Neoplasia Diagnosis After Multi-target Stool DNA Is Enhanced Among Lowest Baseline Detectors

BACKGROUND AND AIMS

Variation in colorectal neoplasia detection limits the effectiveness of screening colonoscopy. By evaluating neoplasia detection rates of individual colonoscopists, we aimed to quantify the effects of pre-procedural knowledge of a positive (+) multi-target stool DNA (mt-sDNA) on colonoscopy quality metrics.

METHODS

We retrospectively identified physicians who performed a high volume of + mt-sDNA colonoscopies; colorectal neoplasia at post-mt-sDNA colonoscopy was recorded. These colonoscopists were stratified into quartiles based on baseline adenoma detection rates. Baseline colonoscopy adenoma detection rates and sessile serrated lesion detection rates were compared to post-mt-sDNA colonoscopy neoplasia diagnosis rates among each quartile. Withdrawal times were measured from negative exams.

RESULTS

During the study period (2014-17) the highest quartile of physicians by volume of post-mt-sDNA colonoscopies were evaluated. Among thirty-five gastroenterologists, their median screening colonoscopy adenoma detection rate was 32% (IQR, 28-39%) and serrated lesion detection rate was 13% (8-15%). After + mt-sDNA, adenoma diagnosis increased to 47% (36-56%) and serrated lesion diagnosis increased to 31% (17-42%) (both p < 0.0001). Median withdrawal time increased from 10 (7-13) to 12 (10-17) minutes (p < 0.0001) and was proportionate across quartiles. After + mt-sDNA, lower baseline detectors had disproportionately higher rates of adenoma diagnosis in female versus male patients (p = 0.048) and higher serrated neoplasia diagnosis rates among all patients (p = 0.0092).

CONCLUSIONS

Knowledge of + mt-sDNA enriches neoplasia diagnosis compared to average risk screening exams. Adenomatous and serrated lesion diagnosis was magnified among those with lower adenoma detection rates. Awareness of the mt-sDNA result may increase physician attention during colonoscopy. Pre-procedure knowledge of a positive mt-sDNA test improves neoplasia diagnosis rates among colonoscopists with lower baseline adenoma detection rates, independent of withdrawal time.

The Role of Clips in Preventing Delayed Bleeding After Colorectal Polyp Resection: An Individual Patient Data Meta-Analysis

BACKGROUND & AIMS

Nonpedunculated colorectal polyps are normally endoscopically removed to prevent neoplastic progression. Delayed bleeding is the most common major adverse event. Clipping the resection defect has been suggested to reduce delayed bleedings. Our aim was to determine if prophylactic clipping reduces delayed bleedings and to analyze the contribution of polyp characteristics, extent of defect closure, and antithrombotic use.

METHODS

An individual patient data meta-analysis was performed. Studies on prophylactic clipping in nonpedunculated colorectal polyps were selected from PubMed, Embase, Web of Science, and Cochrane database (last selection, April 2020). Authors were invited to share original study data. The primary outcome was delayed bleeding ≤30 days. Multivariable mixed models were used to determine the efficacy of prophylactic clipping in various subgroups adjusted for confounders.

RESULTS

Data of 5380 patients with 8948 resected polyps were included from 3 randomized controlled trials, 2 prospective, and 8 retrospective studies. Prophylactic clipping reduced delayed bleeding in proximal polyps ≥20 mm (odds ratio [OR], 0.62; 95% confidence interval [CI], 0.44-0.88; number needed to treat = 32), especially with antithrombotics (OR, 0.59; 95% CI, 0.35-0.99; number needed to treat = 23; subgroup of anticoagulants/double platelet inhibitors: n = 226; OR, 0.40; 95% CI, 0.16-1.01; number needed to treat = 12). Prophylactic clipping did not benefit distal polyps ≥20 mm with antithrombotics (OR, 1.41; 95% CI, 0.79-2.52).

CONCLUSIONS

Prophylactic clipping reduces delayed bleeding after resection of nonpedunculated, proximal colorectal polyps ≥20 mm, especially in patients using antithrombotics. No benefit was found for distal polyps. Based on this study, patients can be identified who may benefit from prophylactic clipping. (PROSPERO registration number CRD42020104317.).

Establishing key research questions for the implementation of artificial intelligence in colonoscopy: a modified Delphi method

BACKGROUND : Artificial intelligence (AI) research in colonoscopy is progressing rapidly but widespread clinical implementation is not yet a reality. We aimed to identify the top implementation research priorities. METHODS : An established modified Delphi approach for research priority setting was used. Fifteen international experts, including endoscopists and translational computer scientists/engineers, from nine countries participated in an online survey over 9 months. Questions related to AI implementation in colonoscopy were generated as a long-list in the first round, and then scored in two subsequent rounds to identify the top 10 research questions. RESULTS : The top 10 ranked questions were categorized into five themes. Theme 1: clinical trial design/end points (4 questions), related to optimum trial designs for polyp detection and characterization, determining the optimal end points for evaluation of AI, and demonstrating impact on interval cancer rates. Theme 2: technological developments (3 questions), including improving detection of more challenging and advanced lesions, reduction of false-positive rates, and minimizing latency. Theme 3: clinical adoption/integration (1 question), concerning the effective combination of detection and characterization into one workflow. Theme 4: data access/annotation (1 question), concerning more efficient or automated data annotation methods to reduce the burden on human experts. Theme 5: regulatory approval (1 question), related to making regulatory approval processes more efficient. CONCLUSIONS : This is the first reported international research priority setting exercise for AI in colonoscopy. The study findings should be used as a framework to guide future research with key stakeholders to accelerate the clinical implementation of AI in endoscopy.

Active, continual fine tuning of convolutional neural networks for reducing annotation efforts

The splendid success of convolutional neural networks (CNNs) in computer vision is largely attributable to the availability of massive annotated datasets, such as ImageNet and Places. However, in medical imaging, it is challenging to create such large annotated datasets, as annotating medical images is not only tedious, laborious, and time consuming, but it also demands costly, specialty-oriented skills, which are not easily accessible. To dramatically reduce annotation cost, this paper presents a novel method to naturally integrate active learning and transfer learning (fine-tuning) into a single framework, which starts directly with a pre-trained CNN to seek "worthy" samples for annotation and gradually enhances the (fine-tuned) CNN via continual fine-tuning. We have evaluated our method using three distinct medical imaging applications, demonstrating that it can reduce annotation efforts by at least half compared with random selection.

Detection of Postcolonoscopy Colorectal Neoplasia by Multi-target Stool DNA

INTRODUCTION

Significant variability between colonoscopy operators contributes to postcolonoscopy colorectal cancers (CRCs). We aimed to estimate postcolonoscopy colorectal neoplasia (CRN) detection by multi-target stool DNA (mt-sDNA), which has not previously been studied for this purpose.

METHODS

In a retrospective cohort of patients with +mt-sDNA and completed follow-up colonoscopy, positive predictive value (PPV) for endpoints of any CRN, advanced adenoma, right-sided neoplasia, sessile serrated polyps (SSP), and CRC were stratified by the time since previous colonoscopy (0-9, 10, and ≥11 years). mt-sDNA PPV at ≤9 years from previous average-risk screening colonoscopy was used to estimate CRN missed at previous screening colonoscopy.

RESULTS

Among the 850 studied patients with +mt-sDNA after a previous negative screening colonoscopy, any CRN was found in 535 (PPV 63%). Among 107 average-risk patients having +mt-sDNA ≤9 years after last negative colonoscopy, any CRN was found in 67 (PPV 63%), advanced neoplasia in 16 (PPV 15%), right-sided CRN in 48 (PPV 46%), and SSP in 20 (PPV 19%). These rates were similar to those in 47 additional average risk persons with previous incomplete colonoscopy and in an additional 68 persons at increased CRC risk. One CRC (stage I) was found in an average risk patient who was mt-sDNA positive 6 years after negative screening colonoscopy.

DISCUSSION

The high PPV of mt-sDNA 0-9 years after a negative screening colonoscopy suggests that lesions were likely missed on previous examination or may have arisen de novo. mt-sDNA as an interval test after negative screening colonoscopy warrants further study.

ASGE guideline on the role of endoscopy in the management of benign and malignant gastroduodenal obstruction

This American Society for Gastrointestinal Endoscopy guideline provides evidence-based recommendations for the endoscopic management of gastric outlet obstruction (GOO). We applied the Grading of Recommendations, Assessment, Development and Evaluation methodology to address key clinical questions. These include the comparison of (1) surgical gastrojejunostomy to the placement of self-expandable metallic stents (SEMS) for malignant GOO, (2) covered versus uncovered SEMS for malignant GOO, and (3) endoscopic and surgical interventions for the management of benign GOO. Recommendations provided in this document were founded on the certainty of the evidence, balance of benefits and harms, considerations of patient and caregiver preferences, resource utilization, and cost-effectiveness.

American Society for Gastrointestinal Endoscopy guideline on the role of endoscopy in familial adenomatous polyposis syndromes

Familial adenomatous polyposis (FAP) syndrome is a complex entity, which includes FAP, attenuated FAP, and MUTYH-associated polyposis. These patients are at significant risk for colorectal cancer and carry additional risks for extracolonic malignancies. In this guideline, we reviewed the most recent literature to formulate recommendations on the role of endoscopy in this patient population. Relevant clinical questions were how to identify high-risk individuals warranting genetic testing, when to start screening examinations, what are appropriate surveillance intervals, how to identify endoscopically high-risk features, and what is the role of chemoprevention. A systematic literature search from 2005 to 2018 was performed, in addition to the inclusion of seminal historical studies. Most studies were from worldwide registries, which have compiled years of data regarding the natural history and cancer risks in this cohort. Given that most studies were retrospective, recommendations were based on epidemiologic data and expert opinion. Management of colorectal polyps in FAP has not changed much in recent years, as colectomy in FAP is the standard of care. What is new, however, is the developing body of literature on the role of endoscopy in managing upper GI and small-bowel polyposis, as patients are living longer and improved endoscopic technologies have emerged.

Prospective multicenter study to evaluate capsule endoscopy competency using a validated assessment tool

BACKGROUND AND AIMS

Capsule endoscopy (CE) is an established, noninvasive modality for examining the small bowel. Minimum training requirements are based primarily on guidelines and expert opinion. A validated tool to assess the competence of CE is lacking. In this prospective, multicenter study, we determined the minimum number of CE procedures required to achieve competence during gastroenterology fellowship; validated a capsule competency test (CapCT); and evaluated any correlation between CE competence and endoscopy experience.

METHODS

We included second- and third-year gastroenterology fellows from 3 institutions between 2013 and 2018 in a structured CE training program with supervised CE interpretation. Fellows completed the CapCT with a maximal score of 100. For comparison, expert faculty completed the same CapCT. Trainee competence was defined as a score ≥90% compared with the mean expert score. Fellows were tested after 15, 25, and 35 supervised CE interpretations. CapCT was validated using expert consensus and item analysis. Data were collected on the number of previous endoscopies.

RESULTS

A total of 68 trainees completed 102 CapCTs. Fourteen CE experts completed the CapCT with a mean score of 94. Mean scores for fellows after 15, 25, and 35 cases were 83, 86, and 87, respectively. Fellows with at least 25 interpretations achieved a mean score ≥84 in all 3 institutions. CapCT item analysis showed high interobserver agreement among expert faculty (k = 0.85). There was no correlation between the scores and the number of endoscopies performed.

CONCLUSION

After a structured CE training program, gastroenterology fellows should complete a minimum of 25 supervised CE interpretations before assessing competence using the validated CapCT, regardless of endoscopy experience.

ASGE guideline on minimum staffing requirements for the performance of GI endoscopy

Efforts to increase patient safety and satisfaction, a critical concern for health providers, require periodic evaluation of all factors involved in the provision of GI endoscopy services. We aimed to develop guidelines on minimum staffing requirements and scope of practice of available staff for the safe and efficient performance of GI endoscopy. The recommendations in this guideline were based on a systematic review of published literature, results from a nationwide survey of endoscopy directors, along with the expert guidance of the American Society for Gastrointestinal Endoscopy (ASGE) Standards of Practice Committee members, ASGE Practice Operation Committee members, and the ASGE Governing Board.

ASGE guideline on the management of achalasia

Achalasia is a primary esophageal motor disorder of unknown etiology characterized by degeneration of the myenteric plexus, which results in impaired relaxation of the esophagogastric junction (EGJ), along with the loss of organized peristalsis in the esophageal body. The criterion standard for diagnosing achalasia is high-resolution esophageal manometry showing incomplete relaxation of the EGJ coupled with the absence of organized peristalsis. Three achalasia subtypes have been defined based on high-resolution manometry findings in the esophageal body. Treatment of patients with achalasia has evolved in recent years with the introduction of peroral endoscopic myotomy. Other treatment options include botulinum toxin injection, pneumatic dilation, and Heller myotomy. This American Society for Gastrointestinal Endoscopy Standards of Practice Guideline provides evidence-based recommendations for the treatment of achalasia, based on an updated assessment of the individual and comparative effectiveness, adverse effects, and cost of the 4 aforementioned achalasia therapies.

ASGE review of adverse events in colonoscopy

Colonoscopy is the most commonly performed endoscopic procedure and overall is considered a low-risk procedure. However, adverse events (AEs) related to this routinely performed procedure for screening, diagnostic, or therapeutic purposes are an important clinical consideration. The purpose of this document from the American Society for Gastrointestinal Endoscopy's Standards of Practice Committee is to provide an update on estimates of AEs related to colonoscopy in an evidence-based fashion. A systematic review and meta-analysis of population-based studies was conducted for the 3 most common and important serious AEs (bleeding, perforation, and mortality). In addition, this document includes an updated systematic review and meta-analysis of serious AEs (bleeding and perforation) related to EMR and endoscopic submucosal dissection for large colon polyps. Finally, a narrative review of other colonoscopy-related serious AEs and those related to specific colonic interventions is included.

ASGE guideline on the role of endoscopy in the evaluation and management of choledocholithiasis

Each year choledocholithiasis results in biliary obstruction, cholangitis, and pancreatitis in a significant number of patients. The primary treatment, ERCP, is minimally invasive but associated with adverse events in 6% to 15%. This American Society for Gastrointestinal Endoscopy (ASGE) Standard of Practice (SOP) Guideline provides evidence-based recommendations for the endoscopic evaluation and treatment of choledocholithiasis. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework was used to rigorously review and synthesize the contemporary literature regarding the following topics: EUS versus MRCP for diagnosis, the role of early ERCP in gallstone pancreatitis, endoscopic papillary dilation after sphincterotomy versus sphincterotomy alone for large bile duct stones, and impact of ERCP-guided intraductal therapy for large and difficult choledocholithiasis. Comprehensive systematic reviews were also performed to assess the following: same-admission cholecystectomy for gallstone pancreatitis, clinical predictors of choledocholithiasis, optimal timing of ERCP vis-à-vis cholecystectomy, management of Mirizzi syndrome and hepatolithiasis, and biliary stent therapy for choledocholithiasis. Core clinical questions were derived using an iterative process by the ASGE SOP Committee. This body developed all recommendations founded on the certainty of the evidence, balance of risks and harms, consideration of stakeholder preferences, resource utilization, and cost-effectiveness.

Severe Diarrhea in the Setting of Immune Checkpoint Inhibitors

Immune checkpoint inhibitors (ICPIs) are novel therapeutic agents targeting a variety of cancers by enhanced T cell activation. Immune-related adverse events (irAEs) commonly occur with ICPI use and can affect multiple organ systems including the gastrointestinal tract. Due to irAEs, the use of ICPIs is limited in autoimmune diseases. We present a case of microscopic colitis diagnosed after the initiation of nivolumab and a case of ipilimumab colitis and Clostridium difficile in the setting of Crohn's colitis.

Comparison of Small Versus Large Volume Split Dose Preparation for Colonoscopy: A Study of Colonoscopy Outcomes

BACKGROUND

Split dose bowel preparations (SDP) have superior outcomes for colonoscopy as compared to evening before regimens. However, the association of the actual volume of the SDP to colonoscopy outcome measures has not been well studied.

AIMS

Compare adenoma detection rate (ADR), sessile serrated polyp detection rate (SDR), mean bowel cleanse score, and predictors of inadequate exams between small volume SDP and large volume SDP.

METHODS

We have conducted a retrospective study in patients undergoing colonoscopy with small volume SDP versus large volume SDP between July 2014 and December 2014. Basic demographics (age, gender and BMI) along with clinical co-morbidities were recorded. Quality of the bowel preparation, ADR and SDR was compared between these groups. Univariate and multivariable logistic regressions were used to assess the determinants of inadequate exams in each group.

RESULTS

1573 patients with split dose preparation were included in this retrospective study. 58.4% (920/1573) patients took small volume SDP. There was no difference in ADR (37.9 vs. 38.8%, p = 0.2); however, SDR was higher for small volume SDP compared to large volume SDP (11.9 vs. 7.9% p = 0.005). There was no difference in the rate of inadequate exams between the two groups (p = 0.7). A history of diabetes and constipation was associated with inadequate exams only in the small volume SDP.

CONCLUSIONS

SDR was higher in small volume SDP. There was no difference in rate of inadequate exams between the two groups. A history of diabetes and constipation was associated with inadequate exams only in patients with the small volume SDP.

Impact of feedback on adenoma detection rates: Outcomes of quality improvement program

BACKGROUND AND AIM

Feedback has been shown to improve performance in colonoscopy including adenoma detection rate (ADR). The frequency at which feedback should be given is unknown. As part of a quality improvement program, we sought to measure the outcome of providing quarterly and monthly feedback on colonoscopy quality measures.

METHODS

All screening colonoscopies performed at endoscopy unit at Mayo Clinic Arizona by gastroenterologists between October 2010 and December 2012 were reviewed. Quality indicators, including ADR, were extracted for each individual endoscopist, and feedback was provided. The study period was divided into four distinct groups: pre-intervention that served as baseline, quarterly feedback, monthly feedback, and post-intervention. Based on ADR, endoscopists were grouped into "low detectors" (≤ 25%), "average detectors" (26-35%), and "high detectors" (> 35%).

RESULTS

A total of 3420 screening colonoscopies were performed during the study period (555 patients during pre-intervention, 1209 patients during quarterly feedback, 599 during monthly feedback, and 1057 during the post-intervention period) by 16 gastroenterologists. The overall ADR for the group improved from 30.5% to 37.7% (P = 0.003). Compared with the pre-interventional period, all quality indicators measured significantly improved during the monthly feedback and post-intervention periods but not in the quarterly feedback period.

CONCLUSIONS

In our quality improvement program, monthly feedback significantly improved colonoscopy quality measures, including ADR, while quarterly feedback did not. The impact of the intervention was most prominent in the "low detectors" group. Results were durable up to 6 months following the intervention.

Is the level of cleanliness using segmental Boston bowel preparation scale associated with a higher adenoma detection rate?

Background

The impact of Boston bowel preparation scale (BBPS) scores on the adenoma detection rate (ADR) in each segment has not been adequately addressed. The aim of this study was to determine the association between segmental or overall ADR and serrated polyp detection rate (SDR) with segmental and total BBPS scores.

Methods

All outpatient screening colonoscopies with documented BBPS scores were retrospectively reviewed at a tertiary institution from January to December 2013. Chi-square tests and logistic regression were used to analyze the detection rates of adenomas and serrated polyps with bowel prep scores. Odds ratios were calculated using logistic regression that controlled for withdrawal time, age, body mass index, diabetes status and sex.

Results

We analyzed 1991 colonoscopies. The overall ADR was 37.5% (95% confidence interval [CI], 35.3-39.6). There was a significant difference in the overall ADR, and in SDR across all bowel category groups, with total BBPS scores of 8 and 9 having lower detection rates than scores of 5, 6 and 7. As the quality of bowel preparation increased, there was a statistical decrease in the ADR (odds ratio [OR] 0.79 [CI 0.66-0.94], P=0.04) of the right colon, while in the left colon, there was a statistical decrease in SDR (OR 0.78, [CI 0.65-0.92] P=0.019).

Conclusion

Segmental ADR and SDR both decreased as prep scores increased, decreasing notably in patients with excellent prep scores of 8 and 9. A possible explanation for this unexpected discrepancy may be related to longer and better visualization of the mucosa when cleansing and suctioning is necessary.

Approach to Incomplete Colonoscopy: New Techniques and Technologies

Colonoscopy is the most widely used screening modality for the detection and removal of colon polyps and for the prevention of colorectal cancer. To identify all colon lesions and reduce the risk of colorectal cancer, it is important to perform a complete colonoscopy. The success of screening colonoscopy depends upon several parameters, including bowel preparation and adenoma detection rate. Incomplete colonoscopy rates vary from 4% to 25% and are associated with higher rates of interval proximal colon cancer. This article reviews the potential causes of and preventive measures for incomplete colonoscopy, as well as techniques and technologies that may improve the rate of complete colonoscopy.

Comparative Validation of Polyp Detection Methods in Video Colonoscopy: Results From the MICCAI 2015 Endoscopic Vision Challenge

Colonoscopy is the gold standard for colon cancer screening though some polyps are still missed, thus preventing early disease detection and treatment. Several computational systems have been proposed to assist polyp detection during colonoscopy but so far without consistent evaluation. The lack of publicly available annotated databases has made it difficult to compare methods and to assess if they achieve performance levels acceptable for clinical use. The Automatic Polyp Detection sub-challenge, conducted as part of the Endoscopic Vision Challenge (http://endovis.grand-challenge.org) at the international conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2015, was an effort to address this need. In this paper, we report the results of this comparative evaluation of polyp detection methods, as well as describe additional experiments to further explore differences between methods. We define performance metrics and provide evaluation databases that allow comparison of multiple methodologies. Results show that convolutional neural networks are the state of the art. Nevertheless, it is also demonstrated that combining different methodologies can lead to an improved overall performance.

Impact and outcomes of research sponsored by the American Society for Gastrointestinal Endoscopy

BACKGROUND AND AIMS

Since 1985, the American Society for Gastrointestinal Endoscopy (ASGE) has awarded grants for endoscopic-related research. The goals of this study were to examine trends in ASGE grant funding and to assess productivity of previous recipients of the ASGE grant awards.

METHODS

This was a retrospective cohort analysis of all research grants awarded by the ASGE through 2009. Measures of academic productivity and self-assessment of the ASGE awards' impact on the recipients' careers were defined by using publicly available resources (eg, National Library of Medicine-PubMed) and administration of an electronic survey to award recipients.

RESULTS

The ASGE awarded 304 grants totaling $12.5 million to 214 unique awardees. Funding increased 7.5-fold between 1985 and 1989 (mean $102,000/year) and between 2005 and 2009 (mean $771,000/year). The majority of awardees were men (83%), were at or below the level of assistant professor (82%), with a median of 3 years of postfellowship experience at the time of the award, and derived from a broad spectrum of institutions as measured by National Institutes of Health funding rank (median 26, interquartile range [IQR] 12-64). Nineteen percent had a master's degree in a research-related field. Awardees' median publications per year increased from 3.5 (IQR 1.2-9.0) before funding to 5.7 (IQR 1.8-9.5) since funding; P = .04, and median h-index scores increased from 3 (IQR 1-8) to 17 (IQR 8-26); P < .001. Multivariate analysis found that the presence of a second advanced degree (eg, masters or doctorate) was independently predictive of high productivity (odds ratio [OR] 2.92; 95% confidence interval [CI], 1.09-7.81). Among 212 unique grant recipients, 82 (40%) completed the online survey. Of the respondents, median peer-reviewed publications per year increased from 3.4 (IQR 1.9-5.5) to 4.5 (IQR 2.0-9.5); P = .17. Ninety-one percent reported that the ASGE grant had a positive or very positive impact on their careers, and 85% of respondents are currently practicing in an academic environment. Most of the grants resulted in at least 1 peer-reviewed publication (67% per Internet-based search and 81% per survey).

CONCLUSIONS

The ASGE research program has grown considerably since 1985, with the majority of grants resulting in at least 1 grant-related publication. Overall academic productivity increased after the award, and the majority of awardees report a positive or very positive impact of the award on their careers. Medical professional societies are an important sponsor of clinical research.

Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?

Training a deep convolutional neural network (CNN) from scratch is difficult because it requires a large amount of labeled training data and a great deal of expertise to ensure proper convergence. A promising alternative is to fine-tune a CNN that has been pre-trained using, for instance, a large set of labeled natural images. However, the substantial differences between natural and medical images may advise against such knowledge transfer. In this paper, we seek to answer the following central question in the context of medical image analysis: Can the use of pre-trained deep CNNs with sufficient fine-tuning eliminate the need for training a deep CNN from scratch? To address this question, we considered four distinct medical imaging applications in three specialties (radiology, cardiology, and gastroenterology) involving classification, detection, and segmentation from three different imaging modalities, and investigated how the performance of deep CNNs trained from scratch compared with the pre-trained CNNs fine-tuned in a layer-wise manner. Our experiments consistently demonstrated that 1) the use of a pre-trained CNN with adequate fine-tuning outperformed or, in the worst case, performed as well as a CNN trained from scratch; 2) fine-tuned CNNs were more robust to the size of training sets than CNNs trained from scratch; 3) neither shallow tuning nor deep tuning was the optimal choice for a particular application; and 4) our layer-wise fine-tuning scheme could offer a practical way to reach the best performance for the application at hand based on the amount of available data.

Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information

This paper presents the culmination of our research in designing a system for computer-aided detection (CAD) of polyps in colonoscopy videos. Our system is based on a hybrid context-shape approach, which utilizes context information to remove non-polyp structures and shape information to reliably localize polyps. Specifically, given a colonoscopy image, we first obtain a crude edge map. Second, we remove non-polyp edges from the edge map using our unique feature extraction and edge classification scheme. Third, we localize polyp candidates with probabilistic confidence scores in the refined edge maps using our novel voting scheme. The suggested CAD system has been tested using two public polyp databases, CVC-ColonDB, containing 300 colonoscopy images with a total of 300 polyp instances from 15 unique polyps, and ASU-Mayo database, which is our collection of colonoscopy videos containing 19,400 frames and a total of 5,200 polyp instances from 10 unique polyps. We have evaluated our system using free-response receiver operating characteristic (FROC) analysis. At 0.1 false positives per frame, our system achieves a sensitivity of 88.0% for CVC-ColonDB and a sensitivity of 48% for the ASU-Mayo database. In addition, we have evaluated our system using a new detection latency analysis where latency is defined as the time from the first appearance of a polyp in the colonoscopy video to the time of its first detection by our system. At 0.05 false positives per frame, our system yields a polyp detection latency of 0.3 seconds.

Using Hybrid Change Strategies to Improve the Patient Experience in Outpatient Specialty Care

The emerging changes in healthcare impose significant burdens on integrated outpatient specialty services with respect to setting patient expectations, handling outside medical records; and coordinating specialty appointments scheduling. Moreover, because of the evolution of the electronic health record and its widespread use, it is critical that patient and physician interaction is maintained and clerical tasks are minimized. In the context of increased government regulation, declining reimbursement, and the rise of new payment models, outpatient practices need to be reimagined so that they are more efficient for the patient and the provider. The redesign of integrated outpatient specialty services can be accomplished only through teamwork, innovation, and efficient use of technology. To address these challenges, the Department of Medicine at Mayo Clinic in Scottsdale, Arizona, implemented an ideal practice design initiative that leveraged a hybrid set of change strategies. The change strategy, which was initiated after examination of current practices and design options, engaged key stakeholders and patients. A number of enablers and barriers to adoption were identified as a result of the implementation experience.

Nonneoplastic polypectomy during screening colonoscopy: the impact on polyp detection rate, adenoma detection rate, and overall cost

BACKGROUND

The frequency of nonneoplastic polypectomy (NNP) and its impact on the polyp detection rate (PDR) is unknown. The correlation between NNP and adenoma detection rate (ADR) and its impact on the cost of colonoscopy has not been investigated.

OBJECTIVE

To determine the rate of NNP in screening colonoscopy, the impact of NNP on the PDR, and the correlation of NNP with ADR. The increased cost of NNP during screening colonoscopy also was calculated.

DESIGN

We reviewed all screening colonoscopies. PDR and ADR were calculated. We then calculated a nonneoplastic polyp detection rate (patients with ≥1 nonneoplastic polyp).

SETTING

Tertiary-care referral center.

PATIENTS

Patients who underwent screening colonoscopies from 2010 to 2011.

INTERVENTIONS

Colonoscopy.

MAIN OUTCOME MEASUREMENTS

ADR, PDR, NNP rate.

RESULTS

A total of 1797 colonoscopies were reviewed. Mean (±standard deviation) PDR was 47.7%±12.0%, and mean ADR was 27.3%±6.9%. The overall NNP rate was 10.4%±7.1%, with a range of 2.4% to 28.4%. Among all polypectomies (n=2061), 276 were for nonneoplastic polyps (13.4%). Endoscopists with a higher rate of nonneoplastic polyp detection were more likely to detect an adenoma (odds ratio 1.58; 95% confidence interval, 1.1-1.2). With one outlier excluded, there was a strong correlation between ADR and NNP (r=0.825; P<.001). The increased cost of removal of nonneoplastic polyps was $32,963.

LIMITATIONS

Retrospective study.

CONCLUSION

There is a strong correlation between adenoma detection and nonneoplastic polyp detection. The etiology is unclear, but nonneoplastic polyp detection rate may inflate the PDR for some endoscopists. NNP also adds an increased cost. Increasing the awareness of endoscopic appearances through advanced imaging techniques of normal versus neoplastic tissue may be an area to improve cost containment in screening colonoscopy.

Quality monitoring in colonoscopy: Time to act

Colonoscopy is the gold standard test for colorectal cancer screening. The primary advantage of colonoscopy as opposed to other screening modalities is the ability to provide therapy by removal of precancerous lesions at the time of detection. However, colonoscopy may miss clinically important neoplastic polyps. The value of colonoscopy in reducing incidence of colorectal cancer is dependent on many factors including, the patient, provider, and facility level. A high quality examination includes adequate bowel preparation, optimal colonoscopy technique, meticulous inspection during withdrawal, identification of subtle flat lesions, and complete polypectomy. Considerable variation among institutions and endoscopists has been reported in the literature. In attempt to diminish this disparity, various approaches have been advocated to improve the quality of colonoscopy. The overall impact of these interventions is not yet well defined. Implementing optimal education and training and subsequently analyzing the impact of these endeavors in improvement of quality will be essential to augment the utility of colonoscopy for the prevention of colorectal cancer.

Automatic polyp detection using global geometric constraints and local intensity variation patterns

This paper presents a new method for detecting polyps in colonoscopy. Its novelty lies in integrating the global geometric constraints of polyps with the local patterns of intensity variation across polyp boundaries: the former drives the detector towards the objects with curvy boundaries, while the latter minimizes the misleading effects of polyp-like structures. This paper makes three original contributions: (1) a fast and discriminative patch descriptor for precisely characterizing patterns of intensity variation across boundaries, (2) a new 2-stage classification scheme for accurately excluding non-polyp edges from an overcomplete edge map, and (3) a novel voting scheme for robustly localizing polyps from the retained edges. Evaluations on a public database and our own videos demonstrate that our method is promising and outperforms the state-of-the-art methods.

A Comprehensive Computer-Aided Polyp Detection System for Colonoscopy Videos

Computer-aided detection (CAD) can help colonoscopists reduce their polyp miss-rate, but existing CAD systems are handicapped by using either shape, texture, or temporal information for detecting polyps, achieving limited sensitivity and specificity. To overcome this limitation, our key contribution of this paper is to fuse all possible polyp features by exploiting the strengths of each feature while minimizing its weaknesses. Our new CAD system has two stages, where the first stage builds on the robustness of shape features to reliably generate a set of candidates with a high sensitivity, while the second stage utilizes the high discriminative power of the computationally expensive features to effectively reduce false positives. Specifically, we employ a unique edge classifier and an original voting scheme to capture geometric features of polyps in context and then harness the power of convolutional neural networks in a novel score fusion approach to extract and combine shape, color, texture, and temporal information of the candidates. Our experimental results based on FROC curves and a new analysis of polyp detection latency demonstrate a superiority over the state-of-the-art where our system yields a lower polyp detection latency and achieves a significantly higher sensitivity while generating dramatically fewer false positives. This performance improvement is attributed to our reliable candidate generation and effective false positive reduction methods.

Mycophenolate mofetil-induced segmental colitis mimicking ischemic colitis

Mycophenolate mofetil (MMF) is a commonly used drug in the prevention of allograft rejection in patients with solid organ transplants. Although diffuse colitis has been described in MMF-related colitis, segmental colitis has not been reported. We report the case of a 64-year-old male on MMF therapy who presented for evaluation of afebrile diarrhea and abdominal pain. Flexible sigmoidoscopy revealed a segmental erythematous mucosa with ulceration in the sigmoid colon, descending colon, splenic flexure and proximal transverse colon. Biopsies of these areas showed dilated damaged crypts, eosinophilic epithelial changes and crypt abscesses with apoptotic bodies consistent with MMF-induced injury. MMF was discontinued, leading to a significant improvement of his symptoms.

Impact of fentanyl in lieu of meperidine on endoscopy unit efficiency: a prospective comparative study in patients undergoing EGD

BACKGROUND

Turnaround time is an important component of endoscopy unit efficiency. Any reduction in the total time from patient arrival in the endoscopy room to departure from the recovery area may translate into better endoscopy unit efficiency.

OBJECTIVE

To evaluate the effects on endoscopy unit efficiency of a change in narcotic choice for moderate sedation in patients undergoing EGD at an ambulatory surgery center.

DESIGN

Prospective, comparative, quality-improvement project.

SETTING

Endoscopy unit of a tertiary-care academic medical center.

PATIENTS

We enrolled consecutive patients (n = 1963) who underwent outpatient EGD by 1 of 5 endoscopists between November 2008 and November 2010.

INTERVENTION

Moderate sedation with midazolam plus fentanyl versus meperidine.

MAIN OUTCOME MEASUREMENTS

Sedation-dependent endoscopy unit efficiency and total procedure time (induction-to-intubation, intubation-to-extubation, and extubation-to-discharge).

RESULTS

Fentanyl was associated with reduced total procedure time by 10.1 minutes resulting from both shorter induction-to-intubation time and extubation-to-discharge time (P < .001). The mean (± SD) sedation-dependent endoscopy unit efficiency was 3.2 (± 1.9) procedures per hour for the meperidine group and 3.9 (± 2.7) procedures per hour for the fentanyl group (P = .012); this would translate into possibly increasing the endoscopy suite efficiency by 22%. Based on dosage equivalency conversion, equal doses of fentanyl and meperidine were used. No sedation-related complications or need for reversal agents were recorded.

LIMITATIONS

No randomization was performed.

CONCLUSION

Compared with meperidine, fentanyl in combination with midazolam was associated with significantly shorter total procedure time. By improving the turnaround time, sedation-dependent endoscopy unit efficiency may be improved by 22%.

Quality metrics in endoscopy

Endoscopy has evolved in the past 4 decades to become an important tool in the diagnosis and management of many digestive diseases. Greater focus on endoscopic quality has highlighted the need to ensure competency among endoscopists. A joint task force of the American College of Gastroenterology and the American Society for Gastrointestinal Endoscopy has proposed several quality metrics to establish competence and help define areas of continuous quality improvement. These metrics represent quality in endoscopy pertinent to pre-, intra-, and postprocedural periods. Quality in endoscopy is a dynamic and multidimensional process that requires continuous monitoring of several indicators and benchmarking with local and national standards. Institutions and practices should have a process in place for credentialing endoscopists and for the assessment of competence regarding individual endoscopic procedures.

Comparison of single- versus double-balloon assisted-colonoscopy for colon examination after previous incomplete standard colonoscopy

BACKGROUND

Colonoscopy fails to achieve cecal intubation in 5-10 % of cases. Many of these cases can be completed using balloon-assisted colonoscopy, either with the single-balloon colonoscopy (SBC) or the double-balloon colonoscopy (DBC) techniques.

AIM

To compare the completion rates of SBC and DBC in patients with previous incomplete conventional colonoscopy.

METHODS

Between August 2009 and July 2011 either SBC or DBC was performed in 53 patients in whom previous conventional colonoscopy did not achieve cecal intubation. The medical records of these 53 patients were reviewed retrospectively for details regarding (1) indication for the initial colonoscopy, (2) patient characteristics, (3) data from the initial colonoscopy, and (4) details on both SBC and DBC.

RESULTS

SBC was successful in intubating the cecum in 100 % (26/26) of patients and DBC was successful in 93 % (25/27) of patients. The median (range) time to reach the cecum was 17 (9-43) min in the SBC group and 20 (7-58) min in the DBC group (P = 0.37). The presence of polyps was an entirely new finding in 35 % (9/26) of patients in the SBC group and 30 % (8/27) of patients in DBC group. Therapeutics were performed in 73 % (19/26) of patients in SBC group and 67 % (18/27) of patients in DBC group.

CONCLUSION

For patients with incomplete conventional colonoscopy, SBC and DBC offer high cecal intubation rates as well as detection of additional polyps and therapeutic capability. Either SBC or DBC can be considered after incomplete conventional colonoscopy.

Increased adenoma detection rate with system-wide implementation of a split-dose preparation for colonoscopy

BACKGROUND

Recent studies using split-dose preparations (SDPs) suggest a significant improvement in the quality of preparation and patient compliance. However, the effects of SDP on other quality indicators of colonoscopy, such as cecal intubation and adenoma detection rates, have not been previously reported, to our knowledge.

OBJECTIVE

The primary objective of this study was to compare polyp detection rates (PDRs) and adenoma detection rates (ADRs) before and after the implementation of an SDP as the preferred bowel preparation. The secondary objectives were to compare the quality of the preparation and colonoscopy completion rates before and after implementation of the SDP.

DESIGN

Retrospective study.

SETTING

Tertiary care medical center.

PATIENTS

Patients undergoing colonoscopy for screening and surveillance of colon polyps and cancer.

INTERVENTIONS

System-wide implementation of SDP.

RESULTS

A total of 3560 patients in the pre-SDP group and 1615 patients in the post-SDP group were included in the study. SDP use increased significantly from 9% to 74% after implementation. In comparison with the pre-SDP group, both PDRs (44.1%-49.5%; P < .001) and ADRs (26.7%-31.8%; P < .001) significantly improved in the post-SDP group. The colonoscopy completion rate significantly increased from 93.6% to 95.5% in the post-SDP group (P = .008). Bowel preparation quality also improved significantly (P < .001) in the post-SDP group.

LIMITATIONS

Retrospective design; not all endoscopists were the same in both periods.

CONCLUSIONS

System-wide implementation of an SDP as the primary choice for colonoscopy significantly improved both PDRs and ADRs, overall quality of the preparation, and colonoscopy completion rates.

Comparison of the yield and miss rate of narrow band imaging and white light endoscopy in patients undergoing screening or surveillance colonoscopy: a meta-analysis

OBJECTIVES

Colonoscopy has an appreciable miss rate for adenomas and colorectal cancer. The goal of advanced endoscopic imaging is to improve lesion detection. Compared with standard definition, high-definition (HD) colonoscopes have the advantage of increased field of visualization and higher resolution; narrow band imaging (NBI) utilizes narrow band filters for enhanced visualization of surface architecture and capillary pattern. The objective of this study was to compare the yield and miss rates of HD-NBI and HD-WLE (white light endoscopy) for the detection of colon polyps using meta-analysis.

METHODS

A recursive literature search of randomized controlled trials (RCTs) comparing the yield of HD-NBI and HD-WLE for detection of colon polyps in patients undergoing screening/surveillance colonoscopy. Authors were contacted for missing data. In RCT with tandem colonoscopy (RCT-t), findings from the first-pass examinations were used in the yield analysis and from the tandem pass for the miss rate analysis. Data on the yield of polyps were extracted, pooled, and analyzed using RevMan 4.2.9 software. Odds ratio (OR) and 95% confidence intervals (CIs) for the pooled data for the yield and miss rates of NBI and WLE were calculated. A fixed effect model (FEM) was used for analyses without, and a random effect model (REM) for analyses with heterogeneity.

RESULTS

The yield analysis revealed no significant difference between HD-NBI and HD-WLE for the detection of adenomas (six studies; n=2,284; OR: 1.01; CI: 0.74-1.37; REM); patients with polyps (six studies; n=2,275; OR: 1.15; CI: 0.8-1.64; REM); patients with adenomas (four studies; n=2,177; OR: 1.0; CI: 0.83-1.20; FEM); detection of adenomas <10 mm (five studies; n=1,618; OR: 1.32; CI: 0.92-1.88; FEM); flat adenomas (five studies; n=1,675; OR: 1.26; CI: 0.62-2.57; REM); and flat adenomas per patient (five studies; n=2,200; OR: 1.63; CI: 0.71-3.74; REM). The miss rate analysis revealed no difference in polyp miss rate (three studies; n=524; OR: 1.17; CI: 0.8-1.71; FEM) or adenoma miss rate (three studies; n=524; OR: 0.65; CI: 0.4-1.06; FEM) between the two techniques.

CONCLUSIONS

Compared with HD-WLE, HD-NBI does not increase the yield of colon polyps, adenomas, or flat adenomas, nor does it decrease the miss rate of colon polyps or adenomas in patients undergoing screening/surveillance colonoscopy.

Successful colonoscopic fecal transplant for severe acute Clostridium difficile pseudomembranous colitis

Clostridium difficile-associated diarrhea has become one of the most common healthcare-associated infections, with significant morbidity and mortality, especially among the elderly in the inpatient setting. The standard approach with metronidazole and vancomycin is not very effective in treating patients with severe colitis and hence other alternatives have been explored. We herein describe the first successful experience of colonoscopic fecal transplant in a case of severe refractory C. difficile pseudomembranous colitis.

Multifocal Carcinoid Tumor of Small Intestine: A Rare Cause of Chronic Obscure Gastrointestinal Bleeding, Suspected on Capsule Endoscopy and Diagnosed on Double Balloon Enteroscopy

We reported a case of multifocal carcinoid tumor of small intestine causing chronic obscure gastrointestinal bleeding, suspected on capsule endoscopy and diagnosed on double balloon enteroscopy.

Sessile serrated adenomas: Demographic, endoscopic and pathological characteristics

AIM: To study the demographic and endoscopic characteristics of patients with sessile serrated adenoma (SSA) in a single center.

METHODS: Patients with SSA were identified by review of the pathology database of Mayo Clinic Arizona from 2005 to 2007. A retrospective chart review was performed to extract data on demographics, polyp characteristics, presence of synchronous adenomatous polyps or cancer, polypectomy methods, and related complications.

RESULTS: One hundred and seventy-one (2.9%) of all patients undergoing colonoscopy had a total of 226 SSAs. The mean (SE) size of the SSAs was 8.1 (0.4) mm; 42% of SSAs were ≤ 5 mm, and 69% were ≤ 9 mm. Fifty-one per cent of SSAs were located in the cecum or ascending colon. Approximately half of the patients had synchronous polyps of other histological types, including hyperplastic and adenomatous polyps. Synchronous adenocarcinoma was present in seven (4%) cases. Ninety-seven percent of polyps were removed by colonoscopy.

CONCLUSION: Among patients with colon polyps, 2.9% were found to have SSAs. Most of the SSAs were located in the right side and were safely managed by colonoscopy.

Complex colon polypectomy

By interrupting the adenoma-carcinoma sequence, endo-scopic polypectomy can prevent the development of colorectal cancer (CRC). Endoscopic polypectomy has, therefore, become an accepted screening and surveillance modality for CRC and has been widely adopted by clinicians, policymakers, and professional organizations as an effective screening tool. Most gastroenterologists are adequately trained to endoscopically excise the majority of polyps found in a routine colonoscopy. However, some polyps, due to their size, location, or configuration, are considered more technically challenging or are associated with an increased risk of complications (such as bleeding or perforation) and, hence, are not routinely resected. These so-called complex polypectomies are the focus of this paper.