What Is CT Colonography?
Computed tomography (CT) colonography is an exciting development in gastrointestinal radiology that has the potential to revolutionize colorectal cancer screening. The procedure uses advanced helical CT in combination with fast computers and software to generate three-dimensional (3D) endoluminal views of the colon (1-4). The test is noninvasive, well tolerated, and a safe technique for evaluating the colon. In comparison with conventional colonoscopy, CT colonography requires no patient sedation; thus, the cost associated with patient monitoring during and after the procedure is reduced. Additional benefits of CT colonography include incidental findings that often affect patient care. These complex factors make the actual economic effect of CT colonography uncertain but potentially more economical than conventional colonoscopy.
Origins of CT Colonography
CT colonography was first popularized by Vining et al at the Bowman Gray School of Medicine (2). The developments that led to CT colonography began with the introduction of slip ring gantry design and helical CT scanning. More recently, the introduction of multisection CT detectors has contributed to making CT colonography an even more practical examination (5). These developments allow the continuous volumetric acquisition of data over a large region of the body. About the same time that helical CT scanners appeared, faster and cheaper computers with specialized 3D graphics software capable of producing realistic 3D endoluminal projections of the colon were developed (6-8).
The Colon Cancer Committee of the American College of Radiology has suggested adopting the term "CT colonography"
). CT colonography was selected because the name clearly defines the procedure as a CT-based study of the colon. The suffix "-graphy" is consistent with the previously used naming convention describing examinations such as cystography, pyelography, and myelography. The committee recommends abandoning the terms "virtual colonoscopy," "CT colonoscopy," "helical 3D CT pneumocolon," and "CT colography." Furthermore, an important part of CT colonography involves the use of axial and two-dimensional (2D) reformatted CT images; thus, the term "virtual" or "3D" may be too narrow a descriptor. A standardized name for this new imaging modality will be important to facilitate the transfer of important information to other health-care specialists.
Occasionally, the term "virtual colonoscopy" has been used to refer only to the 3D endoluminal views. This usage can be mistakenly interpreted as representing the entire CT colonography examination. Use of the term "3D endoluminal view" probably describes the 3D portion of the examination adequately without the additional metaphoric terminology of "virtual colonoscopy."
Requirements for CT Colonography
CT colonography requires adequate bowel cleansing, usually accomplished by administering a bowel preparation regimen the day before. Prior to acquisition of the data, the patient's colon is insufflated with room air or CO2 to provide optimal colonic distention. With CO2 insufflation, the insufflation apparatus generally contains a pressurized CO2 tank connected to a low-pressure bellows via a one-way regulator valve. One vendor has a unit that electronically dispenses CO2 and then maintains a specified colonic pressure. Patients are scanned, usually supine and then prone, with thin-section helical CT. Images are obtained from the dome of the diaphragm through the pelvis to completely cover the colon. At the San Francisco Veterans Affairs Medical Center, University of California, San Francisco, all patients are scanned with 3-mm collimation, with the pitch adjusted so that the total scan time is under 60 seconds per position. CT colonography protocols vary among different centers, with some protocols requiring additional colonic air insufflation between supine and prone scanning.
The axial CT images are downloaded to a computer workstation where the coronal and sagittal 2D reformatted and 3D endoluminal images are generated. There are several postprocessing methods currently used to generate the 3D endoluminal views. The two most widely used methods are surface rendering and volume rendering (6-8).
Importance of CT Colonography
CT colonography is being developed primarily for the identification of colonic polyps and cancer. It offers several potential advantages over other techniques for colon evaluation, including a complete examination of the colon and a rapid examination time. It is noninvasive and does not require sedation. The preliminary sensitivity of CT colonography for the detection of colorectal polyps is promising. If CT colonography proves to be more sensitive than other colon examination methods, it may become the diagnostic examination of choice.
|Large polypoid lesion of the
sigmoid colon with penetration of
the basement membrane and
spicular infiltration into the adjacent soft tissues.
|Small polypoid lesion of the distal
Technology Supports Vital Images CT Colonography Application
for Early Detection of Colon Cancer
Minneapolis, October 30, 2001 Vital Images, Inc. (Nasdaq SmallCap: VTAL), a medical imaging software company, has signed an agreement with PointDx, Inc. to license from PointDx patents related to virtual endoscopy, a technique that uses software to generate images of the inside of hollow organs, such as the colon and vascular structures, from computed tomography (CT) scan data. Access to this technology will allow Vital Images to enhance its CT colonography technology in Vitreaฎ 2, its flagship software product. Further, as the company moves forward, the technology licensed in the patent portfolio will support Vital Images expansion into additional markets with new applications. Under the terms of the agreement, PointDx will also provide technical support and access to clinical data to support Vital Images FDA submissions.
CT colonography is part of Vital Images major growth strategy into the disease screening market, said Albert Emola, Vital Images president and chief executive officer. We have been on the forefront of CT colonography by seeking out and incorporating the most innovative technologies. We expect this technological partnership to reinforce our market-leading position and to strengthen our intellectual property portfolio. More broadly, the licensed patent portfolio covers both domestic and international markets and strengthens our longer-term strategic position across a broad spectrum of potential new applications.
We are excited about working with Vital Images, said Dr. David Vining, president and chief executive officer of PointDx. This dynamic company has the ability to rapidly implement emerging technologies, such as that offered by PointDx.
Dr. Vining, credited as the father of virtual colonoscopy, has developed, along with his team, many of the fundamental concepts used in virtual endoscopy today. Much of the work supporting the patents was performed at the Wake Forest University School of Medicine in Winston-Salem, N.C., where Dr. Vining holds a position as an associate professor of Diagnostic Radiology.
CT colonography, or virtual colonoscopy, is a visualization application that uses two- and three-dimensional images of the colon to allow a physician to look for polyps and tumors in the colon in a less invasive manner than traditional colonoscopy. CT colonography has the potential to improve patient compliance with colon cancer screening recommendations and save lives by detecting pre-cancerous polyps and tumors at an early stage.
Screening More Patients for Colon Cancer with Virtual Colonoscopy
The American Cancer Society estimates that there will be over 98,000 new cases of colon cancer diagnosed in the United States in 2001. Research indicates that early detection and intervention can reduce mortality from the disease by up to 90 percent while also reducing the incidence of non-colon cancers. Furthermore, the American Cancer Society estimates that nearly 74 million people over age 50 are eligible for colon cancer screening. However, only 20 percent of this population base is actually screened each year, and many of these patients are screened with fecal occult blood tests that can yield less-than-optimal results. It is believed that virtual colonoscopy could lead to a substantial increase in the number of patients being screened for colorectal cancer.
Evaluation of Computer Aided Detection in CT Colonography: Potential Applicability to a Screening Population
D. S. Paik, MS, Stanford, CA ? C.F. Beaulieu, MD, PhD ? A. Mani ? R. Prokesh, MD ? J. Yee, MD ? S. Napel, PhD ? R.B. Jeffrey, MD ?
PURPOSE: Computer aided detection (CAD) has the potential to make interpretation of CT colonography (CTC) more efficient by quickly directing attention to likely polyps. While several initial studies in polyp CAD have demonstrated feasibility, they have only been reported for small numbers of carefully selected datasets. In this study, we demonstrate the performance of CAD on data that has both supine and prone scans, is of varying quality, and has low polyp prevalence, and, therefore, is more representative of a screening population.
METHOD AND MATERIALS: Fifty-one patients underwent bowel cleansing and colonic air insufflation, and then were scanned with single- or multi-detector CT. While all patients were imaged in both positions, we selected either supine (N=29) or prone (N=22) position data from each patient for evaluation. Scans were subjectively scored on three 5-point scales: colonic prep quality, distension quality, and imaging artifact. Patient selection included the full range over each quality factor. There were a total of 14 significant polyps (> 8.5 mm) in 9 patients, and 42 patients had no lesions larger than 8.5 mm (18% prevalence). Fiber-optic colonoscopy and prospective expert correlation to CTC were used to establish the gold standard against which CAD was assessed.
RESULTS: For all 51 patients, CAD achieved 92.9% sensitivity with 7.9 false positives (FP) per colon. For supine scans, CAD achieved 90.9% sensitivity at 10.9 FP/colon. 100% sensitivity was reached but at over 400 FP/colon due to a very flat polyp. For prone scans, CAD achieved 100% sensitivity at 4.4 FP/colon. A t-test performed on polyp CAD scores in both supine and prone scans did not reveal a significant difference (p > 0.05).
CONCLUSIONS: Our CAD algorithm is able to detect clinically significant polyps reliably with a low false positive rate in a population that included (1) supine and prone scans, (2) varying prep, distension and imaging quality, and (3) a low polyp prevalence. This suggests a practical role for our CAD algorithm in aiding radiologist interpretation of CTC in a screening population.