Ensembled CNN with artificial bee colony optimization method for esophageal cancer stage classification using SVM classifier.

BACKGROUND: Esophageal cancer (EC) is aggressive cancer with a high fatality rate and a rapid rise of the incidence globally. However, early diagnosis of EC remains a challenging task for clinicians. OBJECTIVE: To help address and overcome this challenge, this study aims to develop and test a new computer-aided diagnosis (CAD) network that combines several machine learning models and optimization methods to detect EC and classify cancer stages. METHODS: The study develops a new deep learning network for the classification of the various stages of EC and the premalignant stage, Barrett's Esophagus from endoscopic images. The proposed model uses a multi-convolution neural network (CNN) model combined with Xception, Mobilenetv2, GoogLeNet, and Darknet53 for feature extraction. The extracted features are blended and are then applied on to wrapper based Artificial Bee Colony (ABC) optimization technique to grade the most accurate and relevant attributes. A multi-class support vector machine (SVM) classifies the selected feature set into the various stages. A study dataset involving 523 Barrett's Esophagus images, 217 ESCC images and 288 EAC images is used to train the proposed network and test its classification performance. RESULTS: The proposed network combining Xception, mobilenetv2, GoogLeNet, and Darknet53 outperforms all the existing methods with an overall classification accuracy of 97.76% using a 3-fold cross-validation method. CONCLUSION: This study demonstrates that a new deep learning network that combines a multi-CNN model with ABC and a multi-SVM is more efficient than those with individual pre-trained networks for the EC analysis and stage classification.

Use of a Cytosponge biomarker panel to prioritise endoscopic Barrett's oesophagus surveillance: a cross-sectional study followed by a real-world prospective pilot.

BACKGROUND: Endoscopic surveillance is recommended for patients with Barrett's oesophagus because, although the progression risk is low, endoscopic intervention is highly effective for high-grade dysplasia and cancer. However, repeated endoscopy has associated harms and access has been limited during the COVID-19 pandemic. We aimed to evaluate the role of a non-endoscopic device (Cytosponge) coupled with laboratory biomarkers and clinical factors to prioritise endoscopy for Barrett's oesophagus. METHODS: We first conducted a retrospective, multicentre, cross-sectional study in patients older than 18 years who were having endoscopic surveillance for Barrett's oesophagus (with intestinal metaplasia confirmed by TFF3 and a minimum Barrett's segment length of 1 cm [circumferential or tongues by the Prague C and M criteria]). All patients had received the Cytosponge and confirmatory endoscopy during the BEST2 (ISRCTN12730505) and BEST3 (ISRCTN68382401) clinical trials, from July 7, 2011, to April 1, 2019 (UK Clinical Research Network Study Portfolio 9461). Participants were divided into training (n=557) and validation (n=334) cohorts to identify optimal risk groups. The biomarkers evaluated were overexpression of p53, cellular atypia, and 17 clinical demographic variables. Endoscopic biopsy diagnosis of high-grade dysplasia or cancer was the primary endpoint. Clinical feasibility of a decision tree for Cytosponge triage was evaluated in a real-world prospective cohort from Aug 27, 2020 (DELTA; ISRCTN91655550; n=223), in response to COVID-19 and the need to provide an alternative to endoscopic surveillance. FINDINGS: The prevalence of high-grade dysplasia or cancer determined by the current gold standard of endoscopic biopsy was 17% (92 of 557 patients) in the training cohort and 10% (35 of 344) in the validation cohort. From the new biomarker analysis, three risk groups were identified: high risk, defined as atypia or p53 overexpression or both on Cytosponge; moderate risk, defined by the presence of a clinical risk factor (age, sex, and segment length); and low risk, defined as Cytosponge-negative and no clinical risk factors. The risk of high-grade dysplasia or intramucosal cancer in the high-risk group was 52% (68 of 132 patients) in the training cohort and 41% (31 of 75) in the validation cohort, compared with 2% (five of 210) and 1% (two of 185) in the low-risk group, respectively. In the real-world setting, Cytosponge results prospectively identified 39 (17%) of 223 patients as high risk (atypia or p53 overexpression, or both) requiring endoscopy, among whom the positive predictive value was 31% (12 of 39 patients) for high-grade dysplasia or intramucosal cancer and 44% (17 of 39) for any grade of dysplasia. INTERPRETATION: Cytosponge atypia, p53 overexpression, and clinical risk factors (age, sex, and segment length) could be used to prioritise patients for endoscopy. Further investigation could validate their use in clinical practice and lead to a substantial reduction in endoscopy procedures compared with current surveillance pathways. FUNDING: Medical Research Council, Cancer Research UK, Innovate UK.

Feasibility and Safety of Tethered Capsule Endomicroscopy in Patients With Barrett's Esophagus in a Multi-Center Study.

BACKGROUND & AIMS: Tethered capsule endomicroscopy (TCE) involves swallowing a small tethered pill that implements optical coherence tomography (OCT) imaging, procuring high resolution images of the whole esophagus. Here, we demonstrate and evaluate the feasibility and safety of TCE and a portable OCT imaging system in patients with Barrett's esophagus (BE) in a multi-center (5-site) clinical study. METHODS: Untreated patients with BE as per endoscopic biopsy diagnosis were eligible to participate in the study. TCE procedures were performed in unsedated patients by either doctors or nurses. After the capsule was swallowed, the device continuously obtained 10-µm-resolution cross-sectional images as it traversed the esophagus. Following imaging, the device was withdrawn through mouth, and disinfected for subsequent reuse. BE lengths were compared to endoscopy findings when available. OCT-TCE images were compared to volumetric laser endomicroscopy (VLE) images from a patient who had undergone VLE on the same day as TCE. RESULTS: 147 patients with BE were enrolled across all sites. 116 swallowed the capsule (79%), 95/114 (83.3%) men and 21/33 (63.6%) women (P = .01). High-quality OCT images were obtained in 104/111 swallowers (93.7%) who completed the procedure. The average imaging duration was 5.55 ± 1.92 minutes. The mean length of esophagus imaged per patient was 21.69 ± 5.90 cm. A blinded comparison of maximum extent of BE measured by OCT-TCE and EGD showed a strong correlation (r = 0.77-0.79). OCT-TCE images were of similar quality to those obtained by OCT-VLE. CONCLUSIONS: The capabilities of TCE to be used across multiple sites, be administered to unsedated patients by either physicians or nurses who are not expert in OCT-TCE, and to rapidly and safely evaluate the microscopic structure of the esophagus make it an emerging tool for screening and surveillance of BE patients. Clinical trial registry website and trial number: NCT02994693 and NCT03459339.

Targeted Hsp70 fluorescence molecular endoscopy detects dysplasia in Barrett's esophagus.

PURPOSE: The incidence of esophageal adenocarcinoma (EAC) has been increasing for decades without significant improvements in treatment. Barrett's esophagus (BE) is best established risk factor for EAC, but current surveillance with random biopsies cannot predict progression to cancer in most BE patients due to the low sensitivity and specificity of high-definition white light endoscopy. METHODS: Here, we evaluated the membrane-bound highly specific Hsp70-specific contrast agent Tumor-Penetrating Peptide (Hsp70-TPP) in guided fluorescence molecular endoscopy biopsy. RESULTS: Hsp70 was significantly overexpressed as determined by IHC in dysplasia and EAC compared with non-dysplastic BE in patient samples (n = 12) and in high-grade dysplastic lesions in a transgenic (L2-IL1b) mouse model of BE. In time-lapse microscopy, Hsp70-TPP was rapidly taken up and internalized  by human BE dysplastic patient-derived organoids. Flexible fluorescence endoscopy of the BE mouse model allowed a specific detection of Hsp70-TPP-Cy5.5 that corresponded closely with the degree of dysplasia but not BE. Ex vivo application of Hsp70-TPP-Cy5.5 to freshly resected whole human EAC specimens revealed a high (> 4) tumor-to-background ratio and a specific detection of previously undetected tumor infiltrations. CONCLUSION: In summary, these findings suggest that Hsp70-targeted imaging using fluorescently labeled TPP peptide may improve tumor surveillance in BE patients.

Wide-area transepithelial sampling for dysplasia detection in Barrett's esophagus: a systematic review and meta-analysis.

BACKGROUND AND AIMS: Seattle protocol forceps biopsy sampling (FB) is currently recommended for surveillance in Barrett's esophagus (BE) but limited by sampling error and lack of compliance. Wide-area transepithelial sampling with 3-dimensional analysis (WATS3D; CDx Diagnostics, Suffern, NY, USA) is reported to increase BE dysplasia detection. We assessed the incremental yield and clinical significance of WATS3D for dysplasia detection over FB in a systematic review and meta-analysis. METHODS: We queried major scientific databases for studies using WATS3D and FB from 2000 to 2020. The primary outcome was the incremental yield of WATS3D-detected dysplasia (defined as a composite of indefinite for dysplasia, low- and high-grade dysplasia [HGD] and esophageal adenocarcinoma [EAC]) over FB. Secondary outcomes were incremental yields of HGD/EAC and rate of reconfirmation of WATS3D dysplasia on subsequent FB. RESULTS: Meta-analysis of 7 eligible studies demonstrated that FB diagnosed dysplasia in 15.9% of cases, whereas the incremental yield with WATS3D was 7.2% (95% confidence interval, 3.9%-11.5%; I2= 92.1%). Meta-analysis of 6 studies demonstrated that FB diagnosed HGD/EAC in 2.3% of patients, whereas the incremental yield with WATS3D was 2.1% (95% confidence interval, .4%-5.3%; I2= 92.7%). Notably, WATS3D was negative in 62.5% of cases where FB identified dysplasia. Two studies reported reconfirmation of WATS3D dysplasia with FB histology in only 20 patients. CONCLUSIONS: WATS3D increases dysplasia detection; however, the clinical significance of this increased dysplasia detection remains uncertain. Data from endoscopic follow-up to ascertain FB histology in patients with dysplasia based solely on WATS3D are needed to determine the optimal clinical application and significance of WATS3D-only dysplasia.

Detection of Barrett's neoplasia with a near-infrared fluorescent heterodimeric peptide.

BACKGROUND: Esophageal adenocarcinoma (EAC) is a molecularly heterogeneous disease with poor prognosis that is rising rapidly in incidence. We aimed to demonstrate specific binding by a peptide heterodimer to Barrett's neoplasia in human subjects. METHODS: Peptide monomers specific for EGFR and ErbB2 were arranged in a heterodimer configuration and labeled with IRDye800. This near-infrared (NIR) contrast agent was topically administered to patients with Barrett's esophagus (BE) undergoing either endoscopic therapy or surveillance. Fluorescence images were collected using a flexible fiber accessory passed through the instrument channel of an upper gastrointestinal endoscope. Fluorescence images were collected from 31 BE patients. A deep learning model was used to segment the target (T) and background (B) regions. RESULTS: The mean target-to-background (T/B) ratio was significantly greater for high grade dysplasia (HGD) and EAC versus BE, low grade dysplasia (LGD), and squamous epithelium. At a T/B ratio of 1.5, sensitivity and specificity of 94.1 % and 92.6 %, respectively, were achieved for the detection of Barrett's neoplasia with an area under the curve of 0.95. No adverse events attributed to the heterodimer were found. EGFR and ErbB2 expression were validated in the resected specimens. CONCLUSIONS: This "first-in-human" clinical study demonstrates the feasibility of detection of early Barrett's neoplasia using a NIR-labeled peptide heterodimer.

Probe-Based Confocal Laser Endomicroscopy and Barrett's Esophagus: Just a Scientific Toy or Significant Improvement in Diagnosis?

BACKGROUND: Barrett's esophagus (BE) requires surveillance to identify potential neoplasia at an early stage. The standard surveillance regimen includes random 4-quadrant biopsies by Seattle protocol. Main limitations of random biopsies are high risk of sampling error, difficulties in histology interpretation, common inadequate classification of pathohistological changes, increased risk of bleeding, and time necessary to acquire the final diagnosis. Probe-based confocal laser endomicroscopy (pCLE) has emerged as a potential tool with an aim to overcome these obvious limitations. SUMMARY: pCLE represents a real-time microscopic imaging method that offers evaluation of epithelial and subepithelial structures with 1,000-fold magnification. In theory, pCLE has potential to eliminate the need for biopsy in BE patients. The main advantages would be real-time diagnosis and decision-making, greater diagnostic accuracy, and evaluation of larger area compared to random biopsies. Clinical pCLE studies in the esophagus show high diagnostic accuracy, and its high negative predictive value offers high reliability and confidence to exclude dysplastic and neoplastic lesions. However, it still cannot replace histopathology due to lower positive predictive value and sensitivity. Key Messages: Despite promising results, its role in routine use in patients with BE remains questionable primarily due to lack of well-organized double-blind randomized trials.

Need for improvement in the evaluation of pre-malignant upper gastro-intestinal lesions in India: Results of a nationwide survey.

BACKGROUND AND AIM: Gastric and esophageal cancers are associated with high morbidity in India. In the absence of formal screening programs in India, it is essential that all elective esophago-gastro-duodenoscopies (EGDs), irrespective of indication, be also considered an opportunity to screen for premalignant lesions. With this premise, we tried to assess the adherence to best practices in the detection of premalignant upper gastro-intestinal lesions (PMUGIL) among endoscopists in India. We also evaluated the adequacy of training, availability of appropriate facilities, and differences between teaching and non-teaching centers. METHODS: We disbursed a survey among endoscopists working in India, through the membership database of the Society of Gastrointestinal Endoscopists of India, by email and instant messaging. The responses were collected and subsequently analyzed. RESULTS: We obtained a total of 422 eligible responses. The adherence to best practices assessed was lower than the set threshold in all except one parameter in both teaching centers and non-teaching centers. Only 58.5% of endoscopists had received training in the detection of PMUGIL. Appropriate image enhanced endoscopy (IEE) facilities were available to only 58.05% of surveyed endoscopists. CONCLUSIONS: Strategies to improve detection of PMUGIL should be directed at improving adherence to best practices, ensuring adequate training of endoscopists in the evaluation of PMUGIL and improving infrastructure.

Esophageal OCT Imaging Using a Paddle Probe Externally Attached to Endoscope.

BACKGROUND AND AIMS: Endoscopic surveillance of Barrett's esophagus (BE) by white light examination is insufficient to diagnose dysplastic change. In this work, we describe an optical imaging method to obtain high-resolution cross-sectional imaging using a paddle-shaped probe affixed to the endoscope tip. METHODS: We integrated Optical Coherence Tomography (OCT), an optical imaging method that produces cross-sectional images, into a paddle probe attached to video endoscope. We acquired images of esophageal epithelium from patients undergoing routine upper GI endoscopy. Images were classified by a reviewer blinded to patient identity and condition, and these results were compared with clinical diagnosis. RESULTS: We successfully captured epithelial OCT images from 30 patients and identified features consistent with both squamous epithelium and Barrett's esophagus. Our blinded image reviewer classified BE versus non-BE with 91.5% accuracy (65/71 image regions), including sensitivity of 84.6% for BE (11/13) and a specificity of 93.1% (54/58). However, in 16 patients, intubation of the probe into the esophagus could not be achieved. CONCLUSIONS: A paddle probe is a feasible imaging format for acquiring cross-sectional OCT images from the esophagus and can provide a structural assessment of BE and non-BE tissue. Probe form factor is the current limiting obstacle, but could be addressed by further miniaturization.

Endoscopic radiofrequency ablation or surveillance in patients with Barrett's oesophagus with confirmed low-grade dysplasia: a multicentre randomised trial.

OBJECTIVE: Due to an annual progression rate of Barrett's oesophagus (BO) with low-grade dysplasia (LGD) between 9% and 13% per year endoscopic ablation therapy is preferred to surveillance. Since this recommendation is based on only one randomised trial, we aimed at checking these results by another multicentre randomised trial with a similar design. DESIGN: A prospective randomised study was performed in 14 centres comparing radiofrequency ablation (RFA) (maximum of 4 sessions) to annual endoscopic surveillance, including patients with a confirmed diagnosis of BO with LGD. Primary outcome was the prevalence of LGD at 3 years. Secondary outcomes were the prevalence of LGD at 1 year, the complete eradication of intestinal metaplasia (CE-IM) at 3 years, the rate of neoplastic progression at 3 years and the treatment-related morbidity. RESULTS: 125 patients were initially included, of whom 82 with confirmed LGD (76 men, mean age 62.3 years) were finally randomised, 40 patients in the RFA and 42 in the surveillance group. At 3 years, CE-IM rates were 35% vs 0% in the RFA and surveillance groups, respectively (p<0.001). At the same time, the prevalence LGD was 34.3% (95% CI 18.6 to 50.0) in the RFA group vs 58.1% (95% CI 40.7 to 75.4) in the surveillance group (OR=0.38 (95% CI 0.14 to 1.02), p=0.05). Neoplastic progression was found in 12.5% (RFA) vs 26.2% (surveillance; p=0.15). The complication rate was maximal after the first RFA treatment (16.9%). CONCLUSION: RFA modestly reduced the prevalence of LGD as well as progression risk at 3 years. The risk-benefit balance of endoscopic ablation therapy should therefore be carefully weighted against surveillance in patients with BO with confirmed LGD. TRIAL REGISTRATION NUMBER: NCT01360541.

Deep-Learning System Detects Neoplasia in Patients With Barrett's Esophagus With Higher Accuracy Than Endoscopists in a Multistep Training and Validation Study With Benchmarking.

BACKGROUND & AIMS: We aimed to develop and validate a deep-learning computer-aided detection (CAD) system, suitable for use in real time in clinical practice, to improve endoscopic detection of early neoplasia in patients with Barrett's esophagus (BE). METHODS: We developed a hybrid ResNet-UNet model CAD system using 5 independent endoscopy data sets. We performed pretraining using 494,364 labeled endoscopic images collected from all intestinal segments. Then, we used 1704 unique esophageal high-resolution images of rigorously confirmed early-stage neoplasia in BE and nondysplastic BE, derived from 669 patients. System performance was assessed by using data sets 4 and 5. Data set 5 was also scored by 53 general endoscopists with a wide range of experience from 4 countries to benchmark CAD system performance. Coupled with histopathology findings, scoring of images that contained early-stage neoplasia in data sets 2-5 were delineated in detail for neoplasm position and extent by multiple experts whose evaluations served as the ground truth for segmentation. RESULTS: The CAD system classified images as containing neoplasms or nondysplastic BE with 89% accuracy, 90% sensitivity, and 88% specificity (data set 4, 80 patients and images). In data set 5 (80 patients and images) values for the CAD system vs those of the general endoscopists were 88% vs 73% accuracy, 93% vs 72% sensitivity, and 83% vs 74% specificity. The CAD system achieved higher accuracy than any of the individual 53 nonexpert endoscopists, with comparable delineation performance. CAD delineations of the area of neoplasm overlapped with those from the BE experts in all detected neoplasia in data sets 4 and 5. The CAD system identified the optimal site for biopsy of detected neoplasia in 97% and 92% of cases (data sets 4 and 5, respectively). CONCLUSIONS: We developed, validated, and benchmarked a deep-learning computer-aided system for primary detection of neoplasia in patients with BE. The system detected neoplasia with high accuracy and near-perfect delineation performance. The Netherlands National Trials Registry, Number: NTR7072.

Prospective development and validation of a volumetric laser endomicroscopy computer algorithm for detection of Barrett's neoplasia.

BACKGROUND AND AIMS: Volumetric laser endomicroscopy (VLE) is an advanced imaging modality used to detect Barrett's esophagus (BE) dysplasia. However, real-time interpretation of VLE scans is complex and time-consuming. Computer-aided detection (CAD) may help in the process of VLE image interpretation. Our aim was to train and validate a CAD algorithm for VLE-based detection of BE neoplasia. METHODS: The multicenter, VLE PREDICT study, prospectively enrolled 47 patients with BE. In total, 229 nondysplastic BE and 89 neoplastic (high-grade dysplasia/esophageal adenocarcinoma) targets were laser marked under VLE guidance and subsequently underwent a biopsy for histologic diagnosis. Deep convolutional neural networks were used to construct a CAD algorithm for differentiation between nondysplastic and neoplastic BE tissue. The CAD algorithm was trained on a set consisting of the first 22 patients (134 nondysplastic BE and 38 neoplastic targets) and validated on a separate test set from patients 23 to 47 (95 nondysplastic BE and 51 neoplastic targets). The performance of the algorithm was benchmarked against the performance of 10 VLE experts. RESULTS: Using the training set to construct the algorithm resulted in an accuracy of 92%, sensitivity of 95%, and specificity of 92%. When performance was assessed on the test set, accuracy, sensitivity, and specificity were 85%, 91%, and 82%, respectively. The algorithm outperformed all 10 VLE experts, who demonstrated an overall accuracy of 77%, sensitivity of 70%, and specificity of 81%. CONCLUSIONS: We developed, validated, and benchmarked a VLE CAD algorithm for detection of BE neoplasia using prospectively collected and biopsy-correlated VLE targets. The algorithm detected neoplasia with high accuracy and outperformed 10 VLE experts. (The Netherlands National Trials Registry (NTR) number: NTR 6728.).

A computer-assisted algorithm for narrow-band imaging-based tissue characterization in Barrett's esophagus.

BACKGROUND AND AIMS: The endoscopic evaluation of narrow-band imaging (NBI) zoom imagery in Barrett's esophagus (BE) is associated with suboptimal diagnostic accuracy and poor interobserver agreement. Computer-aided diagnosis (CAD) systems may assist endoscopists in the characterization of Barrett's mucosa. Our aim was to demonstrate the feasibility of a deep-learning CAD system for tissue characterization of NBI zoom imagery in BE. METHODS: The CAD system was first trained using 494,364 endoscopic images of general endoscopic imagery. Next, 690 neoplastic BE and 557 nondysplastic BE (NDBE) white-light endoscopy overview images were used for refinement training. Subsequently, a third dataset of 112 neoplastic and 71 NDBE NBI zoom images with histologic correlation was used for training and internal validation. Finally, the CAD system was further trained and validated with a fourth, histologically confirmed dataset of 59 neoplastic and 98 NDBE NBI zoom videos. Performance was evaluated using fourfold cross-validation. The primary outcome was the diagnostic performance of the CAD system for classification of neoplasia in NBI zoom videos. RESULTS: The CAD system demonstrated accuracy, sensitivity, and specificity for detection of BE neoplasia using NBI zoom images of 84%, 88%, and 78%, respectively. In total, 30,021 individual video frames were analyzed by the CAD system. Accuracy, sensitivity, and specificity of the video-based CAD system were 83% (95% confidence interval [CI], 78%-89%), 85% (95% CI, 76%-94%), and 83% (95% CI, 76%-90%), respectively. The mean assessment speed was 38 frames per second. CONCLUSION: We have demonstrated promising diagnostic accuracy of predicting the presence/absence of Barrett's neoplasia on histologically confirmed unaltered NBI zoom videos with fast corresponding assessment time.

Expert assessment on volumetric laser endomicroscopy full scans in Barrett's esophagus patients with or without high grade dysplasia or early cancer.

BACKGROUND: Volumetric laser endomicroscopy (VLE) allows for near-microscopic imaging of the superficial esophageal wall and may improve detection of early neoplasia in Barrett's esophagus (BE). Interpretation of a 6-cm long, circumferential VLE "full scan" may however be challenging for endoscopists. We aimed to evaluate the accuracy of VLE experts in correctly diagnosing VLE full scans of early neoplasia and non-dysplastic BE (NDBE). METHODS: 29 VLE full scan videos (15 neoplastic and 14 NDBE) were randomly evaluated by 12 VLE experts using a web-based module. Experts were blinded to the endoscopic BE images and histology. The 15 neoplastic cases contained a subtle endoscopically visible lesion, which on endoscopic resection showed high grade dysplasia or cancer. NDBE cases had no visible lesions and an absence of dysplasia in all biopsies. VLE videos were first scored as "neoplastic" or "NDBE." If neoplastic, assessors located the area most suspicious for neoplasia. Primary outcome was the performance of VLE experts in differentiating between non-dysplastic and neoplastic full scan videos, calculated by accuracy, sensitivity, and specificity. Secondary outcomes included correct location of neoplasia, interobserver agreement, and level of confidence. RESULTS: VLE experts correctly labelled 73 % (95 % confidence interval [CI] 67 % - 79 %) of neoplastic VLE videos. In 54 % (range 27 % - 66 %) both neoplastic diagnosis and lesion location were correct. NDBE videos were consistent with endoscopic biopsies in 52 % (95 %CI 46 % - 57 %). Interobserver agreement was fair (kappa 0.28). High level of confidence was associated with a higher rate of correct neoplastic diagnosis (81 %) and lesion location (73 %). CONCLUSIONS: Identification of subtle neoplastic lesions in VLE full scans by experts was disappointing. Future studies should focus on improving methodologies for reviewing full scans, development of refined VLE criteria for neoplasia, and computer-aided diagnosis of VLE scans.

Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study.

BACKGROUND: The accurate differentiation between T1a and T1b Barrett's-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett's cancer on white-light images. METHODS: Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett's cancer. RESULTS: The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively. CONCLUSION: This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett's cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett's cancer remains challenging for both experts and AI.

Identification of Barrett's esophagus in endoscopic images using deep learning.

BACKGROUND: Development of a deep learning method to identify Barrett's esophagus (BE) scopes in endoscopic images. METHODS: 443 endoscopic images from 187 patients of BE were included in this study. The gastroesophageal junction (GEJ) and squamous-columnar junction (SCJ) of BE were manually annotated in endoscopic images by experts. Fully convolutional neural networks (FCN) were developed to automatically identify the BE scopes in endoscopic images. The networks were trained and evaluated in two separate image sets. The performance of segmentation was evaluated by intersection over union (IOU). RESULTS: The deep learning method was proved to be satisfying in the automated identification of BE in endoscopic images. The values of the IOU were 0.56 (GEJ) and 0.82 (SCJ), respectively. CONCLUSIONS: Deep learning algorithm is promising with accuracies of concordance with manual human assessment in segmentation of the BE scope in endoscopic images. This automated recognition method helps clinicians to locate and recognize the scopes of BE in endoscopic examinations.

Diagnostic accuracy of narrow-band imaging endoscopy with targeted biopsies compared with standard endoscopy with random biopsies in patients with Barrett's esophagus: A systematic review and meta-analysis.

BACKGROUND AND AIM: Endoscopic surveillance for dysplasia in Barrett's esophagus (BE) with random biopsies is the primary diagnostic tool for monitoring clinical progression into esophageal adenocarcinoma. As an alternative, narrow-band imaging (NBI) endoscopy offers targeted biopsies that can improve dysplasia detection. This study aimed to evaluate NBI-guided targeted biopsies' diagnostic accuracy for detecting dysplasia in patients undergoing endoscopic BE surveillance compared with the widely used Seattle protocol. METHODS: Cochrane DTA Register, MEDLINE/PubMed, EMBASE, OpenGrey, and bibliographies of identified papers were searched until 2018. Two independent investigators resolved discrepancies by consensus, study selection, data extraction, and quality assessment. Data on sensitivity, specificity, and predictive values were pooled and analyzed using a random-effects model. RESULTS: Of 9528 identified articles, six studies comprising 493 participants were eligible for quantitative synthesis. NBI-targeted biopsy showed high diagnostic accuracy in detection of dysplasia in BE with a sensitivity of 76% (95% confidence interval [CI]: 0.61-0.91), specificity of 99% (95% CI: 0.99-1.00), positive predictive value of 97% (95% CI: 0.96-0.99), and negative predictive value of 84% (95% CI: 0.69-0.99) for detection of all grades of dysplasia. The receiver-operating characteristic curve for NBI model performance was 0.8550 for detecting all dysplasia. CONCLUSION: Narrow-band imaging-guided biopsy demonstrated high diagnostic accuracy and might constitute a valid substitute for random biopsies during endoscopic surveillance for dysplasia in BE.

SAGES TAVAC safety and efficacy analysis confocal laser endomicroscopy.

BACKGROUND: Confocal laser endomicroscopy (CLE) is a novel endoscopic adjunct that allows real-time in vivo histological examination of mucosal surfaces. By using intravenous or topical fluorescent agents, CLE highlights certain mucosal elements that facilitate an optical biopsy in real time. CLE technology has been used in different organ systems including the gastrointestinal tract. There has been numerous studies evaluating this technology in gastrointestinal endoscopy, our aim was to evaluate the safety, value, and efficacy of this technology in the gastrointestinal tract. METHODS: The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) Technology and Value Assessment Committee (TAVAC) performed a PubMed/Medline database search of clinical studies involving CLE in May of 2018. The literature search used combinations of the keywords: confocal laser endomicroscopy, pCLE, Cellvizio, in vivo microscopy, optical histology, advanced endoscopic imaging, and optical diagnosis. Bibliographies of key references were searched for relevant studies not covered by the PubMed search. Case reports and small case series were excluded. The manufacturer's website was also used to identify key references. The United States Food and Drug Administration (U.S. FDA) Manufacturer And User facility and Device Experience (MAUDE) database was searched for reports regarding the device malfunction or injuries. RESULTS: The technology offers an excellent safety profile with rare adverse events related to the use of fluorescent agents. It has been shown to increase the detection of dysplastic Barrett's esophagus, gastric intraepithelial neoplasia/early gastric cancer, and dysplasia associated with inflammatory bowel disease when compared to standard screening protocols. It also aids in the differentiation and classification of colorectal polyps, indeterminate biliary strictures, and pancreatic cystic lesions. CONCLUSIONS: CLE has an excellent safety profile. CLE can increase the diagnostic accuracy in a number of gastrointestinal pathologies.

Accuracy of artificial intelligence-assisted detection of upper GI lesions: a systematic review and meta-analysis.

BACKGROUND AND AIMS: Artificial intelligence (AI)-assisted detection is increasingly used in upper endoscopy. We performed a meta-analysis to determine the diagnostic accuracy of AI on detection of gastric and esophageal neoplastic lesions and Helicobacter pylori (HP) status. METHODS: We searched Embase, PubMed, Medline, Web of Science, and Cochrane databases for studies on AI detection of gastric or esophageal neoplastic lesions and HP status. After assessing study quality using the Quality Assessment of Diagnostic Accuracy Studies tool, a bivariate meta-analysis following a random-effects model was used to summarize the data and plot hierarchical summary receiver-operating characteristic curves. The diagnostic accuracy was determined by the area under the hierarchical summary receiver-operating characteristic curve (AUC). RESULTS: Twenty-three studies including 969,318 images were included. The AUC of AI detection of neoplastic lesions in the stomach, Barrett's esophagus, and squamous esophagus and HP status were .96 (95% confidence interval [CI], .94-.99), .96 (95% CI, .93-.99), .88 (95% CI, .82-.96), and .92 (95% CI, .88-.97), respectively. AI using narrow-band imaging was superior to white-light imaging on detection of neoplastic lesions in squamous esophagus (.92 vs .83, P < .001). The performance of AI was superior to endoscopists in the detection of neoplastic lesions in the stomach (AUC, .98 vs .87; P < .001), Barrett's esophagus (AUC, .96 vs .82; P < .001), and HP status (AUC, .90 vs .82; P < .001). CONCLUSIONS: AI is accurate in the detection of upper GI neoplastic lesions and HP infection status. However, most studies were based on retrospective reviews of selected images, which requires further validation in prospective trials.

Deep learning algorithm detection of Barrett's neoplasia with high accuracy during live endoscopic procedures: a pilot study (with video).

BACKGROUND AND AIMS: We assessed the preliminary diagnostic accuracy of a recently developed computer-aided detection (CAD) system for detection of Barrett's neoplasia during live endoscopic procedures. METHODS: The CAD system was tested during endoscopic procedures in 10 patients with nondysplastic Barrett's esophagus (NDBE) and 10 patients with confirmed Barrett's neoplasia. White-light endoscopy images were obtained at every 2-cm level of the Barrett's segment and immediately analyzed by the CAD system, providing instant feedback to the endoscopist. At every level, 3 images were evaluated by the CAD system. Outcome measures were diagnostic performance of the CAD system per level and per patient, defined as accuracy, sensitivity, and specificity (ground truth was established by expert assessment and corresponding histopathology), and concordance of 3 sequential CAD predictions per level. RESULTS: Accuracy, sensitivity, and specificity of the CAD system in a per-level analyses were 90%, 91%, and 89%, respectively. Nine of 10 neoplastic patients were correctly diagnosed. The single lesion not detected by CAD showed NDBE in the endoscopic resection specimen. In only 1 NDBE patient, the CAD system produced false-positive predictions. In 75% of all levels, the CAD system produced 3 concordant predictions. CONCLUSIONS: This is one of the first studies to evaluate a CAD system for Barrett's neoplasia during live endoscopic procedures. The system detected neoplasia with high accuracy, with only a small number of false-positive predictions and with a high concordance rate between separate predictions. The CAD system is thereby ready for testing in larger, multicenter trials. (Clinical trial registration number: NL7544.).