Radiographs are limited by X-ray beam attenuation that results in lower image contrast. Also, the increased body thickness through which the X-ray beam must travel results in increased exposure time and introduces motion artifact. The typical setting to obtain a chest radiograph is a kVp of 90-95 and mAs of 2-2.5. However, in obese patients these settings can result in inadequate penetration of the X-ray beams through the patient's body, along with more background scatter (Fig. 3A).


Figure 3
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3A —57-year-old man who weighed 490 lb (223 kg). Reprinted with permission from Uppot R. How obesity hinders image quality and diagnosis in radiology. Bariatrics Today 2005; 1:31-33 (for electronic version, see [4]). Chest radiograph shows limited diagnostic quality image with poor X-ray penetration and poor visualization of lung bases.

 
Some of the solutions that can help in imaging obese patients before acquisition include using a grid and increasing the kVp and mAs (in chest X-ray, increase kVp to 100 and mAs to 4) and after acquisition increasing the film development speed from 400 speed film to 800 speed film and adjusting window and level settings. These equipment adjustments can help in imaging the obese patient (Fig. 3B).


Figure 4
View larger version (102K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3B —57-year-old man who weighed 490 lb (223 kg). Reprinted with permission from Uppot R. How obesity hinders image quality and diagnosis in radiology. Bariatrics Today 2005; 1:31-33 (for electronic version, see [4]). Chest radiograph in same patient after placement of grid and increase in kVp and mAs. This solution improved visualization of lung bases.

 
Fluoroscopy
Fluoroscopy is the most important imaging technique after gastric bypass surgery. At our institution, all patients after laparoscopic gastric bypass are imaged with an oral contrast agent to evaluate the integrity of the anastomosis and exclude gastric outlet obstruction. Because fluoroscopy is primarily limited by the aperture diameter (45 cm) and table weight (350 lb [159 kg]), in patients who exceed the equipment limitations, we use serial abdominal radiographs or CT to obtain the answers to the clinical questions (Fig. 4A, 4B).


Figure 5
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4A —55-year-old man who weighed 479 lb (218 kg) and exceeded table weight for fluoroscopy. Patient underwent barium swallow examination after gastric bypass. Abdominal radiograph in frontal plane obtained with patient standing as part of sequence shows active extravasation.

 

Figure 6
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4B —55-year-old man who weighed 479 lb (218 kg) and exceeded table weight for fluoroscopy. Patient underwent barium swallow examination after gastric bypass. Lateral radiograph confirms gastrocutaneous fistula (arrow).

 
Sonography
Sonography is limited in two ways. First, the increased thickness of body parts in obese patients results in poor penetration of the ultrasound beam beyond the focal depth. Second, the increased attenuation of the ultrasound beam as it passes through subcutaneous and intraperitoneal fat even further compounds the issue of beam penetration. Sonography is attenuated by fat at a rate of 0.63 dB/cm.

Solutions to improving image quality in obese patients include using the lowest frequency transducer available (2 MHz), positioning the transducer to image the organ of interest within the range of the focal length of the transducer, and examining the patient's previous imaging (CT or MRI) to determine the thickness of subcutaneous fat (Fig. 5A, 5B).


Figure 7
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5A —42-year-old woman who weighed 265 lb (120 kg). Axial CT shows thickness (white line) of subcutaneous tissue ultrasound beam has to penetrate before reaching peritoneum.

 

Figure 8
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5B —42-year-old woman who weighed 265 lb (120 kg). Sonogram corresponding to A. Ascites is not seen.

 
CT
If a patient can fit on a CT scanner (weight and girth), image quality is generally adequate. Details of small structures are visible even in the most obese patients. However, there are some limitations, including increased noise due to inadequate beam penetration; limited field of view, which can result in beam-hardening artifact in areas where the patient's body exceeds the size of the field of view; and image quality limitations because of image cropping.

Increased noise—Increased noise is a result of inadequate beam penetration (Fig. 6A). Solutions to decrease noise involve increasing the kVp to 140 and increasing the effective mAs. Two ways that this may be accomplished include decreasing the gantry rotation speed from one rotation in 0.5 second up to one rotation in 1 second to increase the effective mAs and, on MDCT scanners with automated tube current modulation, changing the scanner settings from "fixed mAs" to "automatic mAs," which will allow the scanner to determine the amount of mAs to deliver per body section. Although both of these solutions improve the image quality (Fig. 6B), they increase the radiation dose to the patient.


Figure 9
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6A —39-year-old woman who weighed 413 lb (188 kg). Axial CT image of abdomen with fixed mAs setting resulted in increased noise. Beam-hardening artifact is visualized where patient's body exceeds field of view (arrows).

 

Figure 10
View larger version (159K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6B —39-year-old woman who weighed 413 lb (188 kg). Repeat axial CT of abdomen with equipment setting switched to automatic mAs, allowing scanner to increase mAs and thereby decrease noise.

 

Limited field of view—Standard 70-cm gantry CT instruments have a field of view of 55-65 cm. Increases in gantry diameter on the newer, larger gantry CT scanners also permit increases in the field of view to accommodate larger patients. In areas where the patient's body is larger than the field of view but smaller then the gantry diameter, beam-hardening artifact can limit evaluation of internal organs. One solution is to recognize the artifact during image acquisition and adjust the patient position so that the area of interest does not exceed the field of view.

Dangers of cropping—After a CT study is acquired, technologists may crop the images to focus on internal organ structures at the expense of subcutaneous tissues. Cropping subcutaneous fat can result in the loss of valuable information, especially in patients undergoing evaluation for malignancies or nonspecific abdominal pain for which the abnormality may be within the subcutaneous tissues (Fig. 7A, 7B, 7C).


Figure 11
View larger version (96K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 7A —52-year-old woman who weighed 177 lb (80 kg) and had history of carcinoid. Axial CT in PET/CT study for metastasis. Image was cropped to focus on intraabdominal structures.

 

Figure 12
View larger version (50K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 7B —52-year-old woman who weighed 177 lb (80 kg) and had history of carcinoid. PET portion of PET/CT, which was not cropped, shows area of 18F-FDG uptake (arrow).

 

Figure 13
View larger version (70K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 7C —52-year-old woman who weighed 177 lb (80 kg) and had history of carcinoid. Uncropped axial CT image shows soft-tissue deposit corresponding to area of FDG uptake (arrow).

 

CT benefits of fat—Obesity, however, is not always detrimental to CT image quality. Patients who have predominantly intraperitoneal or retroperitoneal fat have improved visualization of internal organ structures compared with patients with a paucity of intraperitoneal fat because of the better delineation of internal organ structures by the fat (Fig. 8A, 8B).


Figure 14
View larger version (110K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 8A —Axial images in two patients show advantages of fat for CT evaluation. Axial CT image in 56-year-old man with extensive intraabdominal mesenteric fat shows wide separation of mesentery and internal organs, allowing better visualization of internal structures.

 

Figure 15
View larger version (107K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 8B —Axial images in two patients show advantages of fat for CT evaluation. Axial CT in 28-year-old woman with paucity of intraperitoneal fat shows difficulty in visualizing mesenteric structures.

 
MRI
The impact of obesity on MRI represents a dichotomy. Typically MR scanners used in hospitals with a high signal-to-noise ratio (SNR) and strong gradients (≥ 1.5 T) cannot accommodate patients weighing 350 lb (159 kg) or more or who are greater than 60 cm in diameter. On the other hand, typical vertical field open MRI systems can accommodate patients up to 550 lb (250 kg) and offer a range of vertical apertures from 40 to 55 cm but have a lower SNR and weaker gradients [5].

In addition to table weight limits and bore diameter limits, another design factor in MRI that is relevant in imaging obese patients is the bore length. Long bore lengths in cylindric MR scanners, typically 170 cm, may be uncomfortable for obese patients who can become claustrophobic if the entire torso is in the bore. Shorter bore length magnets, 125 cm, on the other hand, may be better tolerated by obese patients because they allow the head and neck area to be outside the bore during scanning. Manufacturers of newer MR scanners recently marketed for the obese population try to address this issue by offering larger bore diameters, shorter bore lengths, and higher table weight limits in a 1.5-T system.

In imaging obese patients, several technical limitations need to be accounted for in MR scanning, including radiofrequency penetration and gradient strengths, limited field of view, scanning time, and radiofrequency energy deposition of the skin.

Radiofrequency penetration and gradient strengths—The disadvantage of low-gradientstrength MR scanners is the low SNR. Increasing gradient strength will not allow improved radiofrequency depth penetration in obese patients but can increase SNR and spectral resolution. Larger body habitus also introduces noise and will therefore decrease the contrast- to-noise ratio (CNR). For receiver coils, the increased distance of inner organs from the coils in obese patients also affects the SNR.

Limited field of view—Larger patients occupy a larger field of view. The maximum field of view for 1.5-T cylindric bore MR scanners is around 40-50 cm. The maximum field of view for vertical field open systems ranges from 35 to 40 cm. Although a larger field of view is advantageous in imaging obese patients, there is an inverse relationship between field of view and image resolution. Larger fields of view can decrease image resolution. Therefore, in imaging obese patients, the smallest possible field of view is used to image the organ of interest without inducing wraparound artifacts (Fig. 9). Some manufacturers have a "no phase wrap" option to prevent wraparound artifact.


Figure 16
View larger version (159K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 9 —T1-weighted, fat-saturated axial gadolinium-enhanced MRI in 44-year-old woman. Wraparound artifact (arrows) is seen because of small field of view. Also in areas where patient's body touches bore, there is inadequate fat saturation.

 

Scanning time—Obesity can affect MRI scanning times. Scanning times are increased because of the larger cross-sectional area and longer craniocaudal dimensions, which require more slices. Increased scanning times can also lead to patient motion, with associated motion artifacts.

Other factors—Other techniques that can aid in improving image quality include using a body coil rather than a phased-array multicoil and using saturation bands to decrease noise from subcutaneous fat. Another factor to consider in MRI of obese patients is the deposition of radiofrequency energy on the skin where it abuts the gantry. At our institution, technologists pad areas of the patient that abut the bore to prevent minor burns.

Nuclear Medicine
In nuclear medicine, obesity degrades image quality by the scatter of photons within the soft tissues, decreasing the SNR (Fig. 10). Also, because the administration of radioisotopes is based on weight, obese patients may exceed the maximum allowable dose and may not be able to receive the proportionate dose of radionuclide for their body weight. Solutions in nuclear medicine imaging of obese patients include using the maximum allowable dose and imaging for a longer time to maximize counts.


Figure 17
View larger version (102K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 10 —Technetium-99m bone scans in 60-year-old man with suspected osteomyelitis. Photon scatter and soft-tissue attenuation limit image quality.

 
Mammography
Although fat appears to be helpful in mammography, with the improved visibility of lesion relative to the surrounding fat, studies have shown that increased BMI is associated with decreased geometric sharpness, decreased image contrast, and higher potential for loss of sharpness because of motion [6].

Interventional Radiology

Obese patients can present scheduling, positioning, and technical challenges to the interventional radiologist. Obese patients may also require high doses of weight-based sedative medications, which may put them at risk for respiratory depression.

Individuals responsible for scheduling interventional procedures should routinely ask for the patient's weight. From our experience, all patients greater than 250 lb (159 kg) should be flagged. In this group of patients, consultation with the referring clinical service and review of prior images by the interventional radiologist before scheduling the procedure can ensure the use of the most appropriate imaging technique and availability of equipment appropriate for the procedure.

Proper patient positioning can also be a problem in obese patients. Several ancillary staff members need to always be available to help move the patient from the stretcher to the procedure table and to help position the patient if necessary. The use of pillows and sandbags is important to secure the patient position before the start of the procedure.

Meticulous planning is important before starting the procedure. Prior imaging can help determine the depth of fat tissue and the most direct approach to the organ of interest. Instruments of appropriate length must be chosen before starting the procedure. Technically, obese patients are more challenging. The accuracy in targeting lesions decreases the deeper the lesion that is to be biopsied or drained.

Sedation of the obese patient can also present challenges. If patients do not tolerate the administered doses of sedatives, the use of a mix of different active sedatives or the assistance of an anesthesiologist may be necessary. Also, if a patient's airway has the potential to be compromised or difficult to access because of body habitus, the patient may not be a candidate for conscious sedation, making general anesthesia mandatory. A few obese patients may require a surgical approach for diagnosis or treatment if suitable imaging guidance cannot be provided.

Summary

The prevalence of obesity is increasing in the United States. Standard, widely deployed imaging equipment is approaching its limits in the ability to image obese patients. Radiologists and technologists need to be aware of the limitations of imaging equipment and the equipment adjustments that can be made to improve image quality in obese patients. Manufacturers of imaging equipment are making design changes to accommodate larger patients and will begin marketing such equipment over the next several years.

References

  1. Department of Health and Human Services. Centers for Disease Control and Prevention. Overweight and obesity: obesity trends: 1991-2001 prevalence of obesity among U.S. adults by state—behavioral risk factor surveillance system (BRFSS) 2001 selfreported data. Available at: www.cdc.gov/nccdphp/dnpa/obesity/trend/prev_reg.htm. Accessed March 20, 2006
  2. Department of Health and Human Services. Centers for Disease Control and Prevention. Overweight and obese: health consequences. Available at: www.cdc.gov/nccdphp/dnpa/obesity/consequences.htm. Accessed October 24, 2006
  3. Rundle RL. U.S.'s obesity woes put a strain on hospitals in unexpected ways. The Wall Street Journal Online Edition. Available at: http://online.wsj.com/article/0,,SB1020194636122710680.djm,00.html. Accessed November 30. 2006
  4. Uppot R. How obesity hinders image quality and diagnosis in radiology. Bariatrics Today 2005;1 : 31-33
  5. Uppot R, Sheehan A, Seethamraju R. Obesity and MR imaging. In: MRI hot topics 2005. Malvern, PA: Seimens Medical Solutions USA
  6. Guest AR, Helvie MA, Chan HP, Hadjiiski LM, Bailey JE, Roubidoux MA. Adverse effects of increased body weight on quantitative measures of mammographic image quality. AJR 2000;175 : 805-810[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 RadioGraphicsHome page
B. Siewert, J. Sosna, A. McNamara, V. Raptopoulos, and J. B. Kruskal
Quality Initiatives: Missed Lesions at Abdominal Oncologic CT: Lessons Learned from Quality Assurance
RadioGraphics, May 1, 2008; 28(3): 623 - 638.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Uppot, R. N.
Right arrow Articles by Mueller, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Uppot, R. N.
Right arrow Articles by Mueller, P. R.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS