Search

Purpose: Inexpensive benchtop training systems offer significant advantages to meet the increasing demand of training surgeons and gastroenterologists in flexible endoscopy. Established scoring systems exist, based on task duration and mistake evaluation. However, they require trained human raters, which limits broad and low-cost adoption. There is an unmet and important need to automate rating with machine learning., Method: We present a general and robust approach for recognizing training tasks from endoscopic training video, which consequently automates task duration computation. Our main technical novelty is to show the performance of state-of-the-art CNN-based approaches can be improved significantly with a novel semi-supervised learning approach, using both labelled and unlabelled videos. In the latter case, we assume only the task execution order is known a priori., Results: Two video datasets are presented: the first has 19 videos recorded in examination conditions, where the participants complete their tasks in predetermined order. The second has 17 h of videos recorded in self-assessment conditions, where participants complete one or more tasks in any order. For the first dataset, we obtain a mean task duration estimation error of 3.65 s, with a mean task duration of 159 s ([Formula: see text] relative error). For the second dataset, we obtain a mean task duration estimation error of 3.67 s. We reduce an average of 5.63% in error to 3.67% thanks to our semi-supervised learning approach., Conclusion: This work is the first significant step forward to automate rating of flexible endoscopy students using a low-cost benchtop trainer. Thanks to our semi-supervised learning approach, we can scale easily to much larger unlabelled training datasets. The approach can also be used for other phase recognition tasks.

Background: Cholecystostomy is commonly performed in high-risk patients with acute cholecystitis. However, internal drainage may be more desirable in patients as it is associated with lower complication rates. This paper describes an image-guided, percutaneous technique for internal gallbladder drainage using a covered lumen-apposing metal stent (LAMS) and assesses its feasibility and safety in a porcine model., Methods: Procedures were performed on 30-kg pigs. Under ultrasound and fluoroscopic guidance, a percutaneous puncture was performed through-and-through the gallbladder into the stomach. A guidewire was placed and a 12Fr sheath was advanced through which a 10-mm LAMS was deployed. Its distal flange was deployed in the gastric lumen, and its proximal flange in the gallbladder. The cholecystoenteric anastomosis was examined by means of endoscopy, laparoscopy, and necropsy., Results: Technical success was 100% (7/7). Procedure times decreased with experience and improvements in technique (median: 22 min). Contrast injection demonstrated free flow through the stent with no leakage. Necropsy confirmed appropriate stent position with good apposition of gallbladder and stomach, and no intraprocedural complications were detected., Conclusions: Image-guided, percutaneous, internal gallbladder drainage using a LAMS is safe and feasible in a porcine model. This technique may be an alternative to endoscopic ultrasound-guided stent placement and external cholecystostomy tube drainage.

Background: Both laparoscopy and endoscopy are image-based procedures, which are less intuitive than traditional open surgery and require extensive training to reach adequate proficiency. Currently, there is lack of understanding as to how the skills in one image-based procedure translate to another, such as endoscopy to laparoscopy and vice versa. The aim of our study was to explore the relationship between endoscopic and laparoscopic skills using a Fundamentals of Laparoscopic Surgery (FLS) trainer, a traditional virtual reality endoscopic trainer and a "desk-top" endoscopic physical simulator., Methods: Senior surgical residents from across Canada participating in an advanced laparoscopic foregut training course were enrolled in the study. Participants were assessed performing the FLS laparoscopic suturing task, the Endobubble 2 task (Simbionix, GI Mentor), and a forward viewing peg transfer on the novel Basics in Endoscopic Skills Training Box (BEST Box)., Results: There was significant correlation between the participant's skill in simulated laparoscopic suturing and simulated endoscopic skill using the BEST box (Pearson coefficient (r) was 0.551 (p = 0.033) and the coefficient of determination (r 2 ) was 0.304). There was a trend towards correlation between laparoscopic suturing time and Endobubble 2 score, but this did not reach statistical significance (r = 0.458, p = 0.086; r 2  = 0.210)., Conclusions: Performance in the two physical simulators, laparoscopic suturing and simulated flexible endoscopy using the BEST box, showed a correlation. This study adds to the growing body of evidence that laparoscopic and endoscopic skills are complementary and has the potential to impact simulation training involving both skill sets.

Background: The fundamentals of laparoscopic surgery (FLS) training box is a validated tool, already accessible to surgical trainees to hone their laparoscopic skills. We aim to investigate the feasibility of adapting the FLS box for the practice and assessment of endoscopic skills. This would allow for a highly available, reusable, low-cost, mechanical trainer., Methods: The design and development process was based on a user-centered design, which is a combination of the design thinking method and cognitive task analysis. The process comprises four phases: empathy, cognitive, prototyping/adaptation, and end user testing. The underlying idea was to utilize as many of the existing components of FLS training to maintain simplicity and cost effectiveness while allowing for the practice of clinically relevant endoscopic skills. A sample size of 18 participants was calculated to be sufficient to detect performance differences between experts and trainees using a two tailed t test with alpha set at 0.05, standard deviation of 5.5, and a power of 80%., Results: Adaptation to the FLS box included two fundamental attachments: a front panel with an insertion point for an endoscope and a shaft which provides additional support and limits movement of the scope. The panel also allows for mounting of retroflexion tasks. Six endoscopic tasks inspired by FLS were designed (two of which utilize existing FLS components). Pilot testing with 38 participants showed high user's satisfaction and demonstrated that the trainer was robust and reliable. Task performance times was able to discriminate between trainees and experts for all six tasks., Conclusions: A mechanical, reusable, low-cost adaptation of the FLS training box for endoscopic skills is feasible and has high user satisfaction. Preliminary testing shows that the simulator is able to discriminate between trainees and experts. Following further validation, this adaptation may act as a supplement to the FES program.