Vasovagal reactions in CT colonography

We describe the occurrence of vasovagal reaction in two patients who underwent CT colonography (CTC). The patients, asymptomatic, were submitted to CTC in one case after right colectomy and in one case for screening purposes. The vasovagal symptoms occurred after pneumocolon and acquisition in the prone decubitus, and included headache, hypotension, bradycardia, cold sweat and pallor, nausea, and diaphoresis. Abdominal pain was also referred. All symptoms resolved within 30 min to 3 h from their onset. In all cases the vasovagal reaction occurred after prone decubitus. CTC images showed a significant distension of the small bowel. Vasovagal reactions are potential complications of CTC.     

Electronic colon-cleansing for CT colonography: diagnostic performance

Background  The purpose of this study was to evaluate whether an electronic-colonic-cleansing (ECC) algorithm is beneficial for the diagnostic performance compared to a CT colonography (CTC) evaluation without electronic cleansing in tagged datasets. Methods  Two blinded readers evaluated CTC datasets from 79 patients with 153 colorectal polyps confirmed by optical colonoscopy. Cases were read in a randomized order with and without the use of electronic colon-cleansing software. Per-polyp sensitivity, per-polyp/per-patient specificity and reading times (with and without ECC) have been calculated and reported. Results  Per-polyp sensitivity for polyps >6 mm without using ECC was 60.4% (Reader 1: 59.7%, Reader 2: 61.1%), while polyps >10 mm were detected with a sensitivity of 58.3% (Reader 1: 66.7%, Reader 2: 50%). On electronically cleansed datasets, the sensitivity was 73.6% (Reader 1: 76.4%; Reader 2: 70.8%) for polyps >6 mm and 83.3% (Reader 1: 83.3%; Reader 2: 83.3%), respectively. Per-patient specificity was 75% without using cleansing (Reader 1: 68%, Reader 2: 82%) and 81.5% using ECC (Reader 1: 86%, Reader 2: 77%). Conclusion  Reading CTC cases using ECC software improves sensitivity in detecting clinically relevant colorectal polyps.     

CT colonography: surveillance in patients with a history of colorectal cancer

Colorectal cancer is a leading cause of morbidity and mortality in the United States. It is also a disease that is preventable if precursor adenomatous polys are removed. Once a diagnosis of colorectal cancer is made, surgical resection is the only means of cure. The ability to resect colorectal cancer for cure is largely dependent upon the stage of tumor at presentation. Once a patient has been treated for colorectal cancer with surgery and in some cases neo-adjuvant or adjuvant therapy, they will present for follow-up. Surveillance is performed on these patients in order to detect local recurrence that if detected early can be surgically resected for cure. Surveillance also allows detection of distant metastatic disease that may in some cases also be cured with resection. Finally, surveillance of the remaining colon is important to detect the development of new or metachronous adenomatoid polyps that if left in place could develop into new colon cancers. Imaging can play a part in patient surveillance to detect recurrent disease at extracolonic sites as well as the development of new colonic lesions. CT colonography is a promising tool for surveillance in patients with a history of colorectal cancer.     

Comparison of routine and unprepped CT colonography augmented by low fiber diet and stool tagging: a pilot study

Background We performed a pilot study examining the feasibility of a new unprepped CT colonography (CTC) strategy: low fiber diet and tagging (unprepped) vs. low fiber diet, tagging and a magnesium citrate cleansing preparation (prepped). Prior reports of tagging were limited in that the residual stool was neither measured and stratified by size nor did prior reports subjectively evaluate the ease of interpretation by a reader experienced in interpreting CTC examinations. Methods Prospective randomized to unprepped n = 14 and prepped n = 14. Colonic segments were subjectively evaluated for residual stool that would potentially interfere with interpretation. Scores were given in the following categories: percentage of residual stool that was touching or nearly touching mucosa, the largest piece of retained stool, effectiveness of tagging, height of residual fluid, degree of distention, ease of interpretation, and reading time. Results Ease of the CT read (scale where 4 = optimal read) averaged 1.3 for the unprepped group and 2.3 for the prepped group. The mean read time averaged 17.5 min for unprepped and 17.9 min for prepped. The degree of distention (scale where 4 = well distended) averaged 3.7 for unprepped and 3.6 for prepped. Supine and prone images combined, the unprepped group had 160 segments with stool; prepped group had 58 segments. The amount of stool covering the mucosa in all segments averaged 1.6 (33%–66% coverage) in the unprepped group and 0.35 (<33% mucosal coverage) in the prepped group. The mean size of the largest piece of stool was 33.67 mm for unprepped and 4.01 mm for prepped. Percentage of tagged stool was not significantly different between the groups (range of 94–98%). The height of residual fluid averaged 8.37 mm for unprepped and 13.4 mm for prepped. Three polyps in three patients were found during optical colonoscopy (OC) in the unprepped group (5, 6, and 10 mm), none of which were prospectively detected at CTC. Three polyps in three patients were detected during OC in the prepped group (5, 10, and 15 mm), two of which were prospectively detected at CTC. Two false-positive lesions were observed at CTC in one patient in the prepped group. Conclusion There was more stool in the unprepped group and while this factor did not slow down the reading time, it made the examination subjectively harder to interpret and likely caused the three polyps in this group to be missed. We conclude that a truly unprepped strategy that leaves significant residual stool, even if well tagged, is not desirable.     

CT colonography: contrast enhancement of benign and malignant colorectal lesions versus fecal residuals

We retrospectively reviewed the computed tomographic colonographic datasets of 22 patients. Mean attenuation values of benign polyps before and after contrast administration were 30 ± 15 HU and 90 ± 18 HU, respectively. Mean attenuation values of colorectal cancer before and after contrast administration were 43 ± 15 HU and 124 ± 18 HU, respectively. The mean attenuation value of solid fecal residuals was 43 ± 15 HU. The difference in attenuation value between precontrast and postcontrast studies of polyps was statistically significant (mean 60 HU, p < 0.01); the same was true for colorectal cancer (mean 81 HU, p < 0.01). The difference between postcontrast density of polyps and cancer with respect to density of solid fecal residuals was statistically significant (p < 0.01). The use of contrast medium could be of help in computed tomographic colonography for discriminating polypoid benign lesions and colorectal cancer from fecal residuals.     

Computer-aided detection for CT colonography: update 2007

Computed tomographic colonography (CTC) is an emerging technique for polyp detection in the colon. However, lesion detection can be challenging due to insufficient patient preparation, chosen CT technique or reader imperfection. The primary goal of computer-aided detection (CAD) for CTC is locating possible polyps, and presenting the reader with these polyp candidates. Other goals are sensitivity improvement and reduction of reading time and inter-observer variability. The multistep CAD procedure typically consists of segmentation of the colonic wall (e.g. region growing); selection of intermediate polyp candidates (curvature analysis, sphere fitting, normal analysis, slope density function ...); classification of final candidates for detection and listing suspicious polyps (location, size and volume). Remaining task for the radiologist is the validation or rejection of the polyp candidates. State-of-the-art CAD systems should require minimal or even no user interaction for the extraction of the colonic wall, offer a computation time less than 10-20 min and high sensitivity and specificity for different polyp sizes and shapes, with a low number of false positives. These systems have the potential to increase radiologist's performance and to decrease inter-reader variability. Besides CAD key techniques we also discuss new developments in CAD and describe recent applications facilitating CTC.     

CT Colonography for Colorectal Cancer Screening

CT colonography (CTC), also known as “virtual colonoscopy,” is a minimally invasive complementary screening method to optical colonoscopy (OC) that is growing in its availability and utilization. As a screening test, it is typically performed in outpatient radiology sites. Examinations are monitored by a radiologist trained in its performance. CTC is selectively performed in a hospital setting for patients with comorbidities that render them at a high risk for OC, such as cardiac disease or screening before organ transplantation. In cases where OC is unsuccessful for at complete colonic inspection, CTC can be performed the same day after OC or planned at a later date to evaluate the remainder of the colon. Candidates for CTC must be able to follow instructions and tolerate colonic insufflation. Patients who have fecal incontinence or significant rectal prolapse are not candidates for CTC because of limited ability to maintain colonic distension. Finally, strict adherence to the recommended colonic preparation before CTC is critical to the examination's success.     

Flat polyps of the colon: accuracy of detection by CT colonography and histologic significance

There is controversy regarding the prevalence, clinical importance, and appropriate screening methods for nonpolypoid (flat and depressed) polyps in the colon. Investigators in Japan have reported higher prevalence of nonpolypoid adenomas in the general population and there have been several reports of higher incidence of high-grade dysplasia in flat adenomas in these Eastern studies. Historically, many Western gastroenterologists have been skeptical of these findings and there have been conflicting studies regarding the prevalence of flat adenomas and incidence of high-grade dysplasia in these lesions. Multiple reasons have been postulated for this apparent difference. Therefore further research into this topic is needed to clarify these issues. In this article we will review the controversy related to the definitions and clinical importance of nonpolypoid neoplasms in the colon, demonstrate the appearance of these unique lesions at CT colonography (CTC) and discuss the accuracy of CTC.     

Colonic perforation during screening CT colonography using automated CO2 insufflation in an asymptomatic adult

CT colonography has become a potential alternative technique to optical colonoscopy for the detection of colorectal polyps and cancer. While considered safer than optical colonoscopy, CT colonography is not without risk. We report a case of colonic perforation during CT colonography using automated CO₂ insufflation and present procedural changes to help minimize the adverse effects of perforation when it occurs.     

On the convergence of EM-like algorithms for image segmentation using Markov random fields

Inference of Markov random field images segmentation models is usually performed using iterative methods which adapt the well-known expectation–maximization (EM) algorithm for independent mixture models. However, some of these adaptations are ad hoc and may turn out numerically unstable. In this paper, we review three EM-like variants for Markov random field segmentation and compare their convergence properties both at the theoretical and practical levels. We specifically advocate a numerical scheme involving asynchronous voxel updating, for which general convergence results can be established. Our experiments on brain tissue classification in magnetic resonance images provide evidence that this algorithm may achieve significantly faster convergence than its competitors while yielding at least as good segmentation results.     

Constrained linear basis sets for HRF modelling using Variational Bayes

FMRI modelling requires flexible haemodynamic response function (HRF) modelling, with the HRF being allowed to vary spatially and between subjects. To achieve this flexibility, voxelwise parameterised HRFs have been proposed; however, inference on such models is very slow. An alternative approach is to use basis functions allowing inference to proceed in the more manageable General Linear Model (GLM) framework. However, a large amount of the subspace spanned by the basis functions produces nonsensical HRF shapes. In this work we propose a technique for choosing a basis set, and then the means to constrain the subspace spanned by the basis set to only include sensible HRF shapes. Penny et al. showed how Variational Bayes can be used to infer on the GLM for FMRI. Here we extend the work of Penny et al. to give inference on the GLM with constrained HRF basis functions and with spatial Markov Random Fields on the autoregressive noise parameters. Constraining the subspace spanned by the basis set allows for far superior separation of activating voxels from nonactivating voxels in FMRI data. We use spatial mixture modelling to produce final probabilities of activation and demonstrate increased sensitivity on an FMRI dataset.     

Novel methods for 3D postoperative analysis of total knee arthroplasty using 2D-3D image registration

Background

Radiographic evaluation of patients after total knee arthroplasty is an important tool for assessing the correct position of the implants and the axis of limb alignment because this will determine long-term durability of the implants. Recently, 2D–3D medical image registration methods are developed for 3D postoperative analysis of total knee arthroplasty. However, most of these techniques have focused only on knee implants.

Methods

A 2D–3D medical image registration is implemented to compute the 3D positions of not only implants but also lower limb bones. The following 3D postoperative analysis methods for total knee arthroplasty are presented in this paper: (1) automatic calculation of relative angles of implants and bones, (2) assessment of external rotation angles of inserted implants, and (3) measurement and comparison of both flexion–extension gap balances. Finally these methods have been applied in five patients who underwent total knee replacements.

Findings

A practical method that can evaluate the patient's knee conditions has been successfully developed. The repeatability and accuracy of 2D–3D registration were around 0.2 mm as obtained from the tests using model bones. Based on the 3D information, the novel methods of postoperative analysis were proposed and successfully applied to the patients.

Interpretation

The 3D positions for both knee implants and lower limb bones can be calculated in order to perform comprehensive postoperative analyses of total knee arthroplasty. The proposed analyses of the postoperative evaluations facilitated various 3D evaluations of the status of implants, alignment of lower limb and gap balances which were not previously feasible.     

Structured patch model for a unified automatic and interactive segmentation framework

We present a novel interactive segmentation framework incorporating a priori knowledge learned from training data. The knowledge is learned as a structured patch model (StPM) comprising sets of corresponding local patch priors and their pairwise spatial distribution statistics which represent the local shape and appearance along its boundary and the global shape structure, respectively. When successive user annotations are given, the StPM is appropriately adjusted in the target image and used together with the annotations to guide the segmentation. The StPM reduces the dependency on the placement and quantity of user annotations with little increase in complexity since the time-consuming StPM construction is performed offline. Furthermore, a seamless learning system can be established by directly adding the patch priors and the pairwise statistics of segmentation results to the StPM. The proposed method was evaluated on three datasets, respectively, of 2D chest CT, 3D knee MR, and 3D brain MR. The experimental results demonstrate that within an equal amount of time, the proposed interactive segmentation framework outperforms recent state-of-the-art methods in terms of accuracy, while it requires significantly less computing and editing time to obtain results with comparable accuracy.     

A new 3-D diagnosis strategy for duodenal malignant lesions using multidetector row CT,CT virtual duodenoscopy,duodenography, and 3-D multicholangiography

Recent advances in multidetector row computed tomography (MD-CT) technology provide new opportunities for clinical diagnoses of various diseases. Here we assessed CT virtual duodenoscopy, duodenography, and three-dimensional (3D) multicholangiography created by MD-CT for clinical diagnosis of duodenal malignant lesions. The study involved seven cases of periduodenal carcinoma (four ampullary carcinomas, two duodenal carcinomas, one pancreatic carcinoma). Biliary contrast medium was administered intravenously, followed by intravenous administration of an anticholinergic agent and oral administration of effervescent granules for expanding the upper gastrointestinal tract. Following intravenous administration of a nonionic contrast medium, an upper abdominal MD-CT scan was performed in the left lateral position. Scan data were processed on a workstation to create CT virtual duodenoscopy, duodenography, 3D multicholangiography, and various postprocessing images, which were then evaluated for their effectiveness as preoperative diagnostic tools. Carcinoma location and extent were clearly demonstrated as defects or colored low-density areas in 3-D multicholangiography images and as protruding lesions in virtual duodenography and duodenoscopy images. These findings were confirmed using multiplanar or curved planar reformation images. In conclusion, CT virtual duodenoscopy, doudenography, 3-D multicholangiography, and various images created by MD-CT alone provided necessary and adequate preoperative diagnostic information.     

Comparison of ventilation/perfusion scintigraphy and helical CT for diagnosis of pulmonary embolism; strategy using clinical data and ancillary findings

Study objective: To address the question whether ventilation/perfusion scintigraphy (SCINT) or helical computed tomography (CT) should be the first hand method for diagnosis of pulmonary embolism (PE). Setting: Departments of radiology, nuclear medicine and internal medicine of a large university hospital. Patients: During 3 years all 128 patients examined for PE with both methods were analysed. The strategy of interpretation behind original clinical reports, i.e. clinical CT and clinical SCINT, was based upon basic criteria for PE, ancillary findings and information from the referring doctor and from previous examinations. Reviewed SCINT and CT reports were obtained from experts in each field blinded to clinical and laboratory data. The findings with respect to PE were classified as no PE, PE or non‐diagnostic. Other pathology than PE was described. A final diagnosis serving as reference was based upon CT, SCINT and other information including clinical follow for 6–24 months. Methods: Planar SCINT was made with ventilation always preceding perfusion. CT was made with contrast injection using 3 mm collimation and table feed of 3 mm s^?1. Results: PE was diagnosed in 32 patients. For clinical and reviewed SCINT sensitivity was 91 and 97%, specificity 96 and 100% and rate of non‐diagnostic findings 10 and 9%, respectively. For clinical and reviewed CT sensitivity was 81 and 78%, specificity 99 and 100% and non‐diagnostic findings was observed in 8 and 1%, respectively. In patients with PE, concordant positive results were obtained with both modalities in 23 of 32 patients (72%). Conclusion: SCINT remains the first hand method because its high sensitivity, general feasibility, low radiation burden and low rate of non‐diagnostic findings in our setting. CT is indispensable when SCINT is not available or its result non‐diagnostic.