Background and study aims
Helicobacter pylori (HP)-associated chronic gastritis can cause mucosal atrophy and intestinal metaplasia, both of which increase the risk of gastric cancer. The accurate diagnosis of HP infection during routine medical checks is important. We aimed to develop a convolutional neural network (CNN), which is a machine-learning algorithm similar to deep learning, capable of recognizing specific features of gastric endoscopy images. The goal behind developing such a system was to detect HP infection early, thus preventing gastric cancer.
Patients and methods
For the development of the CNN, we used 179 upper gastrointestinal endoscopy images obtained from 139 patients (65 were HP-positive: ≥ 10 U/mL and 74 were HP-negative: < 3 U/mL on HP IgG antibody assessment). Of the 179 images, 149 were used as training images, and the remaining 30 (15 from HP-negative patients and 15 from HP-positive patients) were set aside to be used as test images. The 149 training images were subjected to data augmentation, which yielded 596 images. We used the CNN to create a learning tool that would recognize HP infection and assessed the decision accuracy of the CNN with the 30 test images by calculating the sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC).
Results
The sensitivity and specificity of the CNN for the detection of HP infection were 86.7 % and 86.7 %, respectively, and the AUC was 0.956.
Conclusions
CNN-aided diagnosis of HP infection seems feasible and is expected to facilitate and improve diagnosis during health check-ups.
This paper proposes a point groups-based algorithm for point cloud registration. Most of the existing algorithms align two point clouds globally; however, they are unsuitable when the overlapping ratio is low or the inputs do not have strong features. The high accuracy of matched points is conducive for a rigid transformation of point clouds. This study aims to determine the exact matching points to register point clouds. The proposed method is based on point groups that are resampled point clouds. Subsequently, we calculate the multiple average probability (MAP) for each point group and match them by a sparse representation. Finally, the coherent point drift (CPD) algorithm is used to register the matched point groups, and the same transformation is applied to register the point clouds. The experimental results show that in terms of robustness to noise and outliers, our algorithm can register point clouds with a low overlapping ratio.
Point cloud registration is an important part of 3-dimensional information processing. Low overlap ratio, noise, outliers, and missing points considerably influence the registration results. In this paper, we propose a fast and robust point cloud registration method to reduce the impact of these factors. First, the point groups are resampled by point clouds as basic elements for point cloud registration. Second, singular value decomposition is used to decompose the point groups. Third, the depth image of the point groups is calculated, and the sparse feature is obtained using the depth image. Finally, the sparse feature is used to obtain registration results through sparse representation. Under the premise of robustness to low overlap ratio, noise, outliers, and missing points, experimental results show that our algorithm is faster and more accurate than extant methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.