| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
State of the Art |
1 From the Department of Radiology, University of Maryland Hospital, 22 S Greene St, Baltimore, MD 21201-1595. Received July 7, 1998; revision requested August 27; revision received October 2; accepted November 25. Address reprint requests to B.D.
Index terms: Computed tomography (CT), guidance, **.121192 • Fluoroscopy, technology, **.121192 • Interventional procedures • State-of-art reviews
Computed tomography (CT) has been both a standard imaging modality and a guidance technique for percutaneous interventions for more than 20 years (1–5). As an interventional guidance tool, CT has been limited by a lack of real-time capability, in contrast to conventional fluoroscopy and ultrasonography (US). While conventional CT allows determination of appropriate puncture sites, direction of needle insertion, and evaluation of needle placement after insertion, it requires the time-consuming acquisition of multiple single or helical images and does not allow real-time evaluation during the puncture procedure. Conventional CT guidance may be particularly limited in body regions associated with physiologic motion, especially in the chest and upper abdomen.
The introduction of slip ring technology and the continuously rotating x-ray tube in the 1980s made the goal of depicting CT images in real time the subject of considerable technical research. The evolution of new image reconstruction algorithms and high-speed array parallel processor systems for real-time raw data reconstruction and display paved the way for the development of real-time CT fluoroscopy (also known as continuous imaging CT). Other important parallel developments contributing to the evolution of CT fluoroscopy included the introduction of x-ray tubes with much improved heat capacity and more sensitive semiconductor detectors. This technology was developed initially by Toshiba Medical Systems, Tokyo, Japan, and introduced for clinical use in Japan by Katada and colleagues in 1993 (6,7). The first CT fluoroscopy scanner in North America was installed in 1994, and the U.S. Food and Drug Administration granted approval for its routine use in patients in 1996.
Helical or spiral CT scanners with CT fluoroscopy capability are currently available from five vendors: Elscint Medical Systems, Highland Heights, Ohio; GE Medical Systems, Milwaukee, Wis; Picker International, Highland Heights, Ohio; Siemens Medical Systems, Iselin, NJ; and Toshiba America Medical Systems, Tustin, Calif. The technical specifications differ in each system; the frame rate varies from 2 to 8 per second, and maximum duration for continuous CT fluoroscopy varies from 40 to 100 seconds. Additional tube filtration used during CT fluoroscopy for reduction of absorbed and scattered radiation dose is a useful feature but is not available on all systems (manufacturers' data).
At our medical center, over 400 interventional procedures have been performed with CT fluoroscopic guidance to date. In this article, we review our early clinical experience and that of others with real-time CT fluoroscopy for guidance of biopsy, drainage, and other interventional procedures. We also consider the potential future applications of this most recent advance in CT technology.
INSTRUMENTATION
The first scanner with CT fluoroscopy capability was developed in 1993 (7). This machine had a limited field of view, a fixed table position during CT fluoroscopy, and could generate three images per second. Subsequent improvements included a variable field of view of 18–40 cm and collimation of 1–10 mm, automatic or manual table motion during CT fluoroscopy, and a frame rate increase to 6 or 8 frames per second (manufacturer's data). Our own clinical experience has been with an Xpress/SX Aspire CI system (Toshiba Medical Systems, Tokyo), which has a real-time display of 6 frames per second. This scanner has a fast parallel processor system for handling a high volume of continuous raw data in the memory field and for subsequent real-time reconstruction in 256 x 256 display mode. The reconstruction algorithm allows for continuous updating of the image. The first image is created from the 360° of raw data acquired during the 1st second of scanning. Subsequent synchronous addition of new and subtraction of old 60° data sets allows updating of the image six times each second. The first image is created 1.17 seconds after initiation of CT fluoroscopy, and there is a subsequent 0.17-second interscan time. The maximum continuous fluoroscopic time is 100 seconds, and up to 800 seconds of raw data may be stored in memory.
The operator works in the CT room during an interventional procedure, using a control panel that allows control of CT fluoroscopy exposure, table motion, gantry tilt, and a laser light beam. The panel may be covered with a sterile transparent drape, which allows single operator performance of procedures. Alternatively, an assistant may help by adjusting the controls while the operator concentrates on the procedure itself. A foot pedal identical to the traditional fluoroscopy pedal can alternatively be used for CT fluoroscopy exposure. An in-room mobile monitor with "last image hold" is used to view the real-time images during interventional procedures. Additional features that are important for CT fluoroscopy procedures include a large gantry opening (72 cm in the system in operation at our medical center), and a high heat capacity x-ray tube (systems with x-ray tube capacity of up to 7.0 million heat units are available). Once the technologist sets the parameters for CT fluoroscopy, he or she may perform other duties while the radiologist performs the procedure.
Radiation Dosimetry and Protection
Radiation protection measures are necessary during CT fluoroscopy, as the radiologist is in the room during radiation exposure. The radiologist wears standard protection (including a lead apron of 0.5-mm lead equivalent, a 0.5-mm thyroid shield, and lead glasses or goggles) and monitors exposure with an over-apron body radiation badge. Radiation badges worn under the protective apron and thermoluminescent dosimeter finger rings on both hands are additional radiation measurement devices that were routinely used by all operators during the 1st year of operation of CT fluoroscopy at our medical center. These were used only for CT fluoroscopy procedures and were discontinued subsequently because measured doses were always well below statutory limits after the 1st month of operation.
The earliest percutaneous procedures guided by CT fluoroscopy were performed without needle holders. The operators' hands were placed directly in the CT beam, but radiation monitoring data showed the resultant absorbed dose to be excessive (7,8). This led to the development of prototypic needle holders, allowing the operator to insert biopsy and access needles under CT fluoroscopic guidance without primary beam exposure (8,9). Data from a study (10) demonstrated that doses to the operators' hands could be reduced to 0.4 mrad (4 µGy) per second with the use of a 25-cm-long sponge forceps needle holder (Fig 1). Both the "last image hold" feature on the monitor and the capacity for rapid video playback of a CT fluoroscopy sequence help to decrease the need for rechecking images with additional x-ray exposure. A fixed time limit requiring reset to continue (100 seconds in the system we use) also aids in radiation exposure consciousness.
|
Interventional Techniques and Tool Development
As noted above, the manual insertion of biopsy or access needles, guide wires, dilators, or catheters during real-time CT fluoroscopic guidance is impractical due to the high radiation dose to the hands (8). Therefore, percutaneous procedures must be done with the aid of a standoff needle holder device during continuous CT fluoroscopic guidance (Figs 1, 2). Alternatively, an intermittent, discontinuous CT fluoroscopic technique allows biopsy or access needles to be placed manually in an incremental fashion (Fig 3). This technique allows the operator to retain the manual sensitivity that many interventional radiologists have developed with experience, or to generate greater force when passing needles through resistant tissue (Fig 3). If the needle tip has moved out of the collimated beam plane, moving the table by hand during CT fluoroscopy allows it to be located easily. This latter manual table-sliding technique is very helpful to quickly locate the target lesion, needle tip, guide wire, or catheter and is our preferred method for table movement. The gantry angle may be tilted during CT fluoroscopy to facilitate an oblique approach.
|
|
THORACIC APPLICATIONS
When performing a conventional CT-guided biopsy, perhaps no other lesion is potentially more frustrating than a small lung nodule. Such lesions may be difficult to localize consistently on sequential conventional scans due to variations in patient breathing and patient motion. Numerous sections may be taken to localize the nodule, then insert and follow the biopsy needle tip into the lung. Time is spent while the technologist obtains each scan or series, followed by reconstruction, image display, and needle adjustment. After this often painstaking exercise, the final biopsy is performed "blind," without the benefit of fluoroscopic guidance to guarantee that the needle is still in the nodule and that it is being sampled accurately.
The most frequent application of CT fluoroscopy has been for real-time guidance during biopsy of pulmonary nodules (7,14). Although conventional and helical CT have been used for this purpose for many years, lack of real-time visualization of target lesions has been an important limitation, especially for small lesions and in patients who are unable to cooperate with breath holding. Uncooperative patients, whom we normally would not consider for percutaneous biopsy, may be considered suitable if real-time imaging is available (Fig 2). During early experience with CT fluoroscopy, biopsy of lung nodules was performed successfully with a single puncture in 90% of cases (7). Biopsy of very small lung nodules (smaller than 1.5 cm) has been successful with use of CT fluoroscopy in 82% of cases (14). These initial results are similar to success rates published previously for conventional CT guidance, but allowing for both a technical learning curve and the limitations of current needle holder devices, CT fluoroscopy does appear to have the potential to improve the overall success rates for percutaneous lung biopsy. In our own practice, we now attempt biopsy of lesions that we may have considered impractical previously, for example, in the mediastinum or paramediastinal areas of the lung (Figs 2, 3). Real-time imaging allows the dynamic relationship of masses to mediastinal tissues to be assessed during respiration. The phase of respiration may be chosen to allow optimal target alignment and to avoid inadvertent puncture of vital structures. The respective relationship of peripheral lung or pleural masses to the underlying pleura or bone during the respiratory cycle may be assessed also, allowing detection of tumor invasion. CT fluoroscopy allows immediate detection of pneumothorax during a procedure and assists rapid and precise placement of drainage tubes when necessary (12).
CT fluoroscopy has also been shown to be of value in interventions of the pleura: Complex pleural drainages and catheter placement are especially facilitated by the ability to rapidly visualize access needle, guide wire, and catheter position (12). In this study, all of 24 pleural fluid collections were successfully evacuated by using CT fluoroscopy, which proved particularly useful for rapid placement of drainage tubes in patients who were unable to cooperate with breathing instructions and in patients with a narrow window of access. The average total room time was 65 minutes with an average procedure time of 32 minutes and CT fluoroscopy time of 143 seconds. This reflects an average time saving of 25%–30% compared to our own previous experience with conventional CT guidance.
Another innovative use of CT fluoroscopy in the thorax has been for the guidance of transbronchial biopsy via the fiberoptic bronchoscope. With this technique, CT fluoroscopy allows accurate transbronchial needle placement into enlarged peribronchial lymph nodes or masses, with avoidance of major vascular structures (15).
ABDOMINOPELVIC APPLICATIONS
US is a suitable guidance tool for many interventional procedures in the abdomen and pelvis but may be limited when bowel loops obscure the needle path, when lesions are deeply situated in the retroperitoneum, or when the patient is obese. Conventional CT guidance is helpful for retroperitoneal procedures but may be limited in dealing with dynamically moving structures due to respiratory motion or bowel loops.
CT fluoroscopy has been investigated recently as an imaging tool for percutaneous abdominal and pelvic biopsy and therapeutic procedures (11). In that study, 119 percutaneous diagnostic and therapeutic procedures were performed in 97 patients; they were fluid collection aspiration and/or drainage catheter insertion (n = 59), tissue biopsy (n = 49), hepatocellular carcinoma ethanol ablation (n = 6), chemoneurolysis (n = 4), or brachytherapy catheter insertion (n = 1). One hundred twelve (94.1%) procedures were successfully performed by using either continuous CT fluoroscopy with a standoff needle holder or conventional manual insertion, with intermittent CT fluoroscopy to confirm instrument position. Image quality was adequate for needle, guide wire, dilator, or drainage tube placement in all but two hepatic biopsies in which the contrast between the lesion and background liver with use of low-dose CT fluoroscopy was inadequate to allow lesion identification. Conventional CT was used to guide biopsy in these cases. CT fluoroscopic imaging of ethanol distribution during injection facilitated tumor ablation (Fig 4) or neurolytic procedures (Fig 5). In this study, CT fluoroscopy allowed rapid assessment of needle and catheter placement, especially in nonaxial planes. The average fluoroscopy time needed to perform typical biopsy or fluid drainage was 133 and 165 seconds, respectively. The average total CT scanner suite time for biopsy and drainage procedures (ie, time required in the CT scanner room) was 75 minutes and 86 minutes, respectively. The average radiologists' procedure time (in the 78 cases in which data were available) was 46 minutes and 48 minutes, respectively. Little comparative information on the average duration of abdominal or pelvic interventional procedures with conventional CT guidance is available. However, it is our experience that the use of CT fluoroscopy reduces the duration of the radiologists' component of total procedure time for abdominal and pelvic interventional procedures by approximately 30% when compared to conventional CT guidance.
|
|
|
|
|
|
CRANIOSPINAL APPLICATIONS
A report by Kato et al (16) on the clinical use of CT fluoroscopy for interventional guidance included 15 cases of intracranial procedures. These interventions included cerebral hematoma and ventricular system drainage, drainage tube placement, and brain biopsy. All but one of these procedures were successful, though a parietal approach made continuous visualization of the procedure more difficult in some cases. In one patient, the target could not be visualized because of excessive artifact from a metallic stereotactic frame. In this series, a single complication of hematoma formation occurred after a brain biopsy.
A large series of spinal interventional procedures with real-time CT fluoroscopy was reported by Seibel et al (18). These procedures included periradicular spinal nerve root anesthesia therapy in 269 cases, thoracic or lumbar sympathectomy in 21, intratumoral therapy in 10, periarticular therapy in six, laser nucleotomy in six, and facet joint neurolysis in four. Reduction in duration of procedure time was estimated to be approximately 30% on average. These investigators suggested that the safety of spinal interventional procedures was increased by using real-time CT fluoroscopy.
MUSCULOSKELETAL SYSTEM APPLICATIONS
CT fluoroscopy is very suitable for guidance of percutaneous interventional procedures of the musculoskeletal system. It has been used for muscle or bone biopsy, for brachytherapy catheter placement, and for the guidance of sacroiliac joint injection. For sacroiliac joint injection, intermittent CT fluoroscopy combined with dynamic manual table movement allowed an efficient mechanism for performing needle placement in nonorthogonal planes and allowed completion of the procedure in under 15 minutes on average (19).
FUTURE DEVELOPMENTS
Potential further improvements in CT fluoroscopy include the development of even more sensitive semiconductor detectors, thereby improving image quality at the low-millampere doses used for this technique. Such improvements may potentially allow even further reduction in the milliamperage and radiation dose required. The most recently developed scanners are capable of image reconstruction at 8 frames per second, and even higher rates are likely to be achieved in the near future. Another future development related to CT fluoroscopy is the introduction of a virtual reality technique with use of goggles, which superimpose the screen image on the operator's visual field. When the operator wears these goggles, the internal needle position may be superimposed on the operator's view of the external portions of the needle and needle guide. This technique would obviate the operator having to repeatedly switch eye gaze from the patient to the monitor and back during needle placement. As noted above, the development of more suitable biopsy or access needles and immobilization devices is important to realize the full potential of CT fluoroscopy for percutaneous interventional techniques (8,9).
CONCLUSION
Although CT fluoroscopy faces competition as an interventional guidance tool from US, conventional fluoroscopy, and the evolving field of real-time magnetic resonance (MR) imaging, it appears to have a promising future. Although a more expensive guidance modality than US or conventional fluoroscopy, CT fluoroscopy is likely to have an important complementary role in body regions or tissues where neither of the former imaging techniques may be effective. MR imaging guidance is limited currently both by expense and by technical problems with development of nonferromagnetic interventional tools.
Initial studies suggest that CT fluoroscopy has potential for a major impact, especially for radiology services and departments with a substantial interventional workload. Although little firm comparative data are available, the evidence is suggestive that major time savings in the duration of procedures may be realized with CT fluoroscopy compared with conventional CT guidance (7,11,12,18,19). More challenging interventional procedures may be undertaken with the aid of CT fluoroscopic guidance (Figs 2 –6), expanding the possibilities for truly minimal invasive therapeutic procedures. As more users gain access to this technology, newer developments and applications are expected to unfold.
Acknowledgments
We offer sincere thanks to our colleagues, especially Thor Krebs, MD, and Geoff Hastings, MD, for their help in the preparation of the manuscript.
Footnotes
Neither B.D. nor P.A.T. has a direct financial interest in the subject matter and products discussed herein. Both B.D. and P.A.T. have received honorariums for lectures given on behalf of Toshiba America Medical Systems. The Department of Diagnostic Radiology at the University of Maryland has received educational support from Toshiba America Medical Systems in the form of donations of computer equipment.
References
This article has been cited by other articles:
|
|
 |
|
  R YOSHIMATSU, T YAMAGAMI, T KATO, T HIROTA, T MATSUMOTO, J SHIMADA, and T NISHIMURA Percutaneous needle biopsy of lung nodules under CT fluoroscopic guidance with use of the "I-I device" Br. J. Radiol., February 1, 2008; 81(962): 107 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H Cole and C P Lawinski Developments in radiography and fluoroscopy Imaging, June 1, 2006; 18(2): 75 - 81. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Carlson, J. P. Felmlee, C. E. Bender, R. L. Ehman, K. L. Classic, T. L. Hoskin, W. S. Harmsen, and H. H. Hu CT Fluoroscopy-guided Biopsy of the Lung or Upper Abdomen with a Breath-hold Monitoring and Feedback System: A Prospective Randomized Controlled Clinical Trial Radiology, November 1, 2005; 237(2): 701 - 708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Yamagami, T. Kato, S. Iida, T. Hirota, and T. Nishimura Percutaneous Needle Biopsy for Small Lung Nodules Beneath the Rib Under CT Scan Fluoroscopic Guidance With Gantry Tilt Chest, September 1, 2004; 126(3): 744 - 747. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Endo, Y. Kotani, M. Satouchi, Y. Takada, T. Sakamoto, N. Tsubota, and H. Furukawa CT Fluoroscopy-Guided Bronchoscopic Dye Marking for Resection of Small Peripheral Pulmonary Nodules Chest, May 1, 2004; 125(5): 1747 - 1752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Binkert, F. R. Verdun, M. Zanetti, C. W. Pfirrmann, and J. Hodler CT Arthrography of the Glenohumeral Joint: CT Fluoroscopy Versus Conventional CT and Fluoroscopy--Comparison of Image-Guidance Techniques Radiology, October 1, 2003; 229(1): 153 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Sakarya, O. Unal, B. Ozbay, K. Uzun, I. Kati, S. Ozen, and O. Etlik MR Fluoroscopy-guided Transthoracic Fine-Needle Aspiration Biopsy: Feasibility Radiology, August 1, 2003; 228(2): 589 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Yamagami, S. Iida, T. Kato, O. Tanaka, S. Toda, D. Kato, and T. Nishimura Usefulness of New Automated Cutting Needle for Tissue-Core Biopsy of Lung Nodules Under CT Fluoroscopic Guidance Chest, July 1, 2003; 124(1): 147 - 154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Yamagami, S. Iida, T. Kato, O. Tanaka, and T. Nishimura Combining Fine-Needle Aspiration and Core Biopsy Under CT Fluoroscopy Guidance: A Better Way to Treat Patients with Lung Nodules? Am. J. Roentgenol., March 1, 2003; 180(3): 811 - 815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Arrive, O. Rosmorduc, H. Dahan, L. Fartoux, L. Monnier-Cholley, M. Lewin, R. Poupon, and J.-M. Tubiana Percutaneous Acetic Acid Injection for Hepatocellular Carcinoma: Using CT Fluoroscopy to Evaluate Distribution of Acetic Acid Mixed with an Iodinated Contrast Agent Am. J. Roentgenol., January 1, 2003; 180(1): 159 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Yamagami, T. Nakamura, S. Iida, T. Kato, and T. Nishimura Management of Pneumothorax After Percutaneous CT-Guided Lung Biopsy* Chest, April 1, 2002; 121(4): 1159 - 1164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Yamagami, T. Nakamura, T. Kato, S. Matsushima, S. Iida, and T. Nishimura Skin Injury After Radiofrequency Ablation for Hepatic Cancer Am. J. Roentgenol., April 1, 2002; 178(4): 905 - 907. [Full Text] [PDF]
|
|
|
|
|
|
S J Golding and P C Shrimpton Radiation dose in CT: are we meeting the challenge? Br. J. Radiol., January 1, 2002; 75(889): 1 - 4. [Full Text] [PDF]
|
|
|
|
|
|
N Keat Real-time CT and CT fluoroscopy Br. J. Radiol., December 1, 2001; 74(888): 1088 - 1090. [Full Text] [PDF]
|
|
|
|
|
|
E. L. Nickoloff and P. O. Alderson Radiation Exposures to Patients from CT: Reality, Public Perception, and Policy Am. J. Roentgenol., August 1, 2001; 177(2): 285 - 287. [Full Text] [PDF]
|
|
|
|
|
|
W.-Y. Liao, M.-Z. Chen, Y.-L. Chang, H.-D. Wu, C.-J. Yu, P.-H. Kuo, and P.-C. Yang US-guided Transthoracic Cutting Biopsy for Peripheral Thoracic Lesions Less than 3 cm in Diameter Radiology, December 1, 2000; 217(3): 685 - 691. [Abstract] [Full Text]
|
|
|
|
|
|
N. Tomiyama, N. Mihara, M. Maeda, T. Johkoh, T. Kozuka, O. Honda, S. Hamada, S. Yoshida, and H. Nakamura CT-guided Needle Biopsy of Small Pulmonary Nodules: Value of Respiratory Gating Radiology, December 1, 2000; 217(3): 907 - 910. [Abstract] [Full Text]
|
|
|
|
|
|
J. Hierholzer, L. Luo, R. C. Bittner, C. Stroszczynski, R.-J. Schroder, N. Schoenfeld, P. Dorow, R. Loddenkemper, and A. Grassot MRI and CT in the Differential Diagnosis of Pleural Disease Chest, September 1, 2000; 118(3): 604 - 609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Cha, F. T. Lee Jr., J. M. Gurney, B. K. Markhardt, T. F. Warner, F. Kelcz, and D. M. Mahvi CT Versus Sonography for Monitoring Radiofrequency Ablation in a Porcine Liver Am. J. Roentgenol., September 1, 2000; 175(3): 705 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. P. Frush, L. F. Donnelly, and H. G. Chotas Contemporary Pediatric Thoracic Imaging Am. J. Roentgenol., September 1, 2000; 175(3): 841 - 851. [Full Text] [PDF]
|
|
|
|
|
|
S. N. Goldberg, V. Raptopoulos, P. M. Boiselle, K. J. Edinburgh, and A. Ernst Mediastinal Lymphadenopathy: Diagnostic Yield of Transbronchial Mediastinal Lymph Node Biopsy with CT Fluoroscopic Guidance-Initial Experience Radiology, September 1, 2000; 216(3): 764 - 767. [Abstract] [Full Text]
|
|
|
|
 |
|
 P. J. O'Neill, H. L. Flanagan, M. C. Mauney, W. D. Spotnitz, and T. M. Daniel Intrathoracic fibrin sealant application using computed tomography fluoroscopy Ann. Thorac. Surg., July 1, 2000; 70(1): 301 - 302. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. I. Goldberg and A. R. Margulis Gastrointestinal Radiology in the United States: An Overview of the Past 50 Years Radiology, July 1, 2000; 216(1): 1 - 7. [Abstract] [Full Text]
|
|
|
|
|
|
D. H. Sheafor, E. K. Paulson, M. A. Kliewer, D. M. DeLong, and R. C. Nelson Comparison of Sonographic and CT Guidance Techniques: Does CT Fluoroscopy Decrease Procedure Time? Am. J. Roentgenol., April 1, 2000; 174(4): 939 - 942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Maki, W. B. Gefter, and A. Alavi Recent Advances in Pulmonary Imaging Chest, November 1, 1999; 116(5): 1388 - 1402. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
