Unravelling the effect of data augmentation transformations in polyp segmentation

PURPOSE

Data augmentation is a common technique to overcome the lack of large annotated databases, a usual situation when applying deep learning to medical imaging problems. Nevertheless, there is no consensus on which transformations to apply for a particular field. This work aims at identifying the effect of different transformations on polyp segmentation using deep learning.

METHODS

A set of transformations and ranges have been selected, considering image-based (width and height shift, rotation, shear, zooming, horizontal and vertical flip and elastic deformation), pixel-based (changes in brightness and contrast) and application-based (specular lights and blurry frames) transformations. A model has been trained under the same conditions without data augmentation transformations (baseline) and for each of the transformation and ranges, using CVC-EndoSceneStill and Kvasir-SEG, independently. Statistical analysis is performed to compare the baseline performance against results of each range of each transformation on the same test set for each dataset.

RESULTS

This basic method identifies the most adequate transformations for each dataset. For CVC-EndoSceneStill, changes in brightness and contrast significantly improve the model performance. On the contrary, Kvasir-SEG benefits to a greater extent from the image-based transformations, especially rotation and shear. Augmentation with synthetic specular lights also improves the performance.

CONCLUSION

Despite being infrequently used, pixel-based transformations show a great potential to improve polyp segmentation in CVC-EndoSceneStill. On the other hand, image-based transformations are more suitable for Kvasir-SEG. Problem-based transformations behave similarly in both datasets. Polyp area, brightness and contrast of the dataset have an influence on these differences.

Adult Jejuno-jejunal intussusception due to inflammatory fibroid polyp: A case report and literature review

RATIONALE

Intussusception is defined as the invagination or telescoping of a proximal portion of the intestine into the distal portion of the intestine. Intussusception can occur at any age but is more common among children. Most cases of intussusception in adults have a pathological lead point. Inflammatory fibroid polyp (IFP) is a rare benign tumor-like lesion arising from the submucosa of the gastrointestinal tract that can cause intussusception in adults. Here, we report a case of adult intussusception due to IFP. We also present a literature review of 31 reports including 34 cases between 2012 and December 2019, which shows a mean age of 45.4 ± 14.2 years and female dominance (23/34) of intussusception due to IFP.

PATIENT CONCERNS

A 47-year-old man presented with a half-day history of epigastric abdominal pain. Physical examination revealed distension and tenderness of the upper abdomen. Computed tomography (CT) of the abdomen and pelvis demonstrated intussusception of the jejunum along with a suspicious jejunal mass associated with mesenteric lymphadenopathies.

DIAGNOSIS

Intussusception of the jejunum along with a suspicious jejunal mass, and histopathological examination of the resected specimen showed IFP.

INTERVENTIONS

The patient underwent emergency laparotomy. The intussusception was resected without attempts for reduction.

OUTCOMES

The postoperative period was uneventful, and the patient was discharged on the fourth postoperative day.

LESSONS

Intussusception in adults is rare, especially that secondary to IFP. The most commonly used diagnostic tool for adult intussusception is abdominal CT, and the optimal management is resection of the involved bowel segment without reduction if malignancy cannot be ruled out.

Advanced endoscopy technologies to improve the detection and characterisation of colorrectal polyps

Colorectal cancer is a major health problem. An improvement to its survival has been demonstrated by performing colonoscopy screenings and removing its precursor lesions: polyps. However, colonoscopy is not infallible and multiple strategies have been proposed aimed at improving the quality thereof. This report describes the endoscopic systems available to improve the detection and characterization of polyps, the different classifications for histological prediction and the current indications of advanced endoscopic diagnostic techniques.

Endoscopic features, pathological correlates and possible origin of foveolar gastric metaplasia presenting as a duodenal polyp

It has recently been shown that duodenal foveolar gastric metaplasia (FGM) sometimes presents as a polyp. The mechanism by which FGM develops into a polypoid lesion is unknown and it is unclear whether this form of FGM is indistinguishable from other polypoid lesions or whether endoscopists do not recognize it because they are unfamiliar with it. We identified and retrieved archival cases of FGM endoscopically suspicious for adenomatous polyp and examined their pathological, clinical and endoscopic features. Endoscopic features of the 13 identified FGMs presenting as polyps were heterogeneous and overlapping with those of adenomatous polyps. FGM was frequently associated with mucosal and submucosal Brunner's glands, but defining and recognizing hyperplasia of these glands remains difficult. Other pathological features could not explain the development of a polypoid lesion. The endoscopic features of FGM polyps are non-specific, overlapping with those of adenomatous polyps. FGM polyps probably acquire their polypoid aspect due to association with Brunner's gland hyperplasia (BGH), which also arises due to chronic inflammation and damage. Because BGH is ill-defined and difficult to recognize, while FGM is diagnosed easily, this type of polypoid lesions has until now only been recognized based on the presence of FGM, although FGM is most likely a secondary phenomenon and not the primary cause of the polyp.

Reliable Label-Efficient Learning for Biomedical Image Recognition

The use of deep neural networks for biomedical image analysis requires a sufficient number of labeled datasets. To acquire accurate labels as the gold standard, multiple observers with specific expertise are required for both annotation and proofreading. This process can be time-consuming and labor-intensive, making high-quality, and large-annotated biomedical datasets difficult. To address this problem, we propose a deep active learning framework that enables the active selection of both informative queries and reliable experts. To measure the uncertainty of the unlabeled data, a dropout-based strategy is integrated with a similarity criterion for both data selection and random error elimination. To select the reliable labelers, we adopt an expertise estimator to learn the expertise levels of labelers via offline-testing and online consistency evaluation. The proposed method is applied to classification tasks on two types of medical images including confocal endomicroscopy images and gastrointestinal endoscopic images. The annotations are acquired from multiple labelers with diverse levels of expertise. The experiments demonstrate the efficiency and promising performance of the proposed method compared to a set of baseline methods.

The Serrated Polyp Pathway: Is It Time to Alter Surveillance Guidelines?

PURPOSE OF REVIEW

In this manuscript, we review current surveillance guidelines for serrated polyps (SPs) and discuss how recent studies inform the selection of appropriate surveillance intervals for patients with SPs.

RECENT FINDINGS

Large and/or proximal SPs, particularly sessile serrated polyps (SSPs), are associated with increased risk of both synchronous and metachronous neoplasia, including advanced adenomas and colorectal cancer (CRC). Persons harboring multiple SSPs or dysplastic SSPs are at the highest risk. Moreover, a high percentage of large and/or proximal SPs are reclassified as SSPs when read by trained gastrointestinal pathologists, even if they were originally reported as hyperplastic polyps. These findings support the adoption of surveillance guidelines that prescribe closer surveillance of large and/or proximal SPs, regardless of subtype. SSPs remain a challenge to reliably identify, resect, and diagnose via histology. The increased risk of future neoplasia in patients with SSPs is likely driven by a combination of underdetection, inadequate removal, misclassification, and biology. Until further evidence emerges, we support guidelines that recommend close surveillance of patients with a history of large and/or proximal SPs and SSPs specifically in order to mitigate the threat of interval CRC.

Narrow-band imaging endoscopy is advantageous over conventional white light endoscopy for the diagnosis and treatment of children with Peutz-Jeghers syndrome

Using narrow-band imaging (NBI), the micro morphology of polyp surfaces was visualized at high resolution when the contrast between the lesions and the surrounding tissue areas was intensified. The objective of the study was to compare NBI and conventional white light endoscopy (C-WLI) for diagnostic efficacy and treatment of children with Peutz-Jeghers syndrome (PJS), a rare autosomal dominant-genetic disease.We retrospectively analyzed the clinical data of 18 patients diagnosed with PJS and 72 diagnosed with juvenile polyps during the same time period. Endoscopy was used to observe the morphology of polyps, which were classified according to the Kudo pit pattern. Eighty-two and 76 polyps were resected using endoscopy from PJS and juvenile polyp patients, respectively, and pathologically examined for comparison with NBI or C-WLI endoscopic observations.With NBI, 57 (69.5%) type I, 10 (12.2%) type II, and 13 (15.9%) type III polyps were identified in 82 (100%) polyps excised from 18 PJS patients, whereas 69 (88.5%) type I, 5 (6.4%) type II and 4 (5.1%) type III were identified in 78 (100%) of polyps removed from 72 juvenile polyp patients. The combined ratios of types II and III were higher in PJS (28.1%) than in juvenile polyp patients (11.5%). Pathological diagnosis identified 69 of 82 PJS polyps as hamartoma and 13 as adenoma, whereas NBI detected 10 adenomas and C-WLI only 4. The sensitivity, specificity, and accuracy of NBI in the diagnosis of adenoma were 76.9%, 97.1%, and 93.9%, respectively, whereas those of C-WLI were 30.8%, 94.2%, and 84.1%.NBI endoscopy had higher sensitivity and specificity than C-WLI for the diagnosis of adenomatous polyps and is recommended for the diagnosis, characterization, and resection of polyps in children with PJS.

Epidemiological aspects of endoscopic resections of colorectal polyps in patients at an endoscopy training center in the Santos region, Brasil

INTRODUCTION

Colorectal polyps are structures that project from the surface of the mucosal layer of the large intestine. They are classified as neoplastic or non-neoplastic. Early detection of pre-neoplastic lesions is important for preventing colorectal cancer. These can be resected so as to decrease the morbidity and mortality rates. Colonoscopy is the gold-standard procedure for diagnosing and resecting precursor lesions.

OBJECTIVE

To evaluate the epidemiological, endoscopic and histological aspects of endoscopic resection of lesions of the colon and rectum at a training center.

MATERIALS AND METHOD

A search was conducted in the database of our institution covering the period from January 2011 to July 2014. Cases that underwent endoscopic resection of polyps and/or colorectal lesions were selection. The following variables were defined: general data on the patients (age, gender and indication from the examination) and data on the polypoid lesion (number, histological type and topographic distribution).

RESULTS

678 lesions were identified in 456 examinations. Regarding sex, 242 (53.1%) were female and 214 (46.9%) were male. The mean age was 64.54 years, with extremes of 5 and 94 years. The most frequent locations were the rectum (21%) and sigmoid (20%). Histologically, 34.7% were hyperplastic polyps and 58.9% were adenomatous polyps, of which 74.1% were tubular, 10.6% tubulovillous, 2% villous and 13% indeterminate; and 1.7% were adenocarcinomas. In 65.4% of the cases, the examination showed that only one polyps was present, while 34.6% had two or more lesions.

CONCLUSION

In our clinic, with a mean of 250 examinations/month, the parameters evaluated were compatible with the results reported in the literature.

Inflammatory fibroid polyps of the gastrointestinal tract: a 14-year CT study at a single institution

PURPOSE

To investigate the computed tomography (CT) features of pathology-proven inflammatory fibroid polyps (IFPs) in the gastrointestinal tract.

METHODS

This retrospective series study included 27 patients with pathology-proven IFPs in the stomach (n = 16), small (n = 9), and large (n = 2) intestine, who underwent contrast-enhanced CT. Two radiologists reviewed the CT images of the patients to determine in consensus the long diameter, shape, margin, contour, and growth pattern of the lesions, the presence of an ulcer and overlying mucosal hyperenhancement, the lesion enhancement patterns including the homogeneity and the degree of contrast enhancement, and the presence of intussusception and obstruction. The CT results and clinical data of the gastric and intestinal lesions were compared.

RESULTS

The IFPs typically manifested as well-defined (89%), round or ovoid (81%), slightly lobulated-contoured (70%) masses with a purely endoluminal growth pattern (96%) and an overlying mucosal hyperenhancement (67%). Lesion homogeneity and the degree of contrast enhancement varied. The intestinal IFPs were significantly larger (3.5 vs. 2 cm), more symptomatic (82% vs. 19%), and more frequently associated with intussusception (73% vs. 0%) and obstruction (46% vs. 6%) than the gastric lesions (p ≤ 0.027).

CONCLUSIONS

The characteristic CT features of IFPs were well-defined, round or ovoid, lobulated-contoured, and endoluminal masses with overlying mucosal hyperenhancement and various enhancement patterns. IFP should be included in the differential diagnosis of patients with a soft-tissue mass in the gastrointestinal tract, especially if a large endoluminal mass in the small intestine is accompanied by intussusception.

Ultrasound virtual endoscopy: Polyp detection and reliability of measurement in an in vitro study with pig intestine specimens

AIM: To present our initial experience regarding the feasibility of ultrasound virtual endoscopy (USVE) and its measurement reliability for polyp detection in an in vitro study using pig intestine specimens.

METHODS: Six porcine intestine specimens containing 30 synthetic polyps underwent USVE, computed tomography colonography (CTC) and optical colonoscopy (OC) for polyp detection. The polyp measurement defined as the maximum polyp diameter on two-dimensional (2D) multiplanar reformatted (MPR) planes was obtained by USVE, and the absolute measurement error was analyzed using the direct measurement as the reference standard.

RESULTS: USVE detected 29 (96.7%) of 30 polyps, remaining a 7-mm one missed. There was one false-positive finding. Twenty-six (89.7%) of 29 reconstructed images were clearly depicted, while 29 (96.7%) of 30 polyps were displayed on CTC with one false-negative finding. In OC, all the polyps were detected. The intraclass correlation coefficient was 0.876 (95%CI: 0.745-0.940) for measurements obtained with USVE. The pooled absolute measurement errors ± the standard deviations of the depicted polyps with actual sizes ≤ 5 mm, 6-9 mm, and ≥ 10 mm were 1.9 ± 0.8 mm, 0.9 ± 1.2 mm, and 1.0 ± 1.4 mm, respectively.

CONCLUSION: USVE is reliable for polyp detection and measurement in in vitro study.

[Inflammatory fibroid polyp as the cause of upper gastrointestinal bleeding.]

The inflammatory fibroid polyp or Vanek's tumor is a benign polypoid lesion that can be found in any section of the digestive tract. It is most frequently located in the stomach and when found in this organ, the symptoms include upper gastrointestinal bleeding (often silent), anemia and melena. The case described below, on the contrary, presents a gastric inflammatory fibroid polyp which developed acute upper gastrointestinal bleeding and required surgical treatment.

Approach to the endoscopic resection of duodenal lesions

Duodenal polyps or lesions are uncommonly found on upper endoscopy. Duodenal lesions can be categorized as subepithelial or mucosally-based, and the type of lesion often dictates the work-up and possible therapeutic options. Subepithelial lesions that can arise in the duodenum include lipomas, gastrointestinal stromal tumors, and carcinoids. Endoscopic ultrasonography with fine needle aspiration is useful in the characterization and diagnosis of subepithelial lesions. Duodenal gastrointestinal stromal tumors and large or multifocal carcinoids are best managed by surgical resection. Brunner's gland tumors, solitary Peutz-Jeghers polyps, and non-ampullary and ampullary adenomas are mucosally-based duodenal lesions, which can require removal and are typically amenable to endoscopic resection. Several anatomic characteristics of the duodenum make endoscopic resection of duodenal lesions challenging. However, advanced endoscopic techniques exist that enable the resection of large mucosally-based duodenal lesions. Endoscopic papillectomy is not without risk, but this procedure can effectively resect ampullary adenomas and allows patients to avoid surgery, which typically involves pancreaticoduodenectomy. Endoscopic mucosal resection and its variations (such as cap-assisted, cap-band-assisted, and underwater techniques) enable the safe and effective resection of most duodenal adenomas. Endoscopic submucosal dissection is possible but very difficult to safely perform in the duodenum.

A computer-aided diagnostic system for intestinal polyps identified by wireless capsule endoscopy

Small bowel polyps present in images acquired by wireless capsule endoscopy are more difficult to detect using computer-aided diagnostic (CAD) systems. We aimed to identify the optimum morphological characteristics that best describe a polyp and convert them into feature vectors used for automatic detection of polyps present in images acquired by wireless capsule endoscopy (WCE). We prospectively included 54 patients with clinical indications for WCE. Initially, physicians analyzed all images acquired, identifying the frames that contained small bowel polyps. Subsequently, all images were analyzed using an automated computer-aided diagnostic system designed and implemented to convert physical characteristics into vectors of numeric values. The data set was completed with texture and color information, and then analyzed by a feed forward back propagation artificial neural network (ANN) trained to identify the presence of polyps in WCE frames. Overall, the neural network had 93.75% sensitivity, 91.38% specificity, 85.71% positive predictive value (PPV) and 96.36% negative predictive value (NPV). In comparison, physicians' diagnosis indicated 94.79% sensitivity, 93.68% specificity, 89.22% PPV and 97.02% NPV, thus showing that ANN diagnosis was similar to that of human interpretation. Computer-aided diagnostic of small bowel polyps, based on morphological features detection methods, emulation and neural networks classification, seems efficient, fast and reliable for physicians.

[Experience of low volume split-dose bowel preparation for computed tomography colonography]

We propose a new preparation method for the computed tomography colonography (CTC). This method consists of giving a small volume (400 ml) of cleansing solution on the day before the examination and the same volume of solution on the day of the examination [low volume split-dose (LVSD) method]. Using this method, we compared the volume of residual fluid in the colon, the CT value of the residual fluid, and the quality of stool tagging with those for patients undergoing the conventional bowel preparation method. Polyp detectability of the CTC using this method and the acceptability of the preparation were also investigated. The volume of residual fluid in the colon with this method was smaller than that with the conventional method. The CT value of the residual fluid with this method was higher than that with the conventional method. Visual assessment of the quality of stool tagging with this method gave similar results to those obtained using the conventional method. The sensitivities were 95% for 5-10 mm polyps and 100% for polyps larger than 10 mm. The PPVs were 91% for 5-10 mm polyps and 100% for polyps larger than 10 mm. These results appear to be as good as in previous reports. In the questionnaires, about 80% of the answers were favorable regarding the volume and the taste of laxative. We conclude that LVSD bowel preparation method for CTC maintains polyp detectability and is better tolerated.

Surgical alternatives in the treatment of intestinal intussusceptions resulting from polyps in adults

Adult intussusception is an uncommon disease requiring surgical intervention. The aim of this study is to discuss the surgical alternatives and share our experience in the treatment of adult patients with intussusceptions formed as a result of polyps. The retrospective study included 16 adult patients who underwent surgery after the diagnosis of intestinal invaginations resulting from polyps between the years 2000 and 2011. Sixteen patients (seven males and nine females; mean age, 48.18 years; range, 18 to 76 years) presented with intestinal intussusceptions. Although a preoperative diagnosis was carried out in 11 (68.75%) patients, the diagnosis was made intraoperatively in five patients (31.25%). Among the patients, seven (43.8%) had undergone emergency surgeries and nine (52.8) had elective surgery. The invagination in 12 patients (75%) was located in the small intestine, in two patients (12.5%) in the colon, and in a further two patients (12.5%), it was ileocecally located. Ten patients (62.5%) had segmental resection + anastomosis; three patients underwent (18.8%) segmental resection + enterostomy, and three (18.8%) received hemicolectomies. In adults, surgical treatment is always the primary option in intussusceptions resulting from polyps. Although the surgical method of choice in colonically located ones is en bloc resection without reduction, because the polyps located in the small intestine are usually of a benign nature, segmental resection with reduction should be performed in elective surgery and segmental resection without reduction should be performed in emergency cases.

[Transcutaneous sonography of gastrointestinal tumours]

This pictorial essay shows typical sonographic findings of gastrointestinal tumors from the esophagus to the rectum. The results are often incidental findings in routine examinations and demonstrate that the investigation of the GI-tract should always be part of abdominal sonography. Special tips like the examination of the stomach filled with liquid are required particularly in targeted questions.

Endoscopic ultrasound for differential diagnosis of duodenal lesions

PURPOSE

Duodenal tumors are rare and require a different management from that of esophagogastric neoplasia. The present study retrospectively analyses the endoscopic ultrasound (EUS) features of duodenal tumors of both epithelial and subepithelial origin.

MATERIALS AND METHODS

During a 12 year period, all duodenal tumors with histologic confirmation by surgery or biopsy were collected including endoscopic and endosonographic images. EUS images were analyzed for specific features (echogenicity, wall layer structure and relation, outer margins) to possibly distinguish epithelial (polyps and carcinoma versus lymphoma) and subepithelial (tumor type) tumors.

RESULTS

53/80 cases had histologic confirmation (mean age 53.1 ± 11.4 years, m:f = 33:20), 31 were epithelial (13 adenomas, 12 carcinomas, 6 lymphomas) and 22 subepithelial (11 GISTs, 7 Brunneromas, 1 lipoma, 3 NETs). EUS did not recognize carcinomas in 2/13 adenomas. EUS features suggesting carcinoma were loss of wall layers and irregular margins. 5/6 lymphomas showed inhomogeneous thickening with layers partially recognizable. Tumor type of subepithelial lesions correlated with echogenicity: GIST tumors were mostly (62.5 %) hypocheoic with the 3 malignant cases being characterized by heterogeneous echopattern with irregular outer margins. Of the hyperechoic lesions, lipomas had a homogeneous whitish appearance, while NET and Brunneromas were less hyperechoic. In the latter, the endoscopic aspect was also helpful for differential diagnosis. Accuracy of combined endoscopic/EUS imaging for all duodenal lesions was 84.9 % (45/53). No procedural complications occurred among all patients that received EUS examinations.

CONCLUSION

EUS contributes to the differential diagnosis of epithelial lesions known to be malignant; in subepithelial tumors, tissue confirmation is still required.

Sonographic diagnosis of juvenile polyps in children

The aim of this study was to assess the diagnostic value of ultrasonography for juvenile polyps in children and their sonographic characteristics. A retrospective analysis was performed of the ultrasound findings in 27 children who were diagnosed preoperatively with juvenile polyp within the intestinal tract by ultrasonography and then confirmed by colonoscopy, laparotomy and histopathology. The ultrasonic finding common to all polyps was an isolated intraluminal nodular or massive protrusion, associated with multiple mesh-like fluid areas of different sizes. In 25 children, surrounding pedicle-like low echoes of varying lengths were seen connecting with the polyps to form "mushroom" sign. The color Doppler showed abundant blood flow signals within all polyps and pedicles in a shape of a branch or an umbrella. For seven children with an intussusception, the polyp shadow was detected in the cervical part or interior of the intussusception. Ultrasonography is, thus, considered to be a feasible method for diagnosing intestinal juvenile polyp.

Vanek's polyp or inflammatory fibroid polyp. Case report and review of the literature

BACKGROUND

Intussusception in an adult must make us suspect the presence of a tumor (benign or potentially dangerous) as the most frequent cause. Accurate diagnosis is of great importance in order to provide appropriate treatment and improve patient prognosis.

CLINICAL CASE

We report the case of a 42-year-old male with abdominal pain. We performed a CT and found a small bowel intussusception. Definitive diagnosis according to the surgical specimen was inflammatory fibroid polyp (Vanek's polyp).

CONCLUSIONS

Vanek's polyp is a benign lesion that occurs most frequently in the stomach and secondarily in the small bowel. Generally, it is uncommon, and its etiology is not completely known. Accurate diagnosis is done with immunohistochemistry. Because of the consequences that depend on the size and location of the lesion, it may be considered a malignant lesion. Treatment is resection.

Adult intussusception with perforation and secondary peritonitis. Case report

BACKGROUND

Intussuception is an uncommon condition in adults. It is usually secondary to an organic lesion that may be malignant. The most common clinical presentation is as a partial bowel obstruction that requires surgical management. Preoperative diagnosis remains difficult; therefore, this paper presents a case report and a brief review of adult intussusception.

CLINICAL CASE

We present the case of a 24-year-old female with a 36-h evolution of lower abdominal pain with nausea, vomiting and diarrhea and a previous episode 8 days earlier. Pneumoperitoneum was observed on chest x-ray and surgery was decided upon. Peritonitis due to ileoileal intussusception was found, caused by an inflammatory fibroid polyp with microperforations. Small bowel resection with end-to-end anastomosis was performed and the patient had an uneventful recovery.

CONCLUSIONS

Adult intussusception is an infrequent condition with nonspecific symptoms such as pain, nausea and vomiting. With more frequent use of tomography in patients with abdominal pain, correct diagnosis can be achieved. Treatment requires resection of the involved bowel without attempted reduction.

Intraoperative ultrasound in colorectal surgery

PURPOSE

To assess the accuracy of intraoperative ultrasound (IOUS) as a localizing technique for colorectal resections, and its impact on surgical management.

METHODS

Twenty-five patients (15 men and 10 women; mean age, 74.4 years) with early cancers (p T1), or polyps, not amenable to endoscopic removal were selected. IOUS was used as a sole method of intraoperative localization. Its performance was evaluated through review of preoperative colonoscopy reports, intraoperative findings, histopathology reports, and clinical follow-up.

RESULTS

The lesions were situated in the cecum (n = 5), ascending colon (n = 3), transverse colon (n = 4), descending colon (n = 7), and rectum (n = 6). IOUS technique allowed correct localization in 24 of 25 patients, visualization of the bowel wall, and its penetration by malignant tumors. In rectal lesions, IOUS showed clearly the tumor and its margin, which facilitated performance sphincter-sparing procedure.

CONCLUSION

In patients with small polyps and early cancers of colon and rectum, IOUS may be effectively used as a sole method of intraoperative localization and provide additional information that may alter decision making with regard to surgical technique.

Computer-aided detection of colorectal polyps: can it improve sensitivity of less-experienced readers? Preliminary findings

PURPOSE

To determine whether computer-aided detection (CAD) applied to computed tomographic (CT) colonography can help improve sensitivity of polyp detection by less-experienced radiologist readers, with colonoscopy or consensus used as the reference standard.

MATERIALS AND METHODS

The release of the CT colonographic studies was approved by the individual institutional review boards of each institution. Institutions from the United States were HIPAA compliant. Written informed consent was waived at all institutions. The CT colonographic studies in 30 patients from six institutions were collected; 24 images depicted at least one confirmed polyp 6 mm or larger (39 total polyps) and six depicted no polyps. By using an investigational software package, seven less-experienced readers from two institutions evaluated the CT colonographic images and marked or scored polyps by using a five-point scale before and after CAD. The time needed to interpret the CT colonographic findings without CAD and then to re-evaluate them with CAD was recorded. For each reader, the McNemar test, adjusted for clustered data, was used to compare sensitivities for readers without and with CAD; a Wilcoxon signed-rank test was used to analyze the number of false-positive results per patient.

RESULTS

The average sensitivity of the seven readers for polyp detection was significantly improved with CAD-from 0.810 to 0.908 (P=.0152). The number of false-positive results per patient without and with CAD increased from 0.70 to 0.96 (95% confidence interval for the increase: -0.39, 0.91). The mean total time for the readings was 17 minutes 54 seconds; for interpretation of CT colonographic findings alone, the mean time was 14 minutes 16 seconds; and for review of CAD findings, the mean time was 3 minutes 38 seconds.

CONCLUSION

Results of this feasibility study suggest that CAD for CT colonography significantly improves per-polyp detection for less-experienced readers.

Feasibility study of computed tomography colonography using limited bowel preparation at normal and low-dose levels study

The purpose was to evaluate low-dose CT colonography without cathartic cleansing in terms of image quality, polyp visualization and patient acceptance. Sixty-one patients scheduled for colonoscopy started a low-fiber diet, lactulose and amidotrizoic-acid for fecal tagging 2 days prior to the CT scan (standard dose, 5.8–8.2 mSv). The original raw data of 51 patients were modified and reconstructed at simulated 2.3 and 0.7 mSv levels. Two observers evaluated the standard dose scan regarding image quality and polyps. A third evaluated the presence of polyps at all three mSv levels in a blinded prospective way. All observers were blinded to the reference standard: colonoscopy. At three times patients were given questionnaires relating to their experiences and preference. Image quality was sufficient in all patients, but significantly lower in the cecum, sigmoid and rectum. The two observers correctly identified respectively 10/15 (67%) and 9/15 (60%) polyps ≥10 mm, with 5 and 8 false-positive lesions (standard dose scan). Dose reduction down to 0.7 mSv was not associated with significant changes in diagnostic value (polyps ≥10 mm). Eighty percent of patients preferred CT colonography and 13% preferred colonoscopy (P<0.001). CT colonography without cleansing is preferred to colonoscopy and shows sufficient image quality and moderate sensitivity, without impaired diagnostic value at dose-levels as low as 0.7 mSv.

[Virtual colonoscopy--and then? Relevance of small colorectal polyps]

Colorectal carcinomas are the second most frequent malignant tumors in Germany and originate predominantly from benign polyps (adenoma-carcinoma sequence). Optical colonoscopy is still the gold standard for diagnosis and treatment colorectal polyps. However, CT colonography ("virtual colonoscopy") provides an alternative procedure with similar diagnostic performance but without the possibility of endoscopic polypectomy. Because of the high prevalence (approx. 30 %) of colon polyps in the screening population, virtual colonoscopy is useful for screening only, if a small number of the lesions must be subsequently removed endoscopically. Most of these lesions must represent such a low risk of malignant transformation in order for a control examination to be sufficient. According to a comprehensive review of the literature, the risk potential of small polyps up to a size of 5 - 6 mm can be summarized as follows: Up to 75 % of all polyps found in screening programs belong to the group of "small polyps". Only approx. 1.5-2 % of these contain advanced atypia and less than 0.05 % are carcinomas. Growth speed lies in the range of approx. 0.4 mm in 2 years, but regression can be observed as well. The very few existing growth studies cannot provide reliable information about how quickly these small lesions may turn malignant.

CONCLUSION

The risk potential of small polyps can be considered to be low. Nevertheless, because only poor data is available, a definitive statement cannot be made about the spontaneous course of these lesions. Therefore, further studies examining the natural growth behavior of small polyps are necessary, before control examinations (in intervals of two to three years) rather than immediate endoscopic polypectomy can be recommended as an adequate approach in screening programs.

[New imaging methods for bowel imaging]

Advances in ultra fast high resolution imaging using MRI and CT allow the detailed depiction of the gastrointestinal tract within a breathold. In this article we give an overview on new diagnostic avenues depicting the stomach, small bowel and colon. Additionally recently introduced techniques such as "virtual endoscopy" and "virtual pathology" are explained.

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.

Occult colorectal polyps on CT colonography: implications for surveillance

OBJECTIVE

Our purpose was to determine the prevalence of polyps that are invisible on CT colonography (CTC) in a population previously screened for colorectal neoplasms. Differences in the prevalence of occult polyps in various populations might help explain the discordant reported sensitivities for polyp detection in published reports of CTC.

SUBJECTS AND METHODS

Seventy-five consecutive patients who had been previously screened for polyps underwent same-day colonoscopy and CTC. Many of the patients had personal histories of previous polypectomies and were undergoing surveillance colonoscopy. The scans were interpreted prospectively by an experienced radiologist. Polyps missed prospectively on CTC were analyzed retrospectively by three experienced radiologists and categorized as perception errors (visible in retrospect), technical errors (e.g., obscured by feces or fluid), or occult (invisible).

RESULTS

Thirty polyps 5 mm or larger were found at colonoscopy, 18 of which were missed prospectively on CTC. Of the 18 missed polyps, 12 could not be identified in retrospect, even though they were located in clean, dry, well-distended colonic segments. These were classified as occult. Ten of the 12 occult polyps showed flat morphology on review of colonoscopy video recordings. Of the remaining six missed polyps, two were classified as perception errors, two as technical errors, and two as a combination of technical and perception error.

CONCLUSION

In this population, colonographically occult polyps were common and accounted for more detection failures than perception errors and technical errors combined. The high prevalence of occult polyps may be explained by the fact that previous screening may have led to removal of easy-to-see polyps, creating a study population with a higher percentage of hard-to-see polyps.

Factors influencing the detection of abnormalities in barium enemas performed by junior radiologists

AIM

To study the effects of patient variability, technical examination and radiologist-related factors on the radiological diagnosis of barium enema studies.

MATERIALS AND METHODS

Twenty-one residents (12 women and nine men, age 26-39 years) performed 523 double contrast barium enemas in a university hospital. Diagnoses were classified as normal (n = 203), diverticulosis (n = 224), polyps (n = 25), tumours (n = 20) and other (n = 46); five were missing diagnoses. Binary and multinomial logistic regressions were used to study the relations between the independent variables and the radiological diagnoses.

RESULTS

Patients' advanced age (p < 0.001) and obesity (p = 0.041) were significant positive predictors for the detection of any abnormalities, as was good or moderate study quality (p = 0.023). However, patients' obesity was negatively associated with the detection of polyps (p = 0.037). Pathological findings were associated with higher dose-area products (p = 0.047) than normal cases. Residents' advanced age (p = 0.029) and experience (a high number of previously performed barium enemas; p = 0.009) were negative predictors for the detection of pathological findings.

CONCLUSION

Patients' obesity hampers polyp detection, possibly due to decreased image quality or coexisting diverticula. The most experienced and oldest residents missed positive findings, possibly due to insufficient consultation of supervising radiologists, emphasizing the need for their optimal cooperation.

[Potential use of current X-ray contrast studies in the detection of small intestinal polyps in the Peutz-Jeghers syndrome]

The paper presents the results of studies of the small bowel in 9 patients with the Peutz-Jeghers syndrome. The indications for the study were the clinical picture of gastrointestinal bleeding and the symptoms of ileus. The author's intubation enterographic procedure using the better composition of barium suspension and an infusion system for administering contract substances into the small bowel, an improved fractional contrasting procedure, and a procedure employing the agent Entero-view was applied for contrasting the small bowel. All the procedures revealed an obvious picture of polyposis of the small bowel. The minimum size of detected tumors was 0.3-0.4 cm in diameter. Emphasis is laid on the quantitative image of neoplasms with Entero-view and the therapeutic effect of intubation enterography in evolving small-small intestinal intussusception.

[Jejunal intussusception of an inflammatory fibrinoid polyp]

Intussusception of parts of the intestine in adulthood is a rare cause of ileus. In more than 90 % of all cases there is a malignant or benign tumour situated in the parts of the intestine affected. We describe a 55-year-old patient who was admitted to hospital as an emergency because of signs of a proximal intestinal obstruction. Abdominal sonography suggested the diagnosis of a typical jejunal intussusception, whereas gastroscopy yielded no pathological findings. Laparotomy was performed immediately and confirmed the diagnosis. After intraoperative de-invagination 9 cm of jejunum and an inflammatory fibrinoid polyp of 3 cm were resected.

Detection of colorectal polyps by multislice CT colonography with ultra-low-dose technique: comparison with high-resolution videocolonoscopy

BACKGROUND

This prospective study compared multislice CT colonography with ultra-low-dose technique to high-resolution videocolonoscopy as the standard for detection of colorectal cancer and polyps.

METHODS

After standard bowel preparation, 115 patients underwent multislice CT colonography with an ultra-low-dose multislice CT colonography protocol immediately before videocolonoscopy. After noise reduction by using a mathematical algorithm, ultra-low-dose multislice CT colonographic images were analyzed in blinded fashion, and the results were compared with the results of high-resolution videocolonoscopy.

RESULTS

A total of 150 lesions were detected by high-resolution videocolonoscopy in 115 patients. For ultra-low-dose multislice CT colonography, sensitivities for detection of polyps less than 5 mm in size, 5 to 10 mm, and greater than 10 mm in diameter were 76%, 91%, and 100%, respectively. Although the sensitivity for detection of flat lesions was only 50%, the sensitivity and the specificity for detection of polyps 5 mm or greater in size were 94% and 84%, respectively. For adenomatous lesions greater than 5 mm in size, sensitivity was 94% and specificity was 92%. The overall specificity was 79%. The calculated effective radiation dose ranged between 0.75 and 1.25 mSv.

CONCLUSIONS

Compared with high-resolution videocolonoscopy, ultra-low-dose multislice CT colonography has excellent sensitivity and specificity for detection of colorectal lesions 5 mm or greater in size, and the radiation exposure is relatively low. However, before this technique can be generally recommended for colorectal screening, further improvement in the detection of flat and extremely small lesions must be achieved.

Use of multidetector-row CT colonography for detection of colorectal neoplasia in patients referred via the Department of Health "2-Week-wait" initiative

AIM

Patients referred under the Department of Health 2-week wait initiative with symptoms of colorectal cancer frequently undergo whole-colon examination. We investigated the use of computed tomography (CT) colonography as an alternative to colonoscopy in this scenario.

MATERIALS AND METHODS

Fifty-four consecutive patients, referred via the 2-week wait initiative and scheduled for colonoscopy, consented to undergo multidetector CT colonography immediately before endoscopy. The site and morphology of any polyp or cancer detected by CT was noted and comparison made with subsequent colonoscopy.

RESULTS

Colonoscopy detected polyps or cancer in 29 patients (53.7%). CT colonography prospectively detected 18 of 41 (44%) polyps of 1-5 mm, three of four (75%) polyps of 6-9 mm, four of four (100%) polyps 10 mm or larger, and five of six (83%) cancers. The missed cancer occurred early in the series and was a perceptive error. The overall sensitivity, specificity, positive predictive value and negative predictive value of CT colonography for cancer and polyps 10 mm or greater on a per patient basis were 90, 100, 100 and 98%, respectively. CT detected one renal cancer and one colonic cancer, initially missed due to incomplete colonoscopy.

CONCLUSION

CT colonography is a robust technique for investigation of symptomatic patients. The learning curve must be overcome for optimal performance.

CT colonography: Protocol optimization with multi-detector row CT--study in an anthropomorphic colon phantom

PURPOSE

To determine optimal detector collimation, section thickness, and tube current for multi-detector row computed tomography (CT) colonography.

MATERIALS AND METHODS

An anthropomorphic colon phantom with simulated polyps of varying size (2, 6, 8, 10, and 12 mm) was examined by using multi-detector row CT with varying combinations of detector collimation (4 x 1.0 mm and 4 x 2.5 mm), dose per section (10, 20, 40, 60, 80, 100, and 140 mAs), and section thickness/reconstruction interval (1.25/0.6, 2.0/1.0, 3.0/1.0, and 5.0/2.0 mm). Polyp depiction, longitudinal polyp distortion, and presence of rippling artifacts were assessed on reformatted three-dimensional endoluminal images by three reviewers.

RESULTS

Longitudinal distortion and rippling artifacts increased with increasing section thickness and use of broader detector collimation. Polyps 8 mm or larger were depicted with any combination of section thickness, detector collimation, and tube current. Depiction of polyps 6 mm or smaller depended on the detector collimation/reconstructed section thickness and was rated optimal for the 4 x 1.0-mm detector collimation with a section thickness of 1.25 mm. This was also observed for low-dose protocols. Polyps 6 mm or smaller that were not detected with 3-mm section thickness and 4 x 2.5-mm detector collimation were detected with 1.25-mm section thickness and 10 mAs.

CONCLUSION

A narrow detector collimation with thin-section imaging (4 x 1.0-mm detector collimation, 1.25-mm section thickness) is a prerequisite for low-dose (10-mAs) multi-detector row CT colonography.

Multiple somatostatin- and gastrin-containing carcinoids of the duodenum: report of a case treated by pancreas-sparing duodenectomy

A 59-year-old Japanese man was incidentally discovered to have multiple polyps in the duodenum by endoscopy during a health checkup. Laboratory studies showed an elevated level of serum somatostatin. The mutation of the MEN1 gene was not observed. An endoscopic examination revealed multiple polypoid lesions in the bulbus and in the second portion of the duodenum. An upper gastrointestinal series also showed the multiple polypoid lesions in the bulbus, and in the descending and horizontal portions of the duodenum. The biopsy specimen showed small monotonous cells with round nuclei compatible with a carcinoid tumor. The tumor cells were positive for somatostatin and gastrin. Based on the clinical diagnosis of multiple carcinoids of the duodenum, a pancreas-sparing duodenectomy was performed. Macroscopic findings showed about 30 polypoid lesions throughout the duodenum. The biggest one was not over 10 mm in size. A histological examination revealed well-demarcated carcinoid tumors located in the submucosal layer. Immunohistochemically, the tumors were diffusely positive for somatostatin and were scatteringly positive for gastrin. The patient's postoperative course of treatment was uneventful and his postoperative serum somatostatin went down to normal range. The pancreas-sparing duodenectomy is a safe and effective treatment in patients with multiple carcinoids of the duodenum, that are smaller than 1 cm in size.