Artificial intelligence (AI) using deep-learning (DL) has emerged as a breakthrough computer technology. By the era of big data, the accumulation of an enormous number of digital images and medical records drove the need for the utilization of AI to efficiently deal with these data, which have become fundamental resources for a machine to learn by itself. Among several DL models, the convolutional neural network showed outstanding performance in image analysis. In the field of gastroenterology, physicians handle large amounts of clinical data and various kinds of image devices such as endoscopy and ultrasound. AI has been applied in gastroenterology in terms of diagnosis, prognosis, and image analysis. However, potential inherent selection bias cannot be excluded in the form of retrospective study. Because overfitting and spectrum bias (class imbalance) have the possibility of overestimating the accuracy, external validation using unused datasets for model development, collected in a way that minimizes the spectrum bias, is mandatory. For robust verification, prospective studies with adequate inclusion/exclusion criteria, which represent the target populations, are needed. DL has its own lack of interpretability. Because interpretability is important in that it can provide safety measures, help to detect bias, and create social acceptance, further investigations should be performed.
Background Visual inspection, lesion detection, and differentiation between malignant and benign features are key aspects of an endoscopist’s role. The use of machine learning for the recognition and differentiation of images has been increasingly adopted in clinical practice. This study aimed to establish convolutional neural network (CNN) models to automatically classify gastric neoplasms based on endoscopic images. Methods Endoscopic white-light images of pathologically confirmed gastric lesions were collected and classified into five categories: advanced gastric cancer, early gastric cancer, high grade dysplasia, low grade dysplasia, and non-neoplasm. Three pretrained CNN models were fine-tuned using a training dataset. The classifying performance of the models was evaluated using a test dataset and a prospective validation dataset. Results A total of 5017 images were collected from 1269 patients, among which 812 images from 212 patients were used as the test dataset. An additional 200 images from 200 patients were collected and used for prospective validation. For the five-category classification, the weighted average accuracy of the Inception-Resnet-v2 model reached 84.6 %. The mean area under the curve (AUC) of the model for differentiating gastric cancer and neoplasm was 0.877 and 0.927, respectively. In prospective validation, the Inception-Resnet-v2 model showed lower performance compared with the endoscopist with the best performance (five-category accuracy 76.4 % vs. 87.6 %; cancer 76.0 % vs. 97.5 %; neoplasm 73.5 % vs. 96.5 %; P < 0.001). However, there was no statistical difference between the Inception-Resnet-v2 model and the endoscopist with the worst performance in the differentiation of gastric cancer (accuracy 76.0 % vs. 82.0 %) and neoplasm (AUC 0.776 vs. 0.865). Conclusion The evaluated deep-learning models have the potential for clinical application in classifying gastric cancer or neoplasm on endoscopic white-light images.
Endoscopic resection is recommended for gastric neoplasms confined to mucosa or superficial submucosa. The determination of invasion depth is based on gross morphology assessed in endoscopic images, or on endoscopic ultrasound. These methods have limited accuracy and pose an inter-observer variability. Several studies developed deep-learning (DL) algorithms classifying invasion depth of gastric cancers. Nevertheless, these algorithms are intended to be used after definite diagnosis of gastric cancers, which is not always feasible in various gastric neoplasms. This study aimed to establish a DL algorithm for accurately predicting submucosal invasion in endoscopic images of gastric neoplasms. Pre-trained convolutional neural network models were fine-tuned with 2899 white-light endoscopic images. The prediction models were subsequently validated with an external dataset of 206 images. In the internal test, the mean area under the curve discriminating submucosal invasion was 0.887 (95% confidence interval: 0.849–0.924) by DenseNet−161 network. In the external test, the mean area under the curve reached 0.887 (0.863–0.910). Clinical simulation showed that 6.7% of patients who underwent gastrectomy in the external test were accurately qualified by the established algorithm for potential endoscopic resection, avoiding unnecessary operation. The established DL algorithm proves useful for the prediction of submucosal invasion in endoscopic images of gastric neoplasms.
Background/AimsTo assess the therapeutic outcomes of endoscopic resection (ER) of early gastric cancer (EGC) with undifferentiated-type histology. MethodsCases of ER of EGC with undifferentiated-type histology in the Korean endoscopic submucosal dissection (ESD) registry database were identified and reviewed. The immediate outcomes, including en bloc resection, complete resection, and curative resection rates, and long-term outcomes, including recurrence and survival rates, were extracted and analyzed. ResultsFrom 2006 to 2015, 275 EGCs with undifferentiated-type histology from 275 patients were identified. The immediate outcomes were as follows: en bloc resection rate: 92.4%; complete resection rate: 80%; and curative resection rate: 36.4%. Compared to patients with lesions that were beyond the expanded indication, those with expanded indication lesions showed better therapeutic outcomes. There was no difference in immediate outcomes between patients with poorly differentiated adenocarcinoma (PDC) and signet ring cell carcinoma (SRC). However, compared to ER of SRC, ER of PDC had a stronger association with submucosal invasion (41.9% vs. 23.6%, p=0.003). With regard to long-term outcomes, there was no difference between lesions with curative and non-curative resections in the recurrence and mortality rates. These rates also did not differ between PDC and SRC (median follow up: 3.96 years). ConclusionsER confined to expanded indication lesions can be considered for treatment of EGC with undifferentiated-type histology.
Serum pepsinogen assay (sPGA), which reveals serum pepsinogen (PG) I concentration and the PG I/PG II ratio, is a non-invasive test for predicting chronic atrophic gastritis (CAG) and gastric neoplasms. Although various cut-off values have been suggested, PG I ≤70 ng/mL and a PG I/PG II ratio of ≤3 have been proposed. However, previous meta-analyses reported insufficient systematic reviews and only pooled outcomes, which cannot determine the diagnostic validity of sPGA with a cut-off value of PG I ≤70 ng/mL and/or PG I/PG II ratio ≤3. We searched the core databases (MEDLINE, Cochrane Library, and Embase) from their inception to April 2018. Fourteen and 43 studies were identified and analyzed for the diagnostic performance in CAG and gastric neoplasms, respectively. Values for sensitivity, specificity, diagnostic odds ratio, and area under the curve with a cut-off value of PG I ≤70 ng/mL and PG I/PG II ratio ≤3 to diagnose CAG were 0.59, 0.89, 12, and 0.81, respectively and for diagnosis of gastric cancer (GC) these values were 0.59, 0.73, 4, and 0.7, respectively. Methodological quality and ethnicity of enrolled studies were found to be the reason for the heterogeneity in CAG diagnosis. Considering the high specificity, non-invasiveness, and easily interpretable characteristics, sPGA has potential for screening of CAG or GC.
BackgroundControversies persist regarding the effect of prokinetics for the treatment of functional dyspepsia (FD). This study aimed to assess the comparative efficacy of prokinetic agents for the treatment of FD.MethodsRandomized controlled trials (RCTs) of prokinetics for the treatment of FD were identified from core databases. Symptom response rates were extracted and analyzed using odds ratios (ORs). A Bayesian network meta-analysis was performed using the Markov chain Monte Carlo method in WinBUGS and NetMetaXL.ResultsIn total, 25 RCTs, which included 4473 patients with FD who were treated with 6 different prokinetics or placebo, were identified and analyzed. Metoclopramide showed the best surface under the cumulative ranking curve (SUCRA) probability (92.5%), followed by trimebutine (74.5%) and mosapride (63.3%). However, the therapeutic efficacy of metoclopramide was not significantly different from that of trimebutine (OR:1.32, 95% credible interval: 0.27–6.06), mosapride (OR: 1.99, 95% credible interval: 0.87–4.72), or domperidone (OR: 2.04, 95% credible interval: 0.92–4.60). Metoclopramide showed better efficacy than itopride (OR: 2.79, 95% credible interval: 1.29–6.21) and acotiamide (OR: 3.07, 95% credible interval: 1.43–6.75). Domperidone (SUCRA probability 62.9%) showed better efficacy than itopride (OR: 1.37, 95% credible interval: 1.07–1.77) and acotiamide (OR: 1.51, 95% credible interval: 1.04–2.18).ConclusionsMetoclopramide, trimebutine, mosapride, and domperidone showed better efficacy for the treatment of FD than itopride or acotiamide. Considering the adverse events related to metoclopramide or domperidone, the short-term use of these agents or the alternative use of trimebutine or mosapride could be recommended for the symptomatic relief of FD.Electronic supplementary materialThe online version of this article (doi:10.1186/s12876-017-0639-0) contains supplementary material, which is available to authorized users.
BackgroundKorean Red Ginseng (KRG) is a well-known natural product with anticarcinogenic and antioxidant effects. We evaluated the antifatigue effect of KRG in patients with nonalcoholic fatty liver disease (NAFLD).MethodsEighty patients with NAFLD were prospectively randomized to receive 3 wk of KRG or placebo in addition to counseling on healthy eating and regular exercise. Liver function test, proinflammatory cytokines, adiponectin, antioxidant activity, and fatigue score were measured and compared according to the body mass index between the KRG and placebo groups.ResultsThe liver function tests were significantly improved after 3 wk of treatment in both groups. The mean levels (at baseline and after treatment) of tumor necrosis factor-α were 108.0 pg/mL ± 54.8 pg/mL and 92.7 pg/mL ± 39.0 pg/mL (p = 0.018) in the KRG group and 123.1 pg/mL ± 42.1 pg/mL and 127.5 pg/mL ± 62.2 pg/mL (p = 0.694) in the placebo group, respectively. There was a significant difference in change of adiponectin levels between the KRG (7,751.2 pg/mL ± 3,108.1 pg/mL and 8,197.3 pg/mL ± 2,714.5 pg/mL) and placebo groups (7,711.6 pg/mL ± 3,041.3 pg/mL and 7,286.1 pg/mL ± 5,188.7 pg/mL, p = 0.027). In patients with overweight, the fatigue score was significantly decreased in the KRG group (35.0 ± 13.2 and 24.5 ± 8.9, p = 0.019).ConclusionOur results show that KRG might be effective in reducing proinflammatory cytokine and fatigue in overweight patients with NAFLD, in addition to improvements in adiponectin levels.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.