ABSTRACT
BACKGROUND: The gastroscopic examination is a preferred method for the detection of upper gastrointestinal lesions. However, gastroscopic examination has high requirements for doctors, especially for the strict position and quantity of the archived images. These requirements are challenging for the education and training of junior doctors. OBJECTIVE: The purpose of this study is to use deep learning to develop automatic position recognition technology for gastroscopic examination. METHODS: A total of 17182 gastroscopic images in eight anatomical position categories are collected. Convolutional neural network model MogaNet is used to identify all the anatomical positions of the stomach for gastroscopic examination The performance of four models is evaluated by sensitivity, precision, and F1 score. RESULTS: The average sensitivity of the method proposed is 0.963, which is 0.074, 0.066 and 0.065 higher than ResNet, GoogleNet and SqueezeNet, respectively. The average precision of the method proposed is 0.964, which is 0.072, 0.067 and 0.068 higher than ResNet, GoogleNet, and SqueezeNet, respectively. And the average F1-Score of the method proposed is 0.964, which is 0.074, 0.067 and 0.067 higher than ResNet, GoogleNet, and SqueezeNet, respectively. The results of the t-test show that the method proposed is significantly different from other methods (p< 0.05). CONCLUSION: The method proposed exhibits the best performance for anatomical positions recognition. And the method proposed can help junior doctors meet the requirements of completeness of gastroscopic examination and the number and position of archived images quickly.
Subject(s)
Deep Learning , Gastroscopy , Humans , Gastroscopy/methods , Gastroscopy/education , Stomach/anatomy & histology , Stomach/diagnostic imaging , Neural Networks, ComputerABSTRACT
BACKGROUND: A timely diagnosis of early gastric cancer (EGC) can greatly reduce the death rate of patients. However, the manual detection of EGC is a costly and low-accuracy task. The artificial intelligence (AI) method based on deep learning is considered as a potential method to detect EGC. AI methods have outperformed endoscopists in EGC detection, especially with the use of the different region convolutional neural network (RCNN) models recently reported. However, no studies compared the performances of different RCNN series models. OBJECTIVE: This study aimed to compare the performances of different RCNN series models for EGC. METHODS: Three typical RCNN models were used to detect gastric cancer using 3659 gastroscopic images, including 1434 images of EGC: Faster RCNN, Cascade RCNN, and Mask RCNN. RESULTS: The models were evaluated in terms of specificity, accuracy, precision, recall, and AP. Fast RCNN, Cascade RCNN, and Mask RCNN had similar accuracy (0.935, 0.938, and 0.935). The specificity of Cascade RCNN was 0.946, which was slightly higher than 0.908 for Faster RCNN and 0.908 for Mask RCNN. CONCLUSION: Faster RCNN and Mask RCNN place more emphasis on positive detection, and Cascade RCNN places more emphasis on negative detection. These methods based on deep learning were conducive to helping in early cancer diagnosis using endoscopic images.
Subject(s)
Deep Learning , Stomach Neoplasms , Humans , Stomach Neoplasms/diagnostic imaging , Artificial Intelligence , Gastroscopy , Neural Networks, Computer , Early Detection of Cancer/methodsABSTRACT
INTRODUCTION: The influence of sedation on the endoscopic detection rate of upper gastrointestinal (UGI) early cancer (EC) and precancerous lesions, including high-grade intraepithelial neoplasia (HGIN) and low-grade intraepithelial neoplasia, has not been assessed. The aim of this research is to assess whether the use of sedation can help improve the detection rate of UGI EC and precancerous lesions. The second objective is to evaluate its potential influencing factors. METHODS: The study includes 432,202 patients from a multicenter database from January 2012 to July 2019. Information on endoscopic findings and histology biopsies was obtained from endoscopy quality-control system. Associations of sedation with the detection rate of EC and precancerous lesions were assessed. RESULTS: The sedation group has a higher detection rate of UGI EC and HGIN compared with the no-sedation group, whereas the detection rate of low-grade intraepithelial neoplasia was similar between the 2 groups. There were more cases examined by using staining, image enhancement, or magnifying techniques in the sedation group (P < 0.001). And, the mean observation time was also longer in the sedation group (P < 0.001). The type 0-IIb esophageal HGIN and EC cases were significantly increased in the sedation group. No significant difference was detected on lesion subtypes for gastric HGIN and EC according to the Paris classification. More gastric HGIN and EC were detected at gastric body in the sedation group (P = 0.001). DISCUSSION: Sedation may improve the endoscopic detection rate of EC and HGIN in the UGI tract probably through enhancing the use of accessary endoscopic techniques, prolonging observation time, and taking more biopsies in different locations (see Visual Abstract, Supplementary Digital Content 2, http://links.lww.com/AJG/B926).
Subject(s)
Conscious Sedation , Endoscopy, Gastrointestinal , Esophageal Neoplasms/diagnosis , Precancerous Conditions/diagnosis , Stomach Neoplasms/diagnosis , Adult , Aged , Biopsy , Early Detection of Cancer , Esophageal Neoplasms/pathology , Female , Humans , Male , Middle Aged , Neoplasm Grading , Precancerous Conditions/pathology , Propensity Score , Retrospective Studies , Stomach Neoplasms/pathologyABSTRACT
BACKGROUND: Radiofrequency ablation (RFA) and microwave ablation (MWA) are the most frequently used thermal ablation methods for the treatment of liver cancer. Liver abscess is a common and severe complication of thermal ablation treatment. OBJECTIVES: The objective of this study was to determine the incidence and risk factors of liver abscess formation after thermal ablation of liver cancer. MATERIALS AND METHODS: The clinical data of 423 patients who underwent 691 thermal ablation procedures for liver cancer were collected in order to retrospectively analyze the basic characteristics, incidence, and risk factors associated with liver abscess formation. Patients with multiple risk factors for liver abscess formation were enrolled in a risk factor group, and patients with no risk factors were enrolled in a control group. The chi-square test and multiple logistic regression analysis were used to analyze the relationship between the occurrence of liver abscesses and potential risk factors. RESULTS: Two hundred and eight patients underwent 385 RFA procedures, and 185 patients underwent 306 MWA procedures. The total incidence of liver abscesses was 1.7%, while the rates in the RFA group (1.8%) and MWA groups (1.6%) were similar (P > 0.05). The rates of liver abscesses in patients who had child-pugh class B and class C cirrhosis (P = 0.0486), biliary tract disease (P = 0.0305), diabetes mellitus (P = 0.0344), and porta hepatis tumors (P = 0.0123) were 4.0%, 6.7%, 6.5%, and 13.0%, respectively. There was a statistically significant difference between these four groups and the control group (all P < 0.05). The incidence of liver abscesses in the combined ablation and percutaneous ethanol injection (PEI) group (P = 0.0026) was significantly lower than that of the ablation group (P < 0.05). CONCLUSIONS: The incidence of liver abscesses after liver cancer thermal ablation is low. Child-Pugh Class B and Class C cirrhosis, biliary tract disease, diabetes mellitus, and porta hepatis tumors are four significant risk factors. Combined ablation and PEI reduces the rate of liver abscesses.