Development of a novel image-based program to teach narrow-band imaging

Abstract

Objectives: Narrow-band imaging (NBI) is a widely available endoscopic imaging technology; however, uptake of the technique could be improved. Teaching new imaging techniques and assessing trainees' performance can be a challenging exercise during a 1-day workshop.To support NBI training, we developed an online training tool (Medimq) to help experts train novices in NBI bronchoscopy that could assess trainees' performance and provide feedback before the close of the 1-day course. The present study determines whether trainees' capacity to identify relevant pathology increases with the proposed interactive testing method. Methods: Two groups of 20 and 18 bronchoscopists have attended an NBI course where they did a pretest and post-test before and after the main lecture, and a follow-up test 4 weeks later to measure retention of knowledge. We measured their ability to mark normal and abnormal 'biopsy size' areas on bronchoscopic NBI images for biopsy. These markings were compared with areas marked by experts on the same images. Results: The first group results were used to pilot the test. After modifications, the results of the improved test for group 2 showed trainees improved by 32% (total class average normalized gain) in detecting normal or abnormal areas. On follow-up testing, Group 2 improved by 23%. Conclusions:The overall class average normalized gain of 32% shows our test can be used to improve trainees' competency in analyzing NBI Images. The testing method (and tool) can be used to measure the follow up 4 weeks later. Better follow-up test results would be expected with more frequent practice by trainees after the course.

A Web-Based Education Program for Colorectal Lesion Diagnosis with Narrow Band Imaging Classification

“…This suggests that NBI effectively provided enhanced visualization of morphological alterations in vessel patterns and surface patterns of colorectal lesions that are unclear under the WL mode, and that this classification system effec­tively characterized their specific features. Also, higher interobserver agreement using the NBI mode in the expert group (0.71) as well as in the trainee group (0.69) compared to the previously reported data (0.31–0.69) [5, 8, 10] suggests that this simplified classification system might have resulted in this improvement.…”

“…For NBI training, sufficient exposure to key features in NBI findings is required to perform optical diagnosis comfortably [3]. Although didactic lectures of NBI diagnosis could be provided by experienced endoscopists [8-10], availability of and access to such a program are limited. As such, a widely available online, self-learning educational module may be useful.…”

Vascular pattern recognition for narrow-band imaging bronchoscopy

“…This dataset is augmented with data from the Dumas et al NBI training program. 6 Next, we introduce our supervised learning approach to train a convolutional neural network (CNN) to achieve high accuracy given a relatively small dataset.…”

“…These criteria, known as "Shibuya descriptors", summarize pathological features of suspicious vessel structures including complex network of tortuous vessels, spiral and screw-type vessels, abrupt-ending vessels, and dotted vessels. 4 For our work, we consider three vascular groups to entail these pathological patterns, similar to the Dumas et al NBI observer study 6 In particular, we propose a method for identifying pathological patterns observed in NBI bronchoscopic images, thereby enabling us to distinguish (pre-)malignant lesions from benign. Our method takes advantage of a metric learning approach, instead of direct classification, based on Siamese networks to learn pathological features from a small annotated NBI dataset.…”

Bronchoscopic video synchronization for interactive multimodal inspection of bronchial lesions

“…At present, this task is not only impractical but also error prone and likely to result in missed lesion sites. 3,4 As a step toward addressing these problems, the video streams can be recorded for later off-line review, but there still is no available means for relating individual video sources and linking multimodal video findings at correlated sites. Thus, methods still do not exist to help exploit the considerable potential of multimodal bronchoscopy for lesion detection.…”