• Barrett’s esophagus detection
• Convolutional neural networks
• Deep learning
• Multistep training

• Humans
• Esophageal Neoplasms/classification
• Deep Learning
• Barrett Esophagus/classification
• Barrett Esophagus/diagnosis
• Barrett Esophagus/pathology
• Adenocarcinoma/classification
• Adenocarcinoma/pathology
• Neural Networks, Computer
• Algorithms

• BEX 0581-16-0 Alexander von Humboldt-Stiftung
• 306166/2014-3 Conselho Nacional de Desenvolvimento Científico e Tecnológico
• 307066/2017-7 Conselho Nacional de Desenvolvimento Científico e Tecnológic
• 2013/07375-0 Fundação de Amparo à Pesquisa do Estado de São Paulo
• 2014/12236- 1 Fundação de Amparo à Pesquisa do Estado de São Paulo
• 2016/19403-6 Fundação de Amparo à Pesquisa do Estado de São Paulo
• 2017/04847-9 Fundação de Amparo à Pesquisa do Estado de São Paulo
• 2019/08605-5 Fundação de Amparo à Pesquisa do Estado de São Paulo
• EP/T021063/1 Engineering and Physical Sciences Research Council

• classification
• deep learning
• esophagogastroduodenoscopy
• location inference
• minimum shooting points

• S0252-21-1001 National IT Industry Promotion Agency (NIPA)

• Deep learning
• Gastric intestinal metaplasia
• Real-time semantic segmentation

• Humans
• Gastroscopy/methods
• Stomach Neoplasms
• Narrow Band Imaging/methods
• Metaplasia/diagnostic imaging
• Precancerous Conditions/pathology

• Diagnosis, Computer-Assisted
• Esophageal Neoplasms/diagnosis
• Humans
• Machine Learning
• Neural Networks, Computer

• 1D-CNN
• ESCC
• Relief feature selection
• Survival risk prediction

The close relationship of medicine with technology and the particular interest in this symbiosis in recent years has led to the development of several computed artificial intelligence (AI) systems aimed at various areas of medicine. A number of studies have demonstrated that those systems allow accurate diagnoses with histological precision, thus facilitating decision-making by clinicians in real time. In the field of gastroenterology, AI has been applied in the diagnosis of pathologies of the entire digestive tract and their attached glands, and are increasingly accepted for the detection of colorectal polyps and confirming their histological classification. Studies have shown high accuracy, sensitivity, and specificity in relation to expert endoscopists, and mainly in relation to those with less experience. Other applications that are increasingly studied and with very promising results are the investigation of dysplasia in patients with Barrett's esophagus and the endoscopic and histological assessment of colon inflammation in patients with ulcerative colitis. In some cases AI is thus better than or at least equal to human abilities. However, additional studies are needed to reinforce the existing data, and mainly to determine the applicability of this technology in other indications. This review summarizes the state of the art of AI in gastroenterological pathology.

• Artificial intelligence
• Computer-aided diagnosis
• Deep learning
• gastrointestinal endoscopy
• Colorectal polyps
• Dysplasia

• Artificial Intelligence
• Barrett Esophagus
• Endoscopy, Gastrointestinal
• Gastroenterology
• Humans

• Digital pathology
• Oesophageal adenocarcinoma
• Prognostic and predictive biomarkers
• Tumour microenvironment

Intestinal metaplasia refers to the replacement of the differentiated and mature normal mucosal epithelium outside the intestinal tract by the intestinal epithelium. This paper briefly describes the etiology and clinical significance of intestinal metaplasia in Barrett’s esophagus. This article summarizes the impact of intestinal metaplasia on the diagnosis, monitoring, and treatment of Barrett’s esophagus according to different guidelines. We also briefly explore the basis for the endoscopic diagnosis of intestinal metaplasia in Barrett’s esophagus. The identification techniques of goblet cells in Barrett’s esophagus are also elucidated by some scholars. Additionally, we further elaborate on the current treatment methods related to Barrett’s esophagus.

• Barrett’s esophagus (BE)
• endoscopy and pathological identification
• intestinal metaplasia (IM)
• monitoring
• treatment

• Adenocarcinoma
• Barrett's esophagus
• Computer-aided diagnosis
• Explainable artificial intelligence
• Machine learning

• artificial intelligence
• diagnosis
• early esophageal neoplasms
• meta-analysis
• systemic review

• Artificial intelligence
• Deep learning
• Endoscopy
• Esophagogastroduodenoscopy
• Quality control

• Barrett’s esophagus (BE)
• Esophageal adenocarcinoma (EAC)
• advanced imaging
• chromoendoscopy

• Colonoscopy
• Early Detection of Cancer
• Humans
• Neural Networks, Computer
• Surgical Instruments

• Barrett's esophagus
• Dysplasia
• Esophageal adenocarcinoma
• Intestinal metaplasia
• Surveillance

• artificial intelligence
• endoscopy
• lower gastrointestinal tract
• quality
• upper gastrointestinal tract

• Convolutional neural network
• Gastric intestinal metaplasia
• Intelligent diagnosis
• Magnifying narrow-band imaging
• Narrow-band imaging

• Humans
• Metaplasia/diagnostic imaging
• Narrow Band Imaging
• Neural Networks, Computer
• Pilot Projects
• Precancerous Conditions/diagnostic imaging
• Prospective Studies
• Retrospective Studies
• Stomach Neoplasms/diagnostic imaging

• Adenocarcinoma
• Barrett's esophagus
• Generative adversarial networks
• Machine learning

• Adenocarcinoma/diagnostic imaging
• Barrett Esophagus/diagnostic imaging
• Endoscopy
• Esophageal Neoplasms/diagnostic imaging
• Humans
• Machine Learning
• Neural Networks, Computer

Upper gastrointestinal (GI) endoscopic image documentation has provided an efficient, low-cost solution to address quality control for endoscopic reporting. The problem is, however, challenging for computer-assisted techniques, because different sites have similar appearances. Additionally, across different patients, site appearance variation may be large and inconsistent. Therefore, according to the British and modified Japanese guidelines, we propose a set of oesophagogastroduodenoscopy (EGD) images to be routinely captured and evaluate its efficiency for deep learning-based classification methods.

A novel EGD image dataset standardising upper GI endoscopy to several steps is established following landmarks proposed in guidelines and annotated by an expert clinician. To demonstrate the discrimination of proposed landmarks that enable the generation of an automated endoscopic report, we train several deep learning-based classification models utilising the well-annotated images.

We report results for a clinical dataset composed of 211 patients (comprising a total of 3704 EGD images) acquired during routine upper GI endoscopic examinations. We find close agreement between predicted labels using our method and the ground truth labelled by human experts. We observe the limitation of current static image classification scheme for EGD image classification.

Our study presents a framework for developing automated EGD reports using deep learning. We demonstrate that our method is feasible to address EGD image classification and can lead towards improved performance and additionally qualitatively demonstrate its performance on our dataset.

• Artificial intelligence
• Deep learning
• Endoscopy
• Gastroenterology

Oesophageal cancer is a common cancer that continues to have a poor survival. This is largely in part due to its late diagnosis and early metastatic spread. Currently, screening is limited to patients with multiple risk factors via a relatively invasive technique. However, there is a large proportion of patients diagnosed with oesophageal cancer who have not been screened. This has warranted the development of new screening techniques that could be implemented more widely and lead to earlier identification and subsequently improvements in survival rates. This article also explores progress in the surveillance of Barrett’s oesophagus, a pre-malignant condition for the development of oesophageal adenocarcinoma. In recent years, advances in early endoscopic intervention have meant that more patients are considered at an earlier stage for potentially curative treatment.

• Barrett Esophagus/diagnosis
• Early Detection of Cancer
• Esophageal Neoplasms/diagnosis
• Esophageal Neoplasms/prevention & control
• Humans
• Risk Factors

• Barrett surveillance
• artificial intelligence
• esophageal cancer
• machine learning

Barrett’s esophagus is the precursor lesion for esophageal adenocarcinoma. Screening and surveillance of Barrett’s esophagus are undertaken with the goal of earlier detection and lowering the mortality from esophageal adenocarcinoma. The widely used technique is standard esophagogastroduodenoscopy with biopsies per the Seattle protocol for screening and surveillance of Barrett’s esophagus. Surveillance intervals vary depending on the degree of dysplasia with endoscopic eradication therapy confined to patients with Barrett’s esophagus and confirmed dysplasia. In this review, we present various novel techniques for screening of Barrett’s esophagus such as unsedated transnasal endoscopy, cytosponge with trefoil factor-3, balloon cytology, esophageal capsule endoscopy, liquid biopsy, electronic nose, and oral microbiome. In addition, advanced imaging techniques such as narrow band imaging, dye-based chromoendoscopy, confocal laser endomicroscopy, volumetric laser endomicroscopy, and wide-area transepithelial sampling with computer-assisted three-dimensional analysis developed for better detection of dysplasia are also reviewed.

• Barrett’s esophagus
• cytosponge
• dysplasia
• esophageal adenocarcinoma
• gastroesophageal reflux disease
• screening
• surveillance
• trefoil factor
• unsedated transnasal endoscopy

• Barret’s esophagus
• confocal laser endomicroscopy
• enhancement
• fractional differentiation and integration
• grade classification

This study aims to adapt and evaluate the performance of different state-of-the-art deep learning object detection methods to automatically identify esophageal adenocarcinoma (EAC) regions from high-definition white light endoscopy (HD-WLE) images.

Several state-of-the-art object detection methods using Convolutional Neural Networks (CNNs) were adapted to automatically detect abnormal regions in the esophagus HD-WLE images, utilizing VGG’16 as the backbone architecture for feature extraction. Those methods are Regional-based Convolutional Neural Network (R-CNN), Fast R-CNN, Faster R-CNN and Single-Shot Multibox Detector (SSD). For the evaluation of the different methods, 100 images from 39 patients that have been manually annotated by five experienced clinicians as ground truth have been tested.

Experimental results illustrate that the SSD and Faster R-CNN networks show promising results, and the SSD outperforms other methods achieving a sensitivity of 0.96, specificity of 0.92 and F-measure of 0.94. Additionally, the Average Recall Rate of the Faster R-CNN in locating the EAC region accurately is 0.83.

In this paper, recent deep learning object detection methods are adapted to detect esophageal abnormalities automatically. The evaluation of the methods proved its ability to locate abnormal regions in the esophagus from endoscopic images. The automatic detection is a crucial step that may help early detection and treatment of EAC and also can improve automatic tumor segmentation to monitor its growth and treatment outcome.

• Barrett’s esophagus
• Deep learning
• Esophageal adenocarcinoma detection
• HD-WLE