Benchtop Training Research Articles

Development and validation of a novel home-made bench-top training model for retrograde intrarenal surgery.

To develop and validate a low-cost homemade bench-top training model to facilitate retrograde intrarenal surgery (RIRS) training. The RIRS training model (G-Model) was developed using a surgical glove and a recycled ureter access sheath. Fifteen participants including 10 residents and 5 urologists were enrolled. Designed training curriculum for residents was carried out. Face validity, content validity, construct validity and criterion validity evaluation of the G-Model were carried out. The global score of face and content validity was 4.15 ± 0.53 and 4.65 ± 0.29, respectively. For construct validity, the overall modified global rating scale (mGRS) score was significantly improved [12.5 (5.25) vs. 24.0 (5.25), p = 0.004], and the total task time was significantly shortened (39.5 ± 4.48min vs. 24.1 ± 3.81min, p < 0.001) within residents after G-Model training. The baseline mGRS score and total task time of residents were poorer than those of urologists [12.5 (5.25) vs. 32.0 (1.00), p < 0.001; 39.5 ± 4.48min vs. 16.0 ± 1.58min, p < 0.001]. Spearman correlation analysis revealed strong correlations between residents' G-Model and real patient performance. The current study presented a valid low-cost easily accessible RIRS bench-top training model which could facilitate skill acquisition and translate to real-life scenario.

“Sawbones”: A pilot study assessing simulation-based orthopedic training for medical students

IntroSimulation-based training has become a valuable new tool in medical education across the country. The Orthopedic Surgery Interest Group (OSIG) at the University of Miami Miller School of Medicine organized a benchtop training workshop known as “Sawbones” to give medical students essential exposure to basic techniques and instruments commonly used in orthopedic surgery. This pilot study seeks to investigate the participating students’ perceptions of this workshop as part of a potential longitudinal intervention. MethodsA total of 30 medical students (MS1-MS4) with a documented interest in orthopedic surgery were randomly selected via email invitation to participate in this workshop. Students first had a lecture-based training session with faculty on an overview of screw fixation. Participants then formed groups that were headed by an orthopedic resident or attending and took turns fixing fractures on model bones made of synthetic material. Following the session, students were sent an anonymous Qualtrics survey to assess their satisfaction with the workshop. ResultsA total of 22 students (73%), responded to the survey. On a ten-point scale, the average reported interest in orthopedics was 9.2 (SD 0.4). All students (n = 22, 100%) reported that they would like more hands-on orthopedic experiences as a component of their medical education. Nineteen students (86.4%) reported that this training increased their interest in pursuing a career in orthopedic surgery and twenty-one (95.5%) further stated that they would recommend this training to other students. DiscussionBased on the results of this pilot study, Sawbones was promising at engaging students in orthopedics, providing a team environment, and introducing students to orthopedic skills and instrumentation. Providing these sessions in a longitudinal manner could provide opportunities for mentorship and better prepare students for a residency in orthopedics. We recommend a program like Sawbones to better address the lack of Orthopedic exposure in medical school.

Validation of a Bench-Top Training Model for Retrograde Intrarenal Surgery

Objective: To validate the in-house built Styrofoam box bench-top training model for retrograde intrarenal surgery (RIRS). Materials and Methods: This study was performed in the setting of a half-day RIRS course. During the course, participants performed RIRS on a locally built bench-top model. We recruited 26 participants, comprising 20 trainees and 6 experts. Trainees and experts were asked to fill a self-administered questionnaire assessing various components of RIRS to assess the face and content validity. For construct validity, experts using Objective Structured Assessment of Technical Skills (OSATS) rated trainees and the mean OSATS score of junior versus senior residents was compared. Results: As per trainees, the model was 86% (4.3/5) realistic, which was backed by experts who found this model to be 87.6% (4.38/5) realistic. The overall face validity of the model was 86.4% (4.32/5). The overall content validity of this model was 83.4% (4.17/5). Majority of the participants thought that this tool is useful for learning technique (4.38 ± 0.49) and safe-conduct (4.31 ± 0.73) of the procedure. The trainees concurred that the skills acquired are transferable to the operating room (4.23 ± 0.76). However, the construct validity by comparing mean OSATS score of junior versus senior residents was 19.5 ± 3.5 and 23 ± 4.5. A p value of 0.11 could not be established. Conclusion: This study has demonstrated the face, content, and construct validity of the bench-top training model for RIRS. Further evaluation is necessary to compare its effectiveness against other available models to demonstrate concurrent validity.

Automatic task recognition in a flexible endoscopy benchtop trainer with semi-supervised learning.

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. 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. 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 17h 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 3.65s, with a mean task duration of 159s ([Formula: see text] relative error). For the second dataset, we obtain a mean task duration estimation error of 3.67s. We reduce an average of 5.63% in error to 3.67% thanks to our semi-supervised learning approach. 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.

Skills in Open Surgery do not Predict Arthroscopic Skills in Medical Students

Teaching of basic open surgical skills like skin closure has been well established in medical schools' surgical curricula. The same cannot be said for basic arthroscopic skills even though the importance of arthroscopic surgery in Orthopaedics has been well documented. The goal of this study was to evaluate the relationship between basic skills in open surgery and basic arthroscopic skills in a cohort of medical students. Our hypothesis was that performance in open surgical skills does not correlate with skills in arthroscopic surgery. The performance of medical students in basic tasks in open surgery (porcine model: subcutaneous and skin suturing) and arthroscopic surgery (bench top model: arthroscopic triangulation and object transfer) was assessed. For both surgical techniques an introductory course was given followed by a standardized assessment. After 1 week both open and arthroscopic tasks were reassessed. All procedures were recorded and scored by 2 independent observers in a blinded fashion. The study was performed at the skills lab of the Clinic for Orthopaedics and Tumororthopaedics at University Hospital Muenster in Muenster Germany. A cohort of 21 medical students (average age 22.2 years) participated in this study. The cohort consisted of 17 female (81%) and 4 male (19%) students. All students were in the clinical part of their study and had not received any prior surgical education. Interobserver reliability was very high for the arthroscopic tasks and high for the open surgery tasks. No correlation was found between open and arthroscopic skills. For the first assessment the correlation coefficient was r = 0.197 (p = 0.391). For the second assessment the correlation coefficient was r = 0.212 (p = 0.356). Significant improvement from first to second assessment was only found for the arthroscopic tasks (p < 0.0001) while improvement in the open surgery performance failed to reach statistical significance (p = 0.184). The results of this study suggest that performance in open surgical skills does not correlate with performance in arthroscopic skills and should be taught independent from each other. Arthroscopic skills can effectively be taught with bench top training systems in a time- and resource efficient manner, with measurable results even in a 1-day-course setup. Arthroscopic skills training may be offered in undergraduate curricula in addition to open surgical skills training to students with an interest in Orthopaedics.

Time to oxygenation for cannula‐ and scalpel‐based techniques for emergency front‐of‐neck access: a wet lab simulation using an ovine model

Emergency front-of-neck access to achieve a percutaneous airway can be a life-saving intervention, but there is debate about the preferred technique. This prospective, observational study was designed to compare the two most common emergency surgical airway techniques in a wet lab simulation using an ovine model. Forty-three doctors participated. After providing standardised reading, a lecture and dry lab benchtop training, participants progressed to a high-fidelity wet lab simulation. Participants entered an operating theatre where a 'cannot intubate, cannot oxygenate' situation had been declared and were directed to perform emergency front-of-neck access: first with a cannula technique (14-gauge cannula insertion with ventilation using a Rapid-O2® cricothyroidotomy insufflation device); and subsequently, a scalpel-bougie technique (surgical incision, bougie insertion into trachea and then tracheal tube passed over bougie, with ventilation using a self-inflating bag). The primary end-point was time from declaration of 'cannot intubate, cannot oxygenate' to delivery of oxygen via a correctly placed percutaneous device. If a cannula or tracheal tube was not placed within 240s, the attempt was marked as a failure. There was one failure for the cannula approach and 15 for the scalpel-bougie technique (OR 0.07 (95%CI 0.00-0.43); p <0.001). Median (IQR [range]) time to oxygenation, if successful, was 65 (57-78 [28-160]) s for the cannula approach and 90 (74-115 [40-265]) s for the scalpel-bougie technique (p=0.005). In this ovine model, emergency front-of-neck access using a cannula had a lower chance of failure and (when successful) shorter time to first oxygen delivery compared with a scalpel-bougie technique.

Simulation-Based Training Platforms for Arthroscopy: A Randomized Comparison of Virtual Reality Learning to Benchtop Learning

To determine whether a virtual reality (VR) arthroscopy simulator or benchtop (BT) arthroscopy simulator showed superiority as a training tool. Arthroscopic novices were randomized to a training program on a BT or a VR knee arthroscopy simulator. The VR simulator provided user performance feedback. Individuals performed a diagnostic arthroscopy on both simulators before and after the training program. Performance was assessed using wireless objective motion analysis and a global rating scale. The groups (8 in the VR group, 9 in the BT group) were well matched at baseline across all parameters (P > .05). Training on each simulator resulted in significant performance improvements across all parameters (P < .05). BT training conferred a significant improvement in all parameters when trainees were reassessed on the VR simulator (P < .05). In contrast, VR training did not confer improvement in performance when trainees were reassessed on the BT simulator (P > .05). BT-trained subjects outperformed VR-trained subjects in all parameters during final assessments on the BT simulator (P < .05). There was no difference in objective performance between VR-trained and BT-trained subjects on final VR simulator wireless objective motion analysis assessment (P > .05). Both simulators delivered improvements in arthroscopic skills. BT training led to skills that readily transferred to the VR simulator. Skills acquired after VR training did not transfer as readily to the BT simulator. Despite trainees receiving automated metric feedback from the VR simulator, the results suggest a greater gain in psychomotor skills for BT training. Further work is required to determine if this finding persists in the operating room. This study suggests that there are differences in skills acquired on different simulators and skills learnt on some simulators may be more transferable. Further work in identifying user feedback metrics that enhance learning is also required.

Newly acquired arthroscopic skills: Are they transferable during simulator training of other joints?

This randomized study investigates whether novices learning simulation-based arthroscopic skills in one anatomical joint environment can immediately transfer their learnt skills to another joint. Medical students were randomized to a simulated diagnostic knee or shoulder arthroscopic task on benchtop training models. After nine task repetitions over 3weeks on one model, each participant undertook the simulation task of the other anatomical joint. Performance was objectively measured using a validated electromagnetic motion analysis system and a validated global rating scale (GRS). Eighteen students participated; eight started the knee task and ten the shoulder task. All participants demonstrated a learning curve in all parameters during task repetition (time taken, hand path length, number of hand movements and GRS scores; p<0.001) with learning effects >1 SD from initial performance (range 1.1-2.2 SD). When the groups swapped models, there was no immediate evidence of skill transfer, with a significant drop in performance between the final training episode and the transfer task (all parameters p<0.003). In particular, the transfer task performance was no better than the first episode performance on that model by these novices. This study showed basic arthroscopic skills did not immediately transfer to an unfamiliar anatomical environment within a simulated setting. These findings have important clinical implications with regard to surgical training as they potentially challenge the assumption that arthroscopic skills acquired in one joint are universally transferrable to other joints. Future orthopaedic simulation training should aim to deliver exposure to a greater variety of arthroscopic procedures and joint environments. This would allow trainees to become familiar with the different arthroscopic setting before undertaking real surgery and consequently improve patient safety. Therapeutic, Level II.

Feasibility of a Low-Cost Instrumented Trocar for Universal Surgical Procedure Analyses1

The current path to become a surgeon follows the Halstedjan model of apprentice-style training, in which a resident remains under the shadow of a senior surgeon [1,2]. The resident starts by observing the surgeon and then gradually moves on to assist him. Though the process provides immediate mentor feedback through one-on-one interaction, it fails to carry out a systematic and consistent evaluation. All of the assessment is based on a senior surgeon’s subjective appraisal. Moreover, the requirement of performing a real procedure under the supervision of an experienced surgeon puts extra financial pressure on the medical school curriculum [2]. The advent and advancement of minimally invasive surgery (MIS) have made mastering such skills even more difficult, particularly psychomotor. Due to the increased complexity in required surgical skills and due to a need for a more consistent and systematic evaluation, there has been an increasing demand for an objective assessment of surgical skill. Studies in the past have examined motion analyses of surgeon’s hands and tool with a structured checklist to conclude that the motion analyses can serve as one of the measures for objectively assessing surgical skill [3]. A considerable amount of work has been done in the past to develop systems for tracking minimally invasive surgical tools. However, there does not exist a system that can be easily deployed to all three environments, i.e., virtual reality, physical benchtop, and operating room (OR). Most of these systems are only applicable to either virtual reality or physical benchtop training environment. There is only one system, Waseda bioinstrumentation 3, that can be applied in the OR. However, even that system is not validated yet and has an intrusive setup. Moreover, these systems can be very expensive with the average price ranging from $20,000 to $30,000. Hence, there is a strong need for a low-cost, noninvasive means to quantitatively discriminate surgical skill in a manner universally applicable to all environments and various surgical modalities like robotic, laparoscopic and endoscopic surgeries. This work presents the feasibility of such a system. Notably, we evaluate the feasibility of jerk cost as a low-cost alternative to skill evaluation. We hypothesize it can discriminate skill without expensive position tracking. 2 Methods

A Novel Approach to Contextualized Surgical Simulation Training

Decontextualized benchtop simulators play a key role in surgical training. Educational theory highlights the importance of context for effective learning, yet existing full-immersion simulation facilities are prohibitively expensive. This study explored the concept of contextualized training of a key surgical procedure in a novel, low-cost, high-fidelity simulation environment [distributed simulation (DS)] and compared it with decontextualized training in a traditional benchtop simulation. Eleven novice surgeons performed a small bowel anastomosis in a benchtop simulation and in DS. Likert-type questionnaires and semistructured interviews captured surgeons' responses. Surgeons perceived the contextualized DS training as realistic, giving surgeons the opportunity to increase their confidence, motivation, and integration of technical and nontechnical skills. Distributed simulation would be of value after benchtop training and before surgery in a real operating room. Contextualized simulation training allows junior surgeons to gain the necessary confidence before operating in a real operating room. This article presents DS as a solution to widen access to contextualized surgical training.