Artificial intelligence in the diagnosis of gastric precancerous conditions by image-enhanced endoscopy: a multicenter, diagnostic study (with video)

Convolutional Neural Network Model for Intestinal Metaplasia Recognition in Gastric Corpus Using Endoscopic Image Patches

Gastric cancer (GC) is a significant healthcare concern, and the identification of high-risk patients is crucial. Indeed, gastric precancerous conditions present significant diagnostic challenges, particularly early intestinal metaplasia (IM) detection. This study developed a deep learning system to assist in IM detection using image patches from gastric corpus examined using virtual chromoendoscopy in a Western country. Utilizing a retrospective dataset of endoscopic images from Sant'Andrea University Hospital of Rome, collected between January 2020 and December 2023, the system extracted 200 × 200 pixel patches, classifying them with a voting scheme. The specificity and sensitivity on the patch test set were 76% and 72%, respectively. The optimization of a learnable voting scheme on a validation set achieved a specificity of 70% and sensitivity of 100% for entire images. Despite data limitations and the absence of pre-trained models, the system shows promising results for preliminary screening in gastric precancerous condition diagnostics, providing an explainable and robust Artificial Intelligence approach.

A Swin transformer encoder-based StyleGAN for unbalanced endoscopic image enhancement

With the rapid development of artificial intelligence, automated endoscopy-assisted diagnostic systems have become an effective tool for reducing the diagnostic costs and shortening the treatment cycle of patients. Typically, the performance of these systems depends on deep learning models which are pre-trained with large-scale labeled data, for example, early gastric cancer based on endoscopic images. However, the expensive annotation and the subjectivity of the annotators lead to an insufficient and class-imbalanced endoscopic image dataset, and these datasets are detrimental to the training of deep learning models. Therefore, we proposed a Swin Transformer encoder-based StyleGAN (STE-StyleGAN) for unbalanced endoscopic image enhancement, which is composed of an adversarial learning encoder and generator. Firstly, a pre-trained Swin Transformer is introduced into the encoder to extract multi-scale features layer by layer from endoscopic images. The features are subsequently fed into a mapping block for aggregation and recombination. Secondly, a self-attention mechanism is applied to the generator, which adds detailed information of the image layer by layer through recoded features, enabling the generator to autonomously learn the coupling between different image regions. Finally, we conducted extensive experiments on a private intestinal metaplasia grading dataset from a Grade-A tertiary hospital. The experimental results show that the images generated by STE-StyleGAN are closer to the initial image distribution, achieving a Fréchet Inception Distance (FID) value of 100.4. Then, these generated images are used to enhance the initial dataset to improve the robustness of the classification model, and achieved a top accuracy of 86 %.

Accuracy of artificial intelligence-assisted endoscopy in the diagnosis of gastric intestinal metaplasia: A systematic review and meta-analysis

BACKGROUND AND AIMS

Gastric intestinal metaplasia is a precancerous disease, and a timely diagnosis is essential to delay or halt cancer progression. Artificial intelligence (AI) has found widespread application in the field of disease diagnosis. This study aimed to conduct a comprehensive evaluation of AI's diagnostic accuracy in detecting gastric intestinal metaplasia in endoscopy, compare it to endoscopists' ability, and explore the main factors affecting AI's performance.

METHODS

The study followed the PRISMA-DTA guidelines, and the PubMed, Embase, Web of Science, Cochrane, and IEEE Xplore databases were searched to include relevant studies published by October 2023. We extracted the key features and experimental data of each study and combined the sensitivity and specificity metrics by meta-analysis. We then compared the diagnostic ability of the AI versus the endoscopists using the same test data.

RESULTS

Twelve studies with 11,173 patients were included, demonstrating AI models' efficacy in diagnosing gastric intestinal metaplasia. The meta-analysis yielded a pooled sensitivity of 94% (95% confidence interval: 0.92-0.96) and specificity of 93% (95% confidence interval: 0.89-0.95). The combined area under the receiver operating characteristics curve was 0.97. The results of meta-regression and subgroup analysis showed that factors such as study design, endoscopy type, number of training images, and algorithm had a significant effect on the diagnostic performance of AI. The AI exhibited a higher diagnostic capacity than endoscopists (sensitivity: 95% vs. 79%).

CONCLUSIONS

AI-aided diagnosis of gastric intestinal metaplasia using endoscopy showed high performance and clinical diagnostic value. However, further prospective studies are required to validate these findings.

A deep learning model based on magnifying endoscopy with narrow-band imaging to evaluate intestinal metaplasia grading and OLGIM staging: A multicenter study

BACKGROUND AND PURPOSE

Patients with stage III or IV of operative link for gastric intestinal metaplasia assessment (OLGIM) are at a higher risk of gastric cancer (GC). We aimed to construct a deep learning (DL) model based on magnifying endoscopy with narrow-band imaging (ME-NBI) to evaluate OLGIM staging.

METHODS

This study included 4473 ME-NBI images obtained from 803 patients at three endoscopy centres. The endoscopic expert marked intestinal metaplasia (IM) regions on endoscopic images of the target biopsy sites. Faster Region-Convolutional Neural Network model was used to grade IM lesions and predict OLGIM staging.

RESULTS

The diagnostic performance of the model for IM grading in internal and external validation sets, as measured by the area under the curve (AUC), was 0.872 and 0.803, respectively. The accuracy of this model in predicting the high-risk stage of OLGIM was 84.0%, which was not statistically different from that of three junior (71.3%, p = 0.148) and three senior endoscopists (75.3%, p = 0.317) specially trained in endoscopic images corresponding to pathological IM grade, but higher than that of three untrained junior endoscopists (64.0%, p = 0.023).

CONCLUSION

This DL model can assist endoscopists in predicting OLGIM staging using ME-NBI without biopsy, thereby facilitating screening high-risk patients for GC.

Combined gastric and colorectal cancer endoscopic screening may be cost-effective in Europe with the implementation of artificial intelligence: an economic evaluation

BACKGROUND/AIMS

Endoscopic screening for gastric cancer (GC) is not recommended in low-intermediate incidence countries. Artificial intelligence (AI) has high accuracy in GC detection and might increase the cost-effectiveness of screening strategies. We aimed to assess the cost-effectiveness of AI for GC detection in settings with different GC incidence and different accuracies of AI systems.

METHODS

Cost-effectiveness analysis (using Markov model) comparing different screening strategies (no screening versus single esophagogastroduodenoscopy (EGD) at 50 years versus stand-alone EGD every 5/10 years versus combined EGD and screening colonoscopy once or twice per decade in Netherlands, Italy and Portugal) with variable AI accuracy settings. The primary outcome was the incremental cost-effectiveness ratio of the different strategies versus no screening. Deterministic and probabilistic sensitivity analyses were conducted.

RESULTS

Without AI, one single EGD at 50 years (Netherlands, Italy, Portugal), EGD combined with screening colonoscopy once per decade (Italy and Portugal) and EGD combined with screening colonoscopy twice per decade (Portugal) are cost-effective when compared with no screening. If AI increases the accuracy of EGD by at least 1% in comparison to the accuracy of white-light endoscopy accuracy (89%), combined screening twice per decade also becomes cost-effective in Italy. If AI accuracy reaches at least 96%, combined screening once per decade is also cost-effective in the Netherlands.

DISCUSSION

In European countries, AI-assisted EGD may improve the cost-effectiveness of GC screening with combined EGD and screening colonoscopy. The actual effect of AI on cost-effectiveness may vary dependent on the accuracy and costs of the AI system.

An artificial intelligence system for chronic atrophic gastritis diagnosis and risk stratification under white light endoscopy

BACKGROUND AND AIMS

The diagnosis and stratification of gastric atrophy (GA) predict patients' gastric cancer progression risk and determine endoscopy surveillance interval. We aimed to construct an artificial intelligence (AI) system for GA endoscopic identification and risk stratification based on the Kimura-Takemoto classification.

METHODS

We constructed the system using two trained models and verified its performance. First, we retrospectively collected 869 images and 119 videos to compare its performance with that of endoscopists in identifying GA. Then, we included original image cases of 102 patients to validate the system for stratifying GA and comparing it with endoscopists with different experiences.

RESULTS

The sensitivity of model 1 was higher than that of endoscopists (92.72% vs. 76.85 %) at image level and also higher than that of experts (94.87% vs. 85.90 %) at video level. The system outperformed experts in stratifying GA (overall accuracy: 81.37 %, 73.04 %, p = 0.045). The accuracy of this system in classifying non-GA, mild GA, moderate GA, and severe GA was 80.00 %, 77.42 %, 83.33 %, and 85.71 %, comparable to that of experts and better than that of seniors and novices.

CONCLUSIONS

We established an expert-level system for GA endoscopic identification and risk stratification. It has great potential for endoscopic assessment and surveillance determinations.

Deep Learning and Gastric Cancer: Systematic Review of AI-Assisted Endoscopy

BACKGROUND

Gastric cancer (GC), a significant health burden worldwide, is typically diagnosed in the advanced stages due to its non-specific symptoms and complex morphological features. Deep learning (DL) has shown potential for improving and standardizing early GC detection. This systematic review aims to evaluate the current status of DL in pre-malignant, early-stage, and gastric neoplasia analysis.

METHODS

A comprehensive literature search was conducted in PubMed/MEDLINE for original studies implementing DL algorithms for gastric neoplasia detection using endoscopic images. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The focus was on studies providing quantitative diagnostic performance measures and those comparing AI performance with human endoscopists.

RESULTS

Our review encompasses 42 studies that utilize a variety of DL techniques. The findings demonstrate the utility of DL in GC classification, detection, tumor invasion depth assessment, cancer margin delineation, lesion segmentation, and detection of early-stage and pre-malignant lesions. Notably, DL models frequently matched or outperformed human endoscopists in diagnostic accuracy. However, heterogeneity in DL algorithms, imaging techniques, and study designs precluded a definitive conclusion about the best algorithmic approach.

CONCLUSIONS

The promise of artificial intelligence in improving and standardizing gastric neoplasia detection, diagnosis, and segmentation is significant. This review is limited by predominantly single-center studies and undisclosed datasets used in AI training, impacting generalizability and demographic representation. Further, retrospective algorithm training may not reflect actual clinical performance, and a lack of model details hinders replication efforts. More research is needed to substantiate these findings, including larger-scale multi-center studies, prospective clinical trials, and comprehensive technical reporting of DL algorithms and datasets, particularly regarding the heterogeneity in DL algorithms and study designs.

Real-time detection of laryngopharyngeal cancer using an artificial intelligence-assisted system with multimodal data

BACKGROUND

Laryngopharyngeal cancer (LPC) includes laryngeal and hypopharyngeal cancer, whose early diagnosis can significantly improve the prognosis and quality of life of patients. Pathological biopsy of suspicious cancerous tissue under the guidance of laryngoscopy is the gold standard for diagnosing LPC. However, this subjective examination largely depends on the skills and experience of laryngologists, which increases the possibility of missed diagnoses and repeated unnecessary biopsies. We aimed to develop and validate a deep convolutional neural network-based Laryngopharyngeal Artificial Intelligence Diagnostic System (LPAIDS) for real-time automatically identifying LPC in both laryngoscopy white-light imaging (WLI) and narrow-band imaging (NBI) images to improve the diagnostic accuracy of LPC by reducing diagnostic variation among on-expert laryngologists.

METHODS

All 31,543 laryngoscopic images from 2382 patients were categorised into training, verification, and test sets to develop, validate, and internal test LPAIDS. Another 25,063 images from five other hospitals were used as external tests. Overall, 551 videos were used to evaluate the real-time performance of the system, and 200 randomly selected videos were used to compare the diagnostic performance of the LPAIDS with that of laryngologists. Two deep-learning models using either WLI (model W) or NBI (model N) images were constructed to compare with LPAIDS.

RESULTS

LPAIDS had a higher diagnostic performance than models W and N, with accuracies of 0·956 and 0·949 in the internal image and video tests, respectively. The robustness and stability of LPAIDS were validated in external sets with the area under the receiver operating characteristic curve values of 0·965-0·987. In the laryngologist-machine competition, LPAIDS achieved an accuracy of 0·940, which was comparable to expert laryngologists and outperformed other laryngologists with varying qualifications.

CONCLUSIONS

LPAIDS provided high accuracy and stability in detecting LPC in real-time, which showed great potential for using LPAIDS to improve the diagnostic accuracy of LPC by reducing diagnostic variation among on-expert laryngologists.

Color-guided deformable convolution network for intestinal metaplasia severity classification using endoscopic images

Objective. Intestinal metaplasia (IM) is a common precancerous condition for gastric cancer, and the risk of developing gastric cancer increases as IM worsens. However, current deep learning-based methods cannot effectively model the complex geometric structure of IM lesions. To accurately diagnose the severity of IM and prevent the occurrence of gastric cancer, we revisit the deformable convolution network (DCN), propose a novel offset generation method based on color features to guide deformable convolution, named color-guided deformable convolutional network (CDCN).Approach. Specifically, we propose a combined strategy of conventional and deep learning methods for IM lesion areas localization and generating offsets. Under the guidance of offsets, the sample locations of convolutional neural network adaptively adjust to extract discriminate features in an irregular way that conforms to the IM shape.Main results. To verify the effectiveness of CDCN, comprehensive experiments are conducted on the self-constructed IM severity dataset. The experimental results show that CDCN outperforms many existing methods and the accuracy has been improved by 5.39% compared to DCN, reaching 84.17%. Significance. To the best of our knowledge, CDCN is the first method to grade the IM severity using endoscopic images, which can significantly enhance the clinical application of endoscopy, achieving more precise diagnoses.

Gastric Intestinal Metaplasia: Challenges and the Opportunity for Precision Prevention

GIM is a persistent, premalignant lesion whereby gastric mucosa is replaced by metaplastic mucosa resembling intestinal tissue, arising in the setting of chronic inflammation, particularly in the context of Helicobacter pylori. While the overall rates of progression to gastric adenocarcinoma are low, estimated at from 0.25 to 2.5%, there are features that confer a much higher risk and warrant follow-up. In this review, we collate and summarise the current knowledge regarding the pathogenesis of GIM, and the clinical, endoscopic and histologic risk factors for cancer. We examine the current state-of-practice with regard to the diagnosis and management of GIM, which varies widely in the published guidelines and in practice. We consider the emerging evidence in population studies, artificial intelligence and molecular markers, which will guide future models of care. The ultimate goal is to increase the detection of early gastric dysplasia/neoplasia that can be cured while avoiding unnecessary surveillance in very low-risk individuals.

Application of artificial intelligence for improving early detection and prediction of therapeutic outcomes for gastric cancer in the era of precision oncology

Gastric cancer is a leading contributor to cancer incidence and mortality globally. Recently, artificial intelligence approaches, particularly machine learning and deep learning, are rapidly reshaping the full spectrum of clinical management for gastric cancer. Machine learning is formed from computers running repeated iterative models for progressively improving performance on a particular task. Deep learning is a subtype of machine learning on the basis of multilayered neural networks inspired by the human brain. This review summarizes the application of artificial intelligence algorithms to multi-dimensional data including clinical and follow-up information, conventional images (endoscope, histopathology, and computed tomography (CT)), molecular biomarkers, etc. to improve the risk surveillance of gastric cancer with established risk factors; the accuracy of diagnosis, and survival prediction among established gastric cancer patients; and the prediction of treatment outcomes for assisting clinical decision making. Therefore, artificial intelligence makes a profound impact on almost all aspects of gastric cancer from improving diagnosis to precision medicine. Despite this, most established artificial intelligence-based models are in a research-based format and often have limited value in real-world clinical practice. With the increasing adoption of artificial intelligence in clinical use, we anticipate the arrival of artificial intelligence-powered gastric cancer care.

Diagnostic Performance of Deep Learning Models for Gastric Intestinal Metaplasia Detection in Narrow-band Images

Gastric Intestinal Metaplasia (GIM) is one of the precancerous conditions in the gastric carcinogenesis cascade and its optical diagnosis during endoscopic screening is challenging even for seasoned endoscopists. Several solutions leveraging pre-trained deep neural networks (DNNs) have been recently proposed in order to assist human diagnosis. In this paper, we present a comparative study of these architectures in a new dataset containing GIM and non-GIM Narrow-band imaging still frames. We find that the surveyed DNNs perform remarkably well on average, but still measure sizeable inter-fold variability during cross-validation. An additional ad-hoc analysis suggests that these baseline architectures may not perform equally well at all scales when diagnosing GIM.Clinical relevance- Enhanching a clinician's ability to detect and localize intestinal metaplasia can be a crucial tool for gastric cancer management policies.

Adequacy of endoscopic recognition and surveillance of gastric intestinal metaplasia and atrophic gastritis: A multicentre retrospective study in low incidence countries

BACKGROUND

Gastric atrophy (GA) and gastric intestinal metaplasia (GIM) are precursor conditions to gastric adenocarcinoma (GAC) and should be monitored endoscopically in selected individuals. However, little is known about adherence to recommendations in clinical practice in low-risk countries.

OBJECTIVE

The aim of this study was to evaluate endoscopic recognition and adequacy of surveillance for GA and GIM in countries with low GAC prevalence.

METHODS

We retrospectively analysed patients diagnosed with GIM or GA in three centers in The Netherlands and UK between 2012 and 2019. Cases with GIM and/or GA diagnosis at index endoscopy were retrieved through systematic search of pathology databases using 'gastric' and 'intestinal metaplasia' or 'atrophy' keywords. Endoscopy reports were analysed to ascertain accuracy of endoscopic diagnoses. Adequacy of surveillance was assessed following histological diagnosis at the index endoscopy based on ESGE guidelines published in 2012.

RESULTS

We included 396 patients with a median follow-up of 57.2 months. Mean age was 66 years and the rates of antrum-predominant versus extensive GIM were comparable (37% vs 38%). Endoscopic recognition rates were 48.5% for GA and 16.3% for GIM. Surveillance was adequately carried out in 215 of 396 patients (54.3%).

CONCLUSION

In countries with a low incidence of GAC, the rate of endoscopic recognition of gastric pre-cancerous lesions and adherence to surveillance recommendation are low. Substantial improvement is required in endoscopic training and awareness of guidelines recommendation in order to optimise detection and management of pre-malignant gastric conditions.

Real-time gastric intestinal metaplasia diagnosis tailored for bias and noisy-labeled data with multiple endoscopic imaging

This work presents real-time segmentation viz. gastric intestinal metaplasia (GIM). Recently, GIM segmentation of endoscopic images has been carried out to differentiate GIM from a healthy stomach. However, real-time detection is difficult to achieve. Conditions are challenging, and include multiple color modes (white light endoscopy and narrow-band imaging), other abnormal lesions (erosion and ulcer), noisy labels etc. Herein, our model is based on BiSeNet and can overcome the many issues regarding GIM. Application of auxiliary head and additional loss are seen to improve performance as well as enhance multiple color modes accurately. Further, multiple pre-processing techniques are utilized for leveraging detection performance: namely, location-wise negative sampling, jigsaw augmentation, and label smoothing. Finally, the decision threshold can be adjusted separately for each color mode. Work undertaken at King Chulalongkorn Memorial Hospital examined 940 histologically proven GIM images and 1239 non-GIM images, obtained over 173 frames per second (FPS). In terms of accuracy, our model is seen to outperform all baselines. Our results demonstrate sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union (IoU), achieving GIM segmentation values of 91%, 96%, 91%, 91%, 96%, and 55%, respectively.

Diagnosis and segmentation effect of the ME-NBI-based deep learning model on gastric neoplasms in patients with suspected superficial lesions - a multicenter study

Background

Endoscopically visible gastric neoplastic lesions (GNLs), including early gastric cancer and intraepithelial neoplasia, should be accurately diagnosed and promptly treated. However, a high rate of missed diagnosis of GNLs contributes to the potential risk of the progression of gastric cancer. The aim of this study was to develop a deep learning-based computer-aided diagnosis (CAD) system for the diagnosis and segmentation of GNLs under magnifying endoscopy with narrow-band imaging (ME-NBI) in patients with suspected superficial lesions.

Methods

ME-NBI images of patients with GNLs in two centers were retrospectively analysed. Two convolutional neural network (CNN) modules were developed and trained on these images. CNN1 was trained to diagnose GNLs, and CNN2 was trained for segmentation. An additional internal test set and an external test set from another center were used to evaluate the diagnosis and segmentation performance.

Results

CNN1 showed a diagnostic performance with an accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 90.8%, 92.5%, 89.0%, 89.4% and 92.2%, respectively, and an area under the curve (AUC) of 0.928 in the internal test set. With CNN1 assistance, all endoscopists had a higher accuracy than for an independent diagnosis. The average intersection over union (IOU) between CNN2 and the ground truth was 0.5837, with a precision, recall and the Dice coefficient of 0.776, 0.983 and 0.867, respectively.

Conclusions

This CAD system can be used as an auxiliary tool to diagnose and segment GNLs, assisting endoscopists in more accurately diagnosing GNLs and delineating their extent to improve the positive rate of lesion biopsy and ensure the integrity of endoscopic resection.

Diagnostic value of artificial intelligence-assisted endoscopy for chronic atrophic gastritis: a systematic review and meta-analysis

Background and aims

The diagnosis of chronic atrophic gastritis (CAG) under normal white-light endoscopy depends on the endoscopist's experience and is not ideal. Artificial intelligence (AI) is increasingly used to diagnose diseases with good results. This review aimed to evaluate the accuracy of AI-assisted diagnosis of CAG through a meta-analysis.

Methods

We conducted a comprehensive literature search of four databases: PubMed, Embase, Web of Science, and the Cochrane Library. Studies published by November 21, 2022, on AI diagnosis CAG with endoscopic images or videos were included. We assessed the diagnostic performance of AI using meta-analysis, explored the sources of heterogeneity through subgroup analysis and meta-regression, and compared the accuracy of AI and endoscopists in diagnosing CAG.

Results

Eight studies that included a total of 25,216 patients of interest, 84,678 image training set images, and 10,937 test set images/videos were included. The results of the meta-analysis showed that the sensitivity of AI in identifying CAG was 94% (95% confidence interval [CI]: 0.88-0.97, I² = 96.2%), the specificity was 96% (95% CI: 0.88-0.98, I² = 98.04%), and the area under the summary receiver operating characteristic curve was 0.98 (95% CI: 0.96-0.99). The accuracy of AI in diagnosing CAG was significantly higher than that of endoscopists.

Conclusions

AI-assisted diagnosis of CAG in endoscopy has high accuracy and clinical diagnostic value.

Systematic review registration

http://www.crd.york.ac.uk/PROSPERO/, identifier: CRD42023391853.

Current Status of Artificial Intelligence-Based Computer-Assisted Diagnosis Systems for Gastric Cancer in Endoscopy

Artificial intelligence (AI) is gradually being utilized in various fields as its performance has been improving with the development of deep learning methods, availability of big data, and the progression of computer processing units. In the field of medicine, AI is mainly implemented in image recognition, such as in radiographic and pathologic diagnoses. In the realm of gastrointestinal endoscopy, although AI-based computer-assisted detection/diagnosis (CAD) systems have been applied in some areas, such as colorectal polyp detection and diagnosis, so far, their implementation in real-world clinical settings is limited. The accurate detection or diagnosis of gastric cancer (GC) is one of the challenges in which performance varies greatly depending on the endoscopist's skill. The diagnosis of early GC is especially challenging, partly because early GC mimics atrophic gastritis in the background mucosa. Therefore, several CAD systems for GC are being actively developed. The development of a CAD system for GC is considered challenging because it requires a large number of GC images. In particular, early stage GC images are rarely available, partly because it is difficult to diagnose gastric cancer during the early stages. Additionally, the training image data should be of a sufficiently high quality to conduct proper CAD training. Recently, several AI systems for GC that exhibit a robust performance, owing to being trained on a large number of high-quality images, have been reported. This review outlines the current status and prospects of AI use in esophagogastroduodenoscopy (EGDS), focusing on the diagnosis of GC.

Application of artificial intelligence in the endoscopic diagnosis of early gastric cancer, atrophic gastritis, and Helicobacter pylori infection

Gastric cancer (GC) is one of the most serious health problems worldwide. Chronic atrophic gastritis (CAG) is most commonly caused by Helicobacter pylori (H. pylori) infection. Currently, endoscopic detection of early gastric cancer (EGC) and CAG remains challenging for endoscopists, and the diagnostic accuracy of H. pylori infection by endoscopy is approximately 70%. Artificial intelligence (AI) can assist endoscopic diagnosis including detection, prediction of depth of invasion, boundary delineation, and anatomical location of EGC, and has achievable diagnostic ability even comparable to experienced endoscopists. In this review we summarized various AI-assisted systems in the diagnosis of EGC, CAG, and H. pylori infection.

Endoscopic Grading of Gastric Intestinal Metaplasia Using Magnifying and Nonmagnifying Narrow-Band Imaging Endoscopy

Several endoscopic findings obtained by magnifying image-enhanced endoscopy (IEE) are reportedly correlated with gastric intestinal metaplasia (IM); however, the differences between magnifying and nonmagnifying IEE for the diagnosis of gastric IM remain unknown. This study included 100 consecutive patients who underwent narrow-band imaging endoscopy. Four areas of the stomach were evaluated using nonmagnifying and magnifying IEE. Light-blue crest (LBC), white opaque substance (WOS), and endoscopic grading of the gastric IM (EGGIM) were assessed. The concordance rates between nonmagnifying and magnifying IEE were 80.5% for LBC and 93.3% for WOS. The strength of agreement between each observation technique showed good reproducibility, with a kappa value of 0.69 and 0.83 for LBC and WOS, respectively. The individual EGGIM score indicated a good correlation between nonmagnifying and magnifying IEE (concordance rate, 75%; kappa value, 0.67). The prevalence of a high EGGIM score in patients with and without gastric cancer (GC) showed a significant difference both with nonmagnifying IEE (odds ratio (OR), 3.3; 95% confidence interval (CI), 1.2-9.0), and magnifying IEE (OR, 3.1; 95% CI, 1.1-8.9). Nonmagnifying IEE has the potential to stratify the individual risk of GC, similar to magnifying IEE, warranting further investigation with histological assessment.

Systematic review and meta-analysis: Artificial intelligence for the diagnosis of gastric precancerous lesions and Helicobacter pylori infection

BACKGROUND

The endoscopic diagnosis of Helicobacter-pylori(H.pylori) infection and gastric precancerous lesions(GPL), namely atrophic-gastritis and intestinal-metaplasia, still remains challenging. Artificial intelligence(AI) may represent a powerful resource for the endoscopic recognition of these conditions.

AIMS

To explore the diagnostic performance(DP) of AI in the diagnosis of GPL and H.pylori infection.

METHODS

A systematic-review was performed by two independent authors up to September 2021. Inclusion criteria were studies focusing on the DP of AI-system in the diagnosis of GPL and H.pylori infection. The pooled accuracy of studies included was reported.

RESULTS

Overall, 128 studies were found (PubMed-Embase-Cochrane Library) and four and nine studies were finally included regarding GPL and H.pylori infection, respectively. The pooled-accuracy(random effects model) was 90.3%(95%CI 84.3-94.9) and 79.6%(95%CI 66.7-90.0) with a significant heterogeneity[I²=90.4%(95%CI 78.5-95.7);I²=97.9%(97.2-98.6)] for GPL and H.pylori infection, respectively. The Begg's-test showed a significant publication-bias(p = 0.0371) only among studies regarding H.pylori infection. The pooled-accuracy(random-effects-model) was similar considering only studies using CNN-model for the diagnosis of H.pylori infection: 74.1%[(95%CI 51.6-91.3);I²=98.9%(95%CI 98.5-99.3)], Begg's-test(p = 0.1416) did not show publication-bias.

CONCLUSION

AI-system seems to be a good resource for an easier diagnosis of GPL and H.pylori infection, showing a pooled-diagnostic-accuracy of 90% and 80%, respectively. However, considering the high heterogeneity, these promising data need an external validation by randomized control trials and prospective real-time studies.

A deep learning and natural language processing-based system for automatic identification and surveillance of high-risk patients undergoing upper endoscopy: A multicenter study

BACKGROUND

Timely identification and regular surveillance of patients at high risk are crucial for early diagnosis of upper gastrointestinal cancer. However, traditional manual surveillance method is time-consuming, and current surveillance rate is below 50%. Here, we aimed to develop a surveillance system named ENDOANGEL-AS (automatic surveillance) for automatic identification and surveillance of high-risk patients.

METHODS

7874 patients from Renmin Hospital of Wuhan University between May 1 and July 31, 2021 were used as the training set, 6762 patients between August 1 and October 31, 2021 as the internal test set, and 7570 patients from two other hospitals between August 1 and October 31, 2021 as the external test sets. We first extracted descriptions of abnormalities from endoscopic and pathological reports based on natural language processing techniques to identify individuals. Then patients were classified at nine risk levels according to endoscopic and pathological findings, and a deep learning model was trained to identify demarcation line (DL) in gastric low-grade intraepithelial neoplasia (LGIN) using 1561 white-light still images for risk stratification of gastric LGIN. Finally, patients undergoing upper endoscopy were classified and assigned one of ten surveillance intervals according to guidelines. The performance of ENDOANGEL-AS was evaluated and compared with physicians.

FINDINGS

Patient identification module achieved an accuracy of 100% and 99.91% in internal and external test sets, respectively. Risk level classification module achieved an accuracy of 100% and 99.85% in the internal and external test sets, respectively. DL identification module achieved an accuracy of 87.88%. ENDOANGEL-AS on surveillance interval assignment achieved an accuracy of 99.23% and 99.67% in internal and external test sets, respectively. ENDOANGEL-AS had significantly higher accuracy compared with physicians (99.00% vs 38.87%, p < 0.001). The accuracy (63.67%, p < 0.001) of endoscopists with the assistance of ENDOANGEL-AS was significantly improved.

INTERPRETATION

We established a surveillance system that can automatically identify patients and assign surveillance intervals with high accuracy and good transferability.

FUNDING

This work was partly supported by a grant from the Hubei Province Major Science and Technology Innovation Project (2018-916-000-008) and the Fundamental Research Funds for the Central Universities (2042021kf0084).

Low-magnification narrow-band imaging for small gastric neoplasm detection on screening endoscopy

Background and Aims

Microsurface patterns of the gastric mucosa can be observed using magnifying narrow-band imaging (M-NBI). However, the efficacy of M-NBI at low-magnification (LM-NBI) screening for detecting small gastric neoplasms is unclear.

Methods

This prospective study was conducted at a single institution. LM-NBI, defined as minimal magnification that could reveal the microsurface pattern of the gastric mucosa, was performed after routine white-light imaging (WLI) observation of the stomach. Depending on the phase in which the neoplastic lesions were initially found, they were divided into the WLI group and the LM-NBI group, and the characteristics of these neoplastic lesions were investigated accordingly.

Results

Sixty-five epithelial lesions (adenomas or noninvasive carcinomas) of 20 mm or less in diameter were identified in this study. Sixteen lesions were detected only with LM-NBI. Smaller lesions were detected using LM-NBI (P = .01). WLI took about 160 to 260 seconds, while LM-NBI required about 70 to 80 seconds. All lesions in the LM-NBI group had a background of map-like redness (n = 5) or atrophic/metaplastic mucosa (n = 11).

Conclusions

LM-NBI was able to detect lesions overlooked by WLI, especially those in areas of map-like redness or atrophic/metaplastic mucosa of the stomach. Approximately one-quarter of newly diagnosed neoplasms were retrieved on routine examination during an extra 1.5 minutes.

Advances and horizons for artificial intelligence of endoscopic screening and surveillance of gastric and esophageal disease

The development of artificial intelligence in endoscopic assessment of the gastrointestinal tract has shown progressive enhancement in diagnostic acuity. This review discusses the expanding applications for gastric and esophageal diseases. The gastric section covers the utility of AI in detecting and characterizing gastric polyps and further explores prevention, detection, and classification of gastric cancer. The esophageal discussion highlights applications for use in screening and surveillance in Barrett's esophagus and in high-risk conditions for esophageal squamous cell carcinoma. Additionally, these discussions highlight applications for use in assessing eosinophilic esophagitis and future potential in assessing esophageal microbiome changes.

Artificial Intelligence for Upper Gastrointestinal Endoscopy: A Roadmap from Technology Development to Clinical Practice

Stomach cancer is the third deadliest type of cancer in the world (0.86 million deaths in 2017). In 2035, a 20% increase will be observed both in incidence and mortality due to demographic effects if no interventions are foreseen. Upper GI endoscopy (UGIE) plays a paramount role in early diagnosis and, therefore, improved survival rates. On the other hand, human and technical factors can contribute to misdiagnosis while performing UGIE. In this scenario, artificial intelligence (AI) has recently shown its potential in compensating for the pitfalls of UGIE, by leveraging deep learning architectures able to efficiently recognize endoscopic patterns from UGIE video data. This work presents a review of the current state-of-the-art algorithms in the application of AI to gastroscopy. It focuses specifically on the threefold tasks of assuring exam completeness (i.e., detecting the presence of blind spots) and assisting in the detection and characterization of clinical findings, both gastric precancerous conditions and neoplastic lesion changes. Early and promising results have already been obtained using well-known deep learning architectures for computer vision, but many algorithmic challenges remain in achieving the vision of AI-assisted UGIE. Future challenges in the roadmap for the effective integration of AI tools within the UGIE clinical practice are discussed, namely the adoption of more robust deep learning architectures and methods able to embed domain knowledge into image/video classifiers as well as the availability of large, annotated datasets.

From part to whole, operative link on to endoscopic grading of gastric intestinal metaplasia, pathology to endoscopy: gastric intestinal metaplasia graded by endoscopy

Objective: To conduct a systematic review and meta-analysis on the prediction of severity of gastric intestinal metaplasia (GIM) in localized and entire gastric mucosa using endoscopy. Methods: The authors searched Web of Science, PubMed, Embase and Cochrane Central Register of Controlled Trials and performed systematic searches on endoscopic grading of GIM of the entire stomach using Meta-DiSc and Stata. Results: Sensitivity and specificity for the stratified prediction of overall GIM were 0.91 (95% CI: 0.85-0.95) and 0.91 (95% CI: 0.88-0.93), respectively. Sensitivity in predicting the different grades of GIM was higher in operative link on GIM assessment grades 0, III and IV but lower in grades I and II. Conclusion: Digital chromoendoscopy is well suited to predicting the severity of localized and overall GIM.

Real-time semantic segmentation of gastric intestinal metaplasia using a deep learning approach

Background/Aims

Previous artificial intelligence (AI) models attempting to segment gastric intestinal metaplasia (GIM) areas have failed to be deployed in real-time endoscopy due to their slow inference speeds. Here, we propose a new GIM segmentation AI model with inference speeds faster than 25 frames per second that maintains a high level of accuracy.

Methods

Investigators from Chulalongkorn University obtained 802 histological-proven GIM images for AI model training. Four strategies were proposed to improve the model accuracy. First, transfer learning was employed to the public colon datasets. Second, an image preprocessing technique contrast-limited adaptive histogram equalization was employed to produce clearer GIM areas. Third, data augmentation was applied for a more robust model. Lastly, the bilateral segmentation network model was applied to segment GIM areas in real time. The results were analyzed using different validity values.

Results

From the internal test, our AI model achieved an inference speed of 31.53 frames per second. GIM detection showed sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union in GIM segmentation values of 93%, 80%, 82%, 92%, 87%, and 57%, respectively.

Conclusions

The bilateral segmentation network combined with transfer learning, contrast-limited adaptive histogram equalization, and data augmentation can provide high sensitivity and good accuracy for GIM detection and segmentation.

Implementation of artificial intelligence in upper gastrointestinal endoscopy

The application of artificial intelligence (AI) using deep learning has significantly expanded in the field of esophagogastric endoscopy. Recent studies have shown promising results in detecting and differentiating early gastric cancer using AI tools built using white light, magnified, or image‐enhanced endoscopic images. Some studies have reported the use of AI tools to predict the depth of early gastric cancer based on endoscopic images. Similarly, studies based on using AI for detecting early esophageal cancer have also been reported, with an accuracy comparable to that of endoscopy specialists. Moreover, an AI system, developed to diagnose pharyngeal cancer, has shown promising performance with high sensitivity. These reports suggest that, if introduced for regular use in clinical settings, AI systems can significantly reduce the burden on physicians. This review summarizes the current status of AI applications in the upper gastrointestinal tract and presents directions for clinical practice implementation and future research.

Real-time use of artificial intelligence for diagnosing early gastric cancer by magnifying image-enhanced endoscopy: a multicenter diagnostic study (with videos)

BACKGROUND AND AIMS

Endoscopy is a pivotal method for detecting early gastric cancer (EGC). However, skill among endoscopists varies greatly. Here, we proposed a deep learning-based system named ENDOANGEL-ME to diagnose EGC in magnifying image-enhanced endoscopy (M-IEE).

METHODS

M-IEE images were retrospectively obtained from 6 hospitals in China, including 4667 images for training and validation, 1324 images for internal tests, and 4702 images for external tests. One hundred eighty-seven stored videos from 2 hospitals were used to evaluate the performance of ENDOANGEL-ME and endoscopists and to assess the effect of ENDOANGEL-ME on improving the performance of endoscopists. Prospective consecutive patients undergoing M-IEE were enrolled from August 17, 2020 to August 2, 2021 in Renmin Hospital of Wuhan University to assess the applicability of ENDOANGEL-ME in clinical practice.

RESULTS

A total of 3099 patients undergoing M-IEE were enrolled in this study. The diagnostic accuracy of ENDOANGEL-ME for diagnosing EGC was 88.44% and 90.49% in internal and external images, respectively. In 93 internal videos, ENDOANGEL-ME achieved an accuracy of 90.32% for diagnosing EGC, significantly superior to that of senior endoscopists (70.16% ± 8.78%). In 94 external videos, with the assistance of ENDOANGEL-ME, endoscopists showed improved accuracy and sensitivity (85.64% vs 80.32% and 82.03% vs 67.19%, respectively). In 194 prospective consecutive patients with 251 lesions, ENDOANGEL-ME achieved a sensitivity of 92.59% (25/27) and an accuracy of 83.67% (210/251) in real clinical practice.

CONCLUSIONS

This multicenter diagnostic study showed that ENDOANGEL-ME can be well applied in the clinical setting. (Clinical trial registration number: ChiCTR2000035116.).

Gastric epithelial histology and precancerous conditions

The most common histological type of gastric cancer (GC) is gastric adenocarcinoma arising from the gastric epithelium. Less common variants include mesenchymal, lymphoproliferative and neuroendocrine neoplasms. The Lauren scheme classifies GC into intestinal type, diffuse type and mixed type. The WHO classification includes papillary, tubular, mucinous, poorly cohesive and mixed GC. Chronic atrophic gastritis (CAG) and intestinal metaplasia are recommended as common precancerous conditions. No definite precancerous condition of diffuse/poorly/undifferentiated type is recommended. Chronic superficial inflammation and hyperplasia of foveolar cells may be the focus. Presently, the management of early GC and precancerous conditions mainly relies on endoscopy including diagnosis, treatment and surveillance. Management of precancerous conditions promotes the early detection and treatment of early GC, and even prevent the occurrence of GC. In the review, precancerous conditions including CAG, metaplasia, foveolar hyperplasia and gastric hyperplastic polyps derived from the gastric epithelium have been concluded, based on the overview of gastric epithelial histological organization and its renewal.

Deep learning system compared with expert endoscopists in predicting early gastric cancer and its invasion depth and differentiation status (with videos)

BACKGROUND AND AIMS

We aimed to develop and validate a deep learning-based system that covers various aspects of early gastric cancer (EGC) diagnosis, including detecting gastric neoplasm, identifying EGC, and predicting EGC invasion depth and differentiation status. Herein, we provide a state-of-the-art comparison of the system with endoscopists using real-time videos in a nationwide human-machine competition.

METHODS

This multicenter, prospective, real-time, competitive comparative, diagnostic study enrolled consecutive patients who received magnifying narrow-band imaging endoscopy at the Peking University Cancer Hospital from June 9, 2020 to November 17, 2020. The offline competition was conducted in Wuhan, China, and the endoscopists and the system simultaneously read patients' videos and made diagnoses. The primary outcomes were sensitivity in detecting neoplasms and diagnosing EGCs.

RESULTS

One hundred videos, including 37 EGCs and 63 noncancerous lesions, were enrolled; 46 endoscopists from 44 hospitals in 19 provinces in China participated in the competition. The sensitivity rates of the system for detecting neoplasms and diagnosing EGCs were 87.81% and 100%, respectively, significantly higher than those of endoscopists (83.51% [95% confidence interval [CI], 81.23-85.79] and 87.13% [95% CI, 83.75-90.51], respectively). Accuracy rates of the system for predicting EGC invasion depth and differentiation status were 78.57% and 71.43%, respectively, slightly higher than those of endoscopists (63.75% [95% CI, 61.12-66.39] and 64.41% [95% CI, 60.65-68.16], respectively).

CONCLUSIONS

The system outperformed endoscopists in identifying EGCs and was comparable with endoscopists in predicting EGC invasion depth and differentiation status in videos. This deep learning-based system could be a powerful tool to assist endoscopists in EGC diagnosis in clinical practice.

Deep Learning and Device-Assisted Enteroscopy: Automatic Detection of Gastrointestinal Angioectasia

Background and Objectives: Device-assisted enteroscopy (DAE) allows deep exploration of the small bowel and combines diagnostic and therapeutic capacities. Suspected mid-gastrointestinal bleeding is the most frequent indication for DAE, and vascular lesions, particularly angioectasia, are the most common etiology. Nevertheless, the diagnostic yield of DAE for the detection of these lesions is suboptimal. Deep learning algorithms have shown great potential for automatic detection of lesions in endoscopy. We aimed to develop an artificial intelligence (AI) model for the automatic detection of angioectasia DAE images. Materials and Methods: A convolutional neural network (CNN) was developed using DAE images. Each frame was labeled as normal/mucosa or angioectasia. The image dataset was split for the constitution of training and validation datasets. The latter was used for assessing the performance of the CNN. Results: A total of 72 DAE exams were included, and 6740 images were extracted (5345 of normal mucosa and 1395 of angioectasia). The model had a sensitivity of 88.5%, a specificity of 97.1% and an AUC of 0.988. The image processing speed was 6.4 ms/frame. Conclusions: The application of AI to DAE may have a significant impact on the management of patients with suspected mid-gastrointestinal bleeding.

A systematic review on application of deep learning in digestive system image processing

With the advent of the big data era, the application of artificial intelligence represented by deep learning in medicine has become a hot topic. In gastroenterology, deep learning has accomplished remarkable accomplishments in endoscopy, imageology, and pathology. Artificial intelligence has been applied to benign gastrointestinal tract lesions, early cancer, tumors, inflammatory bowel diseases, livers, pancreas, and other diseases. Computer-aided diagnosis significantly improve diagnostic accuracy and reduce physicians' workload and provide a shred of evidence for clinical diagnosis and treatment. In the near future, artificial intelligence will have high application value in the field of medicine. This paper mainly summarizes the latest research on artificial intelligence in diagnosing and treating digestive system diseases and discussing artificial intelligence's future in digestive system diseases. We sincerely hope that our work can become a stepping stone for gastroenterologists and computer experts in artificial intelligence research and facilitate the application and development of computer-aided image processing technology in gastroenterology.

Current situation and prospect of artificial intelligence application in endoscopic diagnosis of Helicobacter pylori infection

With the appearance and prevalence of deep learning, artificial intelligence (AI) has been broadly studied and made great progress in various fields of medicine, including gastroenterology. Helicobacter pylori (H. pylori), closely associated with various digestive and extradigestive diseases, has a high infection rate worldwide. Endoscopic surveillance can evaluate H. pylori infection situations and predict the risk of gastric cancer, but there is no objective diagnostic criteria to eliminate the differences between operators. The computer-aided diagnosis system based on AI technology has demonstrated excellent performance for the diagnosis of H. pylori infection, which is superior to novice endoscopists and similar to skilled. Compared with the visual diagnosis of H. pylori infection by endoscopists, AI possesses voluminous advantages: High accuracy, high efficiency, high quality control, high objectivity, and high-effect teaching. This review summarizes the previous and recent studies on AI-assisted diagnosis of H. pylori infection, points out the limitations, and puts forward prospect for future research.

Status quo and future prospects of artificial neural network from the perspective of gastroenterologists

Artificial neural networks (ANNs) are one of the primary types of artificial intelligence and have been rapidly developed and used in many fields. In recent years, there has been a sharp increase in research concerning ANNs in gastrointestinal (GI) diseases. This state-of-the-art technique exhibits excellent performance in diagnosis, prognostic prediction, and treatment. Competitions between ANNs and GI experts suggest that efficiency and accuracy might be compatible in virtue of technique advancements. However, the shortcomings of ANNs are not negligible and may induce alterations in many aspects of medical practice. In this review, we introduce basic knowledge about ANNs and summarize the current achievements of ANNs in GI diseases from the perspective of gastroenterologists. Existing limitations and future directions are also proposed to optimize ANN’s clinical potential. In consideration of barriers to interdisciplinary knowledge, sophisticated concepts are discussed using plain words and metaphors to make this review more easily understood by medical practitioners and the general public.