Faculty Bibliography

Introduction: EUS-guided FNA and FNB have been stagnating at accuracy levels which falls short of patient and physician hopes. False negative results for malignancy continue to range from 4% (expert centers) to 30%. Multiple passes of the needles often slide along the same tract as the first pass, limiting diagnostic yield. FNA requires 5 passes, fewer for the FNB needles. Rapid on-site cytology may lower the number of passes, but it is logistically and financially challenging, without proving to improve accuracy. Repeated passes bend needles out of shape leading to use of a second needle, with high cost implications. Tumor composition can be a mosaic of desmoplasia, necrosis, cystic areas, and malignant cells. If a needle is in the wrong plane then wrong diagnosis occurs. Often fanning hardly moves the needle. We introduce a new needle prototype to improve diagnostic yield and lower false negatives, with use of a device which has five needles.
Method(s): A multi-pass needle has been designed which ensures five specimens will be from various sites, because the five needles are all penetrating at the same time along different planes within the lesion. This needle apparatus uses the same sheath as a standard 19 gauge needle, but within the sheath it houses a five-lumen plastic tube, which separates between the central 22g needle and the four peripheral 25g needles. The ergonomically-designed handle has a two-phase mechanism which allows the endoscopist to first place the 22g needle within the target lesion, which is the same as the process done today, and to then deploy the four 25g needles, which are pre-bent so as to spread out to a diameter of about one cm, when needles are about 3cm from the scope. The additional needles will each, without fanning, obtain specimens near to the central needle, thus a greater representation of tumor will be obtained, completing the procedure in one pass, with better accuracy due to decreasing false negatives. The two channels for suction apply to the single central needle and to the four peripheral needles. Each needle has its own stylet.
Result(s): The accompanying figure shows the InstaFan, a disruptive technological advance, made to improve EUS biopsies. Three in vivo porcine studies found that the 5 needles delivered resulted in five distinct adequate specimens for pathology diagnosis by the cell block method. Porcine EUS found good visibility of the needles, ease of deployment and no side effects from multiple deployments of the onepass device, both in EUS-guided gastric and duodenal biopsies of porcine pancreas and liver.
Conclusion(s): Future directions of the Onepass will be additional in vivo animal safety studies by additional endoscopists followed by first trials in humans. The goal of the Onepass will be to reduce false negatives, thus increasing optimization of tumor diagnosis. [Formula presented] [Formula presented]
Copyright

BACKGROUND AND AIMS/OBJECTIVE:Artificial intelligence (AI), specifically deep learning, offers the potential to enhance the field of gastrointestinal endoscopy in areas ranging from lesion detection and classification, to quality metrics and documentation. Progress in this field will be measured by whether AI implementation can lead to improved patient outcomes and more-efficient clinical workflow for GI endoscopists. The aims of this article are to report the findings of a multidisciplinary group of experts focusing on issues in artificial intelligence research and applications related to gastroenterology and endoscopy, to review the current status of the field, and to produce recommendations for investigators developing and studying new AI technologies for gastroenterology. METHODS:A multidisciplinary meeting was held on September 28, 2019, bringing together academic, industry, and regulatory experts in diverse fields including gastroenterology, computer and imaging sciences, machine learning, and computer vision, Food and Drug Administration (FDA) and National Institutes of Health (NIH). Recent and ongoing studies in gastroenterology and current technology in AI were presented and discussed, key gaps in knowledge were identified, and recommendations were made for research that would have the highest impact in making advances and implementation in the field of AI to gastroenterology. RESULTS:There was a consensus that AI will transform the field of gastroenterology, particularly endoscopy and image interpretation. Powered by advanced machine learning algorithms, the use of computer vision to endoscopy has the potential to result in better prediction and treatment outcomes for patients with gastroenterology disorders and cancer. Large libraries of endoscopic images, "EndoNet," will be important to facilitate development and application of AI systems. The regulatory environment for implementation of AI systems is evolving, but common outcomes such as colon polyp detection have been highlighted as potential clinical trial endpoints. Other threshold outcomes will be important, as well as clarity on iterative improvement of clinical systems. CONCLUSIONS:Gastroenterology is a prime candidate for early adoption of AI. AI is rapidly moving from an experimental phase to a clinical implementation phase in gastroenterology. It is anticipated that the implementation of AI in gastroenterology over the next decade will have a significant and positive impact on patient care and clinical workflows. Ongoing collaboration among gastroenterologists, industry experts, and regulatory agencies will be important to ensure that progress is rapid and clinically meaningful. However, there are several constraints and areas that will benefit from further exploration, including potential clinical applications, implementation, structure and governance, role of gastroenterologists, and potential impact of AI in gastroenterology.

PURPOSE OF REVIEW/OBJECTIVE:With the intent of maximizing adenoma detection rate (ADR), different devices and colonoscopic innovations have been introduced over the past few years. There are a variety of attachment devices available for use in colonoscopy, all intended to improve ADR. In this review, we evaluate the evidence surrounding the available attachment devices and their impact on ADR. RECENT FINDINGS/RESULTS:Endocuff, Endoring, Transparent cap, G-EYE balloon, and Third Eye Retroscope were all included in this review. Three of the devices, Endocuff, Endoring, and transparent hood or cap, have been shown to increase ADR without significantly altering the use of the colonoscope. Although balloon-assisted colonoscopy is not currently FDA-approved, it has the potential to increase ADR as well. Mechanical enhancement has been shown to complement the current forward-viewing colonoscope, and Endocuff appears to be the most beneficial attachment to improve ADR.

PURPOSE OF REVIEW/OBJECTIVE:This review highlights the history, recent advances, and ongoing challenges of artificial intelligence (AI) technology in colonic polyp detection. RECENT FINDINGS/RESULTS:Hand-crafted AI algorithms have recently given way to convolutional neural networks with the ability to detect polyps in real-time. The first randomized controlled trial comparing an AI system to standard colonoscopy found a 9% increase in adenoma detection rate, but the improvement was restricted to polyps smaller than 10 mm and the results need validation. As this field rapidly evolves, important issues to consider include standardization of outcomes, dataset availability, real-world applications, and regulatory approval. AI has shown great potential for improving colonic polyp detection while requiring minimal training for endoscopists. The question of when AI will enter endoscopic practice depends on whether the technology can be integrated into existing hardware and an assessment of its added value for patient care.

BACKGROUND & AIMS/OBJECTIVE:Mucosal exposure devices improve detection of lesions during colonoscopy and have reduced examination times in uncontrolled studies. We performed a randomized trial of Endocuff Vision vs standard colonoscopy to compare differences in withdrawal time (the primary end point). We proposed that Endocuff Vision would allow complete mucosal inspection in a shorter time without impairing lesion detection. METHODS:Adults older than 40 years undergoing screening or surveillance colonoscopies were randomly assigned to the Endocuff group (n=101, 43.6% women) or the standard colonoscopy group (n=99; 57.6% women). One of 2 experienced endoscopists performed the colonoscopies, aiming for a thorough evaluation of the proximal sides of all haustral folds, flexures, and valves in the shortest time possible. Inspection time was measured with a stopwatch and calculated by subtracting washing, suctioning, polypectomy and biopsy times from total withdrawal time. RESULTS:There were significantly fewer women in the Endocuff arm (P=.0475) but there were no other demographic differences between groups. Mean insertion time with Endocuff was 4.0 min vs 4.4 min for standard colonoscopy (P=.14). Mean inspection time with Endocuff was 6.5 min vs 8.4 min for standard colonoscopy (P<.0001). Numbers of adenomas detected per colonoscopy (1.43 vs 1.07; P=.07), adenoma detection rate (61.4% vs 52%; P=0.21), number of sessile serrated polyps per colonoscopy (0.27 vs 0.21; P=0.12), and sessile serrated polyp detection rate (19.8% vs 11.1%; P= 0.09) were all higher with Endocuff Vision. Results did not differ significantly when we controlled for age, sex, or race. CONCLUSION/CONCLUSIONS:In a randomized trial, we found inclusion of Endocuff in screening or surveillance colonoscopies to decrease examination time without reducing lesion detection. ClinicalTrials.gov no.: NCT03361917.