CAD techniques, challenges, and controversies in computed tomographic colonography.

Computer-aided diagnosis (CAD) for computed tomographic colonography (CTC) automatically detects the locations of suspicious polyps and masses on CTC and provides radiologists with a second opinion. CAD has the potential to increase radiologists' diagnostic performance in the detection of polyps and masses and to decrease variability of the diagnostic accuracy among readers without significantly increasing the reading time. Technical developments have advanced CAD substantially during the past several years, and a fundamental scheme for the detection of polyps has been established. The most recent CAD systems based on this scheme produce a clinically acceptable high sensitivity and a low false-positive rate. However, CAD for CTC is still under active development, and the technology needs to be improved further. This report describes the expected benefits, the current fundamental scheme, the key techniques used for detection of polyps and masses on CTC, the current detection performance, as well as the pitfalls, challenges, controversies, and the future of CAD.

Diagnostic performance of virtual colonoscopy.

The most important diagnostic performance of virtual colonoscopy is the ability to detect colorectal polyps. However, its diagnostic value for evaluation of patients undergoing incomplete colonoscopy and the ability to detect unsuspected extracolonic findings are also important diagnostic features. Because of important differences between published patient populations, a meta-analysis combining data from different studies would not provide a reliable statistic. The by-patient sensitivity for patients with polyps 10 mm and larger ranges from 75% to 100% in most series and was 100% in the two largest series published to date. By-polyp sensitivity for polyps 10 mm and larger ranges from 50% to 100% and was 89-90% in the two largest series. Detailed data on smaller polyps are summarized. Virtual colonoscopy is an excellent test for evaluation of patients with incomplete colonoscopies. The incidence of significant extracolonic findings is 7-13%.

Automated segmentation of colonic walls for computerized detection of polyps in CT colonography.

PURPOSE: A new method for fully automated segmentation of the colonic walls in volumetric CT data was developed for limitation of the search space in computerized detection of polyps. METHOD: For reliable segmentation, an anatomy-oriented approach was used, in which several anatomical structures are segmented in addition to the colon for utilization of their properties. RESULTS: The segmentation method was validated by use of 14 data sets, consisting of cases positive for colonic polyps. We found that the segmented colonic walls included all of the polyps. A subjective rating of the results was performed based on several criteria for visualization of anatomic detail of the colonic wall and mucosal surface. Except for a few cases in which insufflation of the colon was insufficient, all of the results included >95% of the colonic walls. CONCLUSION: This method for colonic wall segmentation is reliable and the segmentation results are applicable in both visualization of the colon and computer-aided diagnosis in the detection of polyps in CT colonography.

Virtual colonoscopy: a novel imaging modality for colorectal cancer.

Advances in computed tomography (CT) technology and computer capabilities have contributed to the development of a new imaging modality for colorectal lesions called CT colonography or virtual colonoscopy. Virtual colonoscopy is a rapid, minimally invasive scan of the cleansed and distended colon. Early work has demonstrated that this test is safe and well tolerated, and that it may be sensitive and specific enough to identify most significant precancerous or cancerous lesions. A number of technical and practical problems remain before virtual colonoscopy can be applied at a population level.

Tumor size on computed tomography scans: is one measurement enough?

BACKGROUND: Bidimensional tumor measurements are used routinely as surrogates for tumor volume. The purpose this study was to determine whether there is any added benefit in bidimensional or tridimensional measurements over a unidimensional measurement. METHODS: Sixty-nine colorectal hepatic metastases on 19 computed tomography scans (1-8 lesions per scan) from 9 patients were analyzed. Five patients contributed 2-4 scans each (mean, 3 scans). The standard volume of these lesions was determined by the "summation of areas" technique. The maximum axial dimension, the product of the greatest axial dimensions, and several volume estimates (based on the volumes of a sphere, an ellipsoid, and a cube) each were correlated with the standard volume. RESULTS: The maximum axial dimension and the product of the greatest axial dimensions correlated equally with tumor volume (correlation coefficient = 0.93). Surrogate measures based on the equations for a sphere and an ellipsoid underestimated tumor volume, whereas the equation for a cube overestimated volume. CONCLUSIONS: When reporting tumor size, there is no significant added benefit in reporting bidimensional or tridimensional measurements over the maximum axial dimension.

Collateral pathways in superior vena caval obstruction as seen on CT.

PURPOSE: Collateral venous pathways occurring with superior vena cava (SVC) obstruction were examined based on CT scans obtained from the thoracic inlet to the pubic symphysis. Similarities and variations from the prior classification scheme were analyzed. METHOD: A retrospective review of our database resulted in a cohort comprising 21 CT scans from 19 patients. The location and frequency of each collateral pathway and the level of the SVC obstruction were tabulated. An accepted classification scheme was applied to the collateral patterns in each case. Additional and atypical features were noted. RESULTS: Fifteen common collateral veins were found that could be grouped into one to four collateral pathways. Unusual shunts, including hepatic parenchymal and pulmonary pathways, were found. Thirteen cases (62%) varied from the standard classification owing to different occlusion levels or presence of other collaterals. No statistically significant relationship between the level of occlusion and the number of collateral pathway groups was found. The most common abdominal collateral veins were those along the liver surface (52.3% of cases), although 18 patients (94.7%) had at least one collateral vein visible in the abdomen. CONCLUSION: The spectrum of venous collateral formations as seen on thoracoabdominal CT scans often includes collaterals at or below the level of the diaphragm, including intrahepatic shunts. Many collateral patterns found in this series could not be classified with the existent classification scheme.

Portal and mesenteric venous calcification in patients with advanced cirrhosis.

OBJECTIVE: The incidence of calcification in the portal and mesenteric venous system was studied in patients with advanced cirrhosis undergoing evaluation for liver transplantation. The significance of portal and mesenteric calcification on liver transplantation was also investigated. CONCLUSION: An 11% incidence of portal and mesenteric venous calcification was found in patients with cirrhosis, which was much higher than anticipated. Two (29%) of seven patients who had calcification present on CT and underwent liver transplantation died at surgery as a result of portal venous thrombosis. Thus, venous calcification seen on CT is a significant finding in patients undergoing liver transplantation.

Role of chest CT in the follow-up of ovarian adenocarcinoma.

OBJECTIVE: We describe the prevalence of metastatic chest disease in ovarian adenocarcinoma as seen on CT. We sought to determine whether routine chest CT added any pertinent information to the follow-up examination of patients with ovarian adenocarcinoma. MATERIALS AND METHODS: Retrospective review of our tumor registry yielded 96 patients with ovarian adenocarcinoma who had only a single primary malignancy and at least one CT scan of the chest, abdomen, and pelvis. CT scans were reviewed to assess the presence of metastatic chest disease in relation to disease activity in the abdomen and pelvis. Chest CT findings were correlated with the physical examination findings and CA-125 levels and were reviewed in consultation with a gynecologic oncologist to select only those patients with chest abnormalities attributable to metastatic disease. RESULTS: A total of 266 CT scans were obtained. Forty (41.7%) of the 96 patients had abnormalities attributable to metastatic chest disease on one or more scans. In the absence of disease progression in the abdomen and pelvis, chest disease progression was seen in only six (2.7%) of the 226 follow-up CT scans. Five of the six patients had rising CA-125 levels. CONCLUSION: Correlation of the findings of abdominal and pelvic CT with the physical findings and the CA-125 levels serves as effective follow-up in patients with ovarian adenocarcinoma. The contribution of additional chest CT in these patients is small.

Comparative viewing modalities for CT cystography.

BACKGROUND: To define the speed and accuracy of two different reconstructive techniques using computed tomography (CT) cystography for the detection and measurement of urinary bladder masses and determine the overall ease of use. METHODS: Ten patients scheduled for cystoscopy for the evaluation of hematuria or bladder masses were studied by means of thin-section CT of the air-distended bladder. Two techniques were employed by two radiologists to blindly interpret the data: conventional two-dimensional data with interactive three-dimensional problem solving (2D3DPS) and surface-shaded display (SSD) three-dimensional images. The results were compared with the data from cystoscopy. RESULTS: Twenty-two (100%) of 22 masses detected on cystoscopy were visualized using the reconstructive techniques. Both modalities were shown to have high accuracy, but only the 2D3DPS had a sensitivity and specificity of 100% for both observers at the patient-level diagnosis. The sensitivities for detecting individual masses for the two observers were 100% and 64% for 2D3DPS and 64% and 70% for SSD. CONCLUSION: Both methods used to display the CT data had a high sensitivity and specificity for masses, but only the 2D3DPS had a sensitivity and specificity of 100% at the patient-level diagnosis, thus making it a feasible imaging modality for cystography. It was also preferred overall for ease of use, high accuracy, and relative low cost.

Efficacy and safety of mangafodipir trisodium (MnDPDP) injection for hepatic MRI in adults: results of the U.S. Multicenter phase III clinical trials. Efficacy of early imaging.

The efficacy of contrast-enhanced magnetic resonance imaging (MRI) for detecting and characterizing, or excluding, hepatic masses was assessed in 404 patients, following the intravenous administration of mangafodipir trisodium (MnDPDP) injection, a hepatic MRI contrast agent. An initial contrast-enhanced computed tomography (CT) examination was followed by unenhanced MRI, injection of MnDPDP (5 micromol/kg IV), and enhanced MRI at 15 minutes post injection. Agreement of the radiologic diagnoses with the patients' final diagnoses was higher for enhanced MRI and for the combined unenhanced and enhanced MRI evaluations than for unenhanced MRI alone or enhanced CT using the clinical diagnosis as the gold standard. Mangafodipir-enhanced MRI uniquely provided additional diagnostic information in 48% of the patients, and patient management was consequently altered in 6% of the patients. MnDPDP-enhanced MRI was comparable or superior to unenhanced MRI and enhanced CT for the detection, classification, and diagnosis of focal liver lesions in patients with known or suspected focal liver disease.

Safety and efficacy of mangafodipir trisodium (MnDPDP) injection for hepatic MRI in adults: results of the U.S. multicenter phase III clinical trials (safety).

The short-term safety of mangafodipir trisodium (MnDPDP) injection was studied in 546 adults with known or suspected focal liver lesions. An initial contrast-enhanced computed tomography examination was followed by unenhanced magnetic resonance imaging (MRI), injection of MnDPDP (5 micromol/kg), and enhanced MRI. Adverse events were reported for 23% of the patients; most were mild to moderate in intensity, did not require treatment, and were not drug related. The most commonly reported adverse events were nausea (7%) and headache (4%). The incidence of serious adverse events was low (nine events in six patients) and not drug related. Injection-associated discomfort was reported for 69% of the patients, and the most commonly reported discomforts included heat (49%) and flushing (33%). Changes in laboratory values and vital signs were generally transient, were not clinically significant, and did not require treatment. There were no clinically significant short-term risks from exposure to MnDPDP.

CT colonography: the next colon screening examination?

Computed tomographic (CT) colonography is a new-generation technique for detecting colorectal neoplasms by using volumetric CT data combined with specialized imaging software. Patient examinations require full colonic preparation, insufflation, and data acquisition with the patient in the supine and prone positions. Current CT technology allows a single image of the colon to be acquired in as little as 20 seconds with a minimum of patient discomfort. Specialized computer software for interpretation usually combines transverse, multiplanar reformation, and three-dimensional endoluminal images for the optimal visualization of the colon and rectum. As of the time this article was written, CT colonography was competitive as a full structural colonic examination for the detection of polyps and cancer. To the authors' knowledge, no study results have yet been reported in a screening population. The unique capabilities of CT colonography include the display of the proximal colon that is inaccessible at colonoscopy because of obstructing colonic lesions or because of incomplete endoscopic examinations and the assessment of extracolonic abdominal and pelvic organs. This abdominopelvic survey potential provides radiologists with an opportunity to discover other potentially life-threatening, asymptomatic conditions. Further technologic developments and validation studies are in progress. CT colonography is an exciting and promising technique with an enormous potential for colorectal screening in the future.

Automated calculation of the centerline of the human colon on CT images.

RATIONALE AND OBJECTIVES: This article presents an evaluation of an automated technique for determining the colon centerline with computed tomographic (CT) data sets. MATERIALS AND METHODS: The technique proceeds as follows. After indication of a voxel in the rectum, voxels corresponding to air were segmented. Points along the colon centerline were estimated on the basis of centers of mass of grown voxels. A second segmentation and centerline calculation was initiated at the cecum. These two centerlines were then averaged. The resulting average was refined by using lumen data obtained perpendicular to the average centerline. The accuracy of the technique was investigated with simulation phantoms. The technique was also evaluated for 40 clinical colon cases. Calculated centerline points were compared with those indicated by radiologists for a randomly selected clinical case. RESULTS: In the simulation studies, the calculated centerline points were, on average, within 2.5 mm of the true centerlines but differed by up to 4 mm in regions of deep folds or sharp turns. In the clinical colon study, 40% of the centerlines were computed with a single seed point and 25% with two seed points. Average centerlines were computed in 1 minute. The root mean square difference between the computed centerline points and those indicated by the radiologists was 4-5 mm (comparable to interobserver variations). CONCLUSION: Accurate centerlines can be determined from colon CT data with this automated technique.

CT findings after laparoscopic repair of ventral hernia.

OBJECTIVE: We describe CT findings after laparoscopic repair of ventral hernia with emphasis on formation of postoperative fluid collections that can mimic recurrent bowel herniation or infected postoperative fluid collections. CONCLUSION: The porous property of the mesh used in laparoscopic repair of ventral hernia allows reaccumulation of fluid in the existing hernia sac or spaces in the subcutaneous tissues that can be created by laparoscopic manipulation. These fluid collections should be expected and are differentiated from infected fluid collections or hernia recurrence by clinical presentation, laboratory data, and lack of ancillary features associated with true hernia such as presence of hernia sac, herniated mesentery, or bowel obstruction.

CT colonography with three-dimensional problem solving for detection of colonic polyps.

OBJECTIVE: We performed CT colonography in patients referred for conventional colonoscopy, interpreted the axial images, and used commercially available software to reconstruct endoluminal perspective views to differentiate polyps from folds. SUBJECTS AND METHODS: We prospectively examined 44 patients (27 men and 17 women; mean age, 58 years old) with CT colonography by interpreting the axial images and using three-dimensional rendering for problem solving only. The CT scans were interpreted by two radiologists who were unaware of patients' histories as revealed by colonoscopic findings. The findings on colonography were compared with those of conventional colonoscopy to determine sensitivity, specificity, time spent on interpretation, and confidence of interpretation. RESULTS: Colonoscopy showed normal findings in 28 patients and 22 polyps in the remaining 16 patients. Six polyps were 8 mm or larger, three were 5-7 mm, and 13 were 5 mm or smaller. The findings of the two observers revealed an overall sensitivity of 50% and 38%, respectively, and a specificity of 93% and 86%, respectively. Sensitivity for polyps larger than 8 mm was 83% and specificity was 100% for both observers. The average amount of time spent on interpretation was 28 min 30 sec (range, 14-65 min). Both observers used the endoluminal view for differentiating folds from polyps in 23 (52%) of 44 patients, which had only minimal impact on interpretation time. CONCLUSION: CT colonography can be performed and the images interpreted using currently available hardware and software by initially using the axial images to search for polyps of significant size. Endoluminal views should be used only when necessary to help distinguish normal folds from fixed raised lesions that are suggestive of polyps.

Imaging of pancreatic transplantation using portal venous and enteric exocrine drainage.

OBJECTIVE: We describe the normal radiologic appearance of pancreatic allografts transplanted using portal venous drainage with enteric drainage of exocrine secretions. We also describe the radiologic appearance of postsurgical complications. MATERIALS AND METHODS: Of 56 patients who received pancreatic transplants using the portal-enteric technique, 24 patients subsequently required radiologic examination for suspected complications involving the pancreatic allograft. Twenty-three patients underwent CT scanning; a total of 58 CT scans were obtained. Nine abdominal sonograms were obtained in five patients, and one patient underwent angiography. The radiologic appearance of each transplant and the complications were analyzed retrospectively and correlated with the clinical course. RESULTS: The most common indications for CT scanning were fever, elevated levels of serum amylase, and evaluation or follow-up of fluid collections. CT showed the normal and abnormal anatomy of the allograft. Abnormal findings seen in the 58 CT scans included fat stranding (30 scans), ascites (21 scans), peripancreatic fluid or pseudocyst (13 scans), and heterogeneity of the allograft (five scans). One patient had pancreatic infarction with pneumatosis and pneumoperitoneum. The allograft was obscured by bowel gas on three sonograms. Four sonograms showed no abnormalities (one Doppler sonogram showed the arterial supply and venous drainage), and one sonogram showed a pseudocyst. In the one patient who underwent angiography, imaging showed no arterial blood flow to the transplant. CONCLUSION: Pancreatic transplantation with portal venous drainage and enteric drainage of exocrine secretions and the complications of such transplantation were revealed with CT, sonography, and angiography. Knowledge of normal anatomic configuration will allow proper interpretation of normal and abnormal findings.

Helical CT examination of potential kidney donors.

OBJECTIVE: The purpose of our study was to evaluate helical CT using axial, coronal, and three-dimensional (3D) reconstruction in the examination of potential kidney donors and to compare the results with angiography and surgery when possible. We also reviewed previously published reports. SUBJECTS AND METHODS: Thirty-six patients underwent unenhanced and enhanced helical CT (3-mm collimation, 150-170 ml of i.v. contrast material injected at 4 ml/sec; pitch 1.5; 17-sec scan delay) with coronal and 3D shaded-surface-display reconstructions made from 1.5-mm overlapping reconstructions. All CT scans were interpreted independently of each other by two observers unaware of other findings. A third observer, who was aware of other findings, also interpreted the images. Results were compared with angiography (24 cases) and surgery (24 cases). Our results are compared with those of other investigators. RESULTS: Axial CT was the best method for detecting accessory arteries (24%) and early branching (10%); it also detected relevant venous and ureteral anatomy and incidental findings. The coronal and 3D images rarely added information that resulted in changed patient treatment. CT findings were concordant with those of digital angiography in 89% of kidneys and were 98% concordant with surgery. CONCLUSION: Helical CT can show arterial, venous, and ureteral anatomy and can also show important incidental findings. If only helical CT is used, a few small accessory vessels and an occasional renal artery stenosis may be missed. Axial images are generally diagnostic and may be supplemented by multiplanar and 3D images read concurrently.