Search

Conducted from April 2019 to July 2020, this study aimed to assess the software accuracy in assisting Z-line segmentation by comparing with medical doctors’ detection results and by recording doctors’ satisfaction in scale, time-of-implementation in interactive mode, and integrated mode. For the development of the Z-line detection algorithm, a dataset of 533 high-definition endoscopic WLI (white-light) images in diverse forms of Z-line were collected. The software was subsequently developed in 4 modes, including manual mode, interactive mode (using Superpixels-BPT), automatic mode (using AI algorithm), and integrated mode (the combination of BPT and U-Net). 30 endoscopic images were assigned to 2 groups of doctors (under 2-year experience and over 5-year experience) for the Z-line detection using the software in 4 modes. Time-of-implementation, number of mouse clicks, satisfaction in scales, and IoU (Interception over Union) metric with expert’s ground-truth are used for assessment. The results showed that IoU metrics of interactive and integrated modes in the experimental dataset was 88% with no statistical difference to the IoU value of manual mode, and mean IoU metrics from the results of 4 modes were high, from 86.7 to 90.8%. The mean values of time-of-implementation in interactive mode and integrated mode were not statistically different from manual mode. The median number of mouse-clicks each use in the interactive mode and the integrated mode were 24.5 and 15.5 times, respectively. The software received good feedbacks from the doctors, with the mean values of satisfaction scores of automatic mode, interactive mode and integrated mode are 7.2, 7.3, and 7.2 respectively. The development of the software for detecting endoscopic anatomy landmarks is a novel and feasible research direction in Vietnam. Further studies could focus on detecting some specific lesions classified according to anatomy landmarks.

The presence of water bubbles, foam, mucus or residual food in the gastrointestinal tract is a common problem during Upper gastrointestinal endoscopy (UGIE). This condition could cause difficulties to detect lesions as well as perform interventions if required for medical doctors. In every medical center all over the world, analyzing endoscopic images with bubbles is becoming a vital practical issue. In this paper, we propose an automatic scheme to segment regions of water bubbles thanks to the recent advantages of a deep neural network - U-net. Based on the segmentation results, we construct a classification model to evaluate the cleaning level of the current examined images. The classification model utilizes features which directly extracted from the segmented areas such as the number of the water bubbles, their concentrating distribution or scattering in the current view. The proposed method is evaluated on a testing dataset validated by the endoscopists. Accuracy of the quality assessment by the proposed techniques achieves the rate of 90%. Therefore, the proposed method presents a feasible tool for automatically eliminating of unclean UGIE images.

In this paper, we aim to develop a diagnostic assistant system for labelling the stomach anatomical locations in upper GastroIntestinal Endoscopy (UGIE) examination. To address this task, we construct an appropriate manner which utilizes both ability of a convolutional neural network (CNN) and supportive interactions between machine and doctors. We solve the problem by a two-phase scheme. The first is a coarse-phase to classify seven major anatomical locations including cardiac orifice, gastric body, fundic, antrum, pyloric ring, lesser curvature and greater curvature using advances of a CNN. The constructed CNN network is compact with high performance and appropriate integration into a Graphic User Interface (GUI). In order to classify with 13 further detailed positions, a GUI is developed so that the endoscopists can conveniently specify the anatomical locations from results of the coarse-phase. In this the fine-phase, the doctors will prune the automatic results as well as specify the more detailed positions of each major location. In the experimental results, the developed application is shown as an efficient way in an UGIE diagnosis. It reduces a significant time from averagely 13:03 minutes in a manual procedure to 4:35 minutes by using the developed system when comparing with trainee endoscopists. The results of specifying anatomical locations satisfied accuracy requirements and showed promising research trend for future application as a computer-aided GIE diagnostic system.