Background and Aims Volumetric laser endomicroscopy (VLE) is an advanced imaging system that provides a near-microscopic resolution scan of the esophageal wall layers up to 3-mm deep. VLE has the ...potential to improve detection of early neoplasia in Barrett’s esophagus (BE). However, interpretation of VLE images is complex because of the large amount of data that need to be interpreted in real time. The aim of this study was to investigate the feasibility of a computer algorithm to identify early BE neoplasia on ex vivo VLE images. Methods We used 60 VLE images from a database of high-quality ex vivo VLE-histology correlations, obtained from BE patients ± neoplasia (30 nondysplastic BE NDBE and 30 high-grade dysplasia/early adenocarcinoma images). VLE features from a recently developed clinical VLE prediction score for BE neoplasia served as input for the algorithm: (1) higher VLE surface than subsurface signal and (2) lack of layering. With this input, novel clinically inspired algorithm features were developed, based on signal intensity statistics and grayscale correlations. For comparison, generic image analysis methods were examined for their performance to detect neoplasia. For classification of the images in the NDBE or neoplastic group, several machine learning methods were evaluated. Leave-1-out cross-validation was used for algorithm validation. Results Three novel clinically inspired algorithm features were developed. The feature “layering and signal decay statistics” showed the optimal performance compared with the other clinically features (“layering” and “signal intensity distribution”) and generic image analyses methods, with an area under the receiver operating characteristic curve (AUC) of .95. Corresponding sensitivity and specificity were 90% and 93%, respectively. In addition, the algorithm showed a better performance than the clinical VLE prediction score (AUC .81). Conclusions This is the first study in which a computer algorithm for BE neoplasia was developed based on VLE images with direct histologic correlates. The algorithm showed good performance to detect BE neoplasia in ex vivo VLE images compared with the performance of a recently developed clinical VLE prediction score. This study suggests that an automatic detection algorithm has the potential to assist endoscopists in detecting early neoplasia on VLE. Future studies on in vivo VLE scans are needed to further validate the algorithm.
There has been a vast increase in GI literature focused on the use of machine learning in endoscopy. The relative novelty of this field poses a challenge for reviewers and readers of GI journals. To ...appreciate scientific quality and novelty of machine learning studies, understanding of the technical basis and commonly used techniques is required. Clinicians often lack this technical background, while machine learning experts may be unfamiliar with clinical relevance and implications for daily practice. Therefore, there is an increasing need for a multidisciplinary, international evaluation on how to perform high-quality machine learning research in endoscopy. This review aims to provide guidance for readers and reviewers of peer-reviewed GI journals to allow critical appraisal of the most relevant quality requirements of machine learning studies. The paper provides an overview of common trends and their potential pitfalls and proposes comprehensive quality requirements in six overarching themes: terminology, data, algorithm description, experimental setup, interpretation of results and machine learning in clinical practice.
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).
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.
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).
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
Display omitted
A prior randomized study (Surveillance versus Radiofrequency Ablation study SURF study) demonstrated that radiofrequency ablation (RFA) of Barrett's esophagus (BE) with confirmed low-grade dysplasia ...(LGD) significantly reduces the risk of esophageal adenocarcinoma. Our aim was to report the long-term outcomes of this study.
The SURF study randomized BE patients with confirmed LGD to RFA or surveillance. For this retrospective cohort study, all endoscopic and histologic data acquired at the end of the SURF study in May 2013 until December 2017 were collected. The primary outcome was rate of progression to high-grade dysplasia (HGD)/cancer. All 136 patients randomized to RFA (n = 68) or surveillance (n = 68) in the SURF study were included. After closure of the SURF study, 15 surveillance patients underwent RFA based on patient preference and study outcomes.
With 40 additional months (interquartile range, 12-51), the total median follow-up from randomization to last endoscopy was 73 months (interquartile range, 46-85). HGD/cancer was diagnosed in 1 patient in the RFA group (1.5%) and in 23 in the surveillance group (33.8%) (P = .000), resulting in an absolute risk reduction of 32.4% (95% confidence interval CI, 22.4%-44.2%) with a number needed to treat of 3.1 (95% CI, 2.3-4.5). Seventy-five of 83 patients (90%; 95% CI, 82.1%-95.0%) treated with RFA for BE reached complete clearance of BE and dysplasia. BE recurred in 7 of 75 patients (9%; 95% CI, 4.6%-18.0%), mostly minute islands or tongues, and LGD in 3 of 75 (4%; 95% CI, 1.4%-11.1%).
RFA of BE with confirmed LGD significantly reduces the risk of malignant progression, with sustained clearance of BE in 91% and LGD in 96% of patients, after a median follow-up of 73 months. (Clinical trial registration number: NTR1198.)
Early neoplasia in Barrett's esophagus is difficult to detect and often overlooked during Barrett's surveillance. An automatic detection system could be beneficial, by assisting endoscopists with ...detection of early neoplastic lesions. The aim of this study was to assess the feasibility of a computer system to detect early neoplasia in Barrett's esophagus.
Based on 100 images from 44 patients with Barrett's esophagus, a computer algorithm, which employed specific texture, color filters, and machine learning, was developed for the detection of early neoplastic lesions in Barrett's esophagus. The evaluation by one endoscopist, who extensively imaged and endoscopically removed all early neoplastic lesions and was not blinded to the histological outcome, was considered the gold standard. For external validation, four international experts in Barrett's neoplasia, who were blinded to the pathology results, reviewed all images.
The system identified early neoplastic lesions on a per-image analysis with a sensitivity and specificity of 0.83. At the patient level, the system achieved a sensitivity and specificity of 0.86 and 0.87, respectively. A trade-off between the two performance metrics could be made by varying the percentage of training samples that showed neoplastic tissue.
The automated computer algorithm developed in this study was able to identify early neoplastic lesions with reasonable accuracy, suggesting that automated detection of early neoplasia in Barrett's esophagus is feasible. Further research is required to improve the accuracy of the system and prepare it for real-time operation, before it can be applied in clinical practice.
Accurate endoscopic differentiation would enable to resect and discard small and diminutive colonic lesions, thereby increasing cost-efficiency. Current classification systems based on narrow band ...imaging (NBI), however, do not include neoplastic sessile serrated adenomas/polyps (SSA/Ps). We aimed to develop and validate a new classification system for endoscopic differentiation of adenomas, hyperplastic polyps and SSA/Ps <10 mm.
We developed the Workgroup serrAted polypS and Polyposis (WASP) classification, combining the NBI International Colorectal Endoscopic classification and criteria for differentiation of SSA/Ps in a stepwise approach. Ten consultant gastroenterologists predicted polyp histology, including levels of confidence, based on the endoscopic aspect of 45 polyps, before and after participation in training in the WASP classification. After 6 months, the same endoscopists predicted polyp histology of a new set of 50 polyps, with a ratio of lesions comparable to daily practice.
The accuracy of optical diagnosis was 0.63 (95% CI 0.54 to 0.71) at baseline, which improved to 0.79 (95% CI 0.72 to 0.86, p<0.001) after training. For polyps diagnosed with high confidence the accuracy was 0.73 (95% CI 0.64 to 0.82), which improved to 0.87 (95% CI 0.80 to 0.95, p<0.01). The accuracy of optical diagnosis after 6 months was 0.76 (95% CI 0.72 to 0.80), increasing to 0.84 (95% CI 0.81 to 0.88) considering high confidence diagnosis. The combined negative predictive value with high confidence of diminutive neoplastic lesions (adenomas and SSA/Ps together) was 0.91 (95% CI 0.83 to 0.96).
We developed and validated the first integrative classification method for endoscopic differentiation of small and diminutive adenomas, hyperplastic polyps and SSA/Ps. In a still image evaluation setting, introduction of the WASP classification significantly improved the accuracy of optical diagnosis overall as well as SSA/P in particular, which proved to be sustainable after 6 months.
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.
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.
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.
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.
IMPORTANCE Barrett esophagus containing low-grade dysplasia is associated with an increased risk of developing esophageal adenocarcinoma, a cancer with a rapidly increasing incidence in the western ...world. OBJECTIVE To investigate whether endoscopic radiofrequency ablation could decrease the rate of neoplastic progression. DESIGN, SETTING, AND PARTICIPANTS Multicenter randomized clinical trial that enrolled 136 patients with a confirmed diagnosis of Barrett esophagus containing low-grade dysplasia at 9 European sites between June 2007 and June 2011. Patient follow-up ended May 2013. INTERVENTIONS Eligible patients were randomly assigned in a 1:1 ratio to either endoscopic treatment with radiofrequency ablation (ablation) or endoscopic surveillance (control). Ablation was performed with the balloon device for circumferential ablation of the esophagus or the focal device for targeted ablation, with a maximum of 5 sessions allowed. MAIN OUTCOMES AND MEASURES The primary outcome was neoplastic progression to high-grade dysplasia or adenocarcinoma during a 3-year follow-up since randomization. Secondary outcomes were complete eradication of dysplasia and intestinal metaplasia and adverse events. RESULTS Sixty-eight patients were randomized to receive ablation and 68 to receive control. Ablation reduced the risk of progression to high-grade dysplasia or adenocarcinoma by 25.0% (1.5% for ablation vs 26.5% for control; 95% CI, 14.1%-35.9%; P < .001) and the risk of progression to adenocarcinoma by 7.4% (1.5% for ablation vs 8.8% for control; 95% CI, 0%-14.7%; P = .03). Among patients in the ablation group, complete eradication occurred in 92.6% for dysplasia and 88.2% for intestinal metaplasia compared with 27.9% for dysplasia and 0.0% for intestinal metaplasia among patients in the control group (P < .001). Treatment-related adverse events occurred in 19.1% of patients receiving ablation (P < .001). The most common adverse event was stricture, occurring in 8 patients receiving ablation (11.8%), all resolved by endoscopic dilation (median, 1 session). The data and safety monitoring board recommended early termination of the trial due to superiority of ablation for the primary outcome and the potential for patient safety issues if the trial continued. CONCLUSIONS AND RELEVANCE In this randomized trial of patients with Barrett esophagus and a confirmed diagnosis of low-grade dysplasia, radiofrequency ablation resulted in a reduced risk of neoplastic progression over 3 years of follow-up. TRIAL REGISTRATION trialregister.nl Identifier: NTR1198
Background
Endoscopic resection for early oesophageal cancer was introduced around 2000 in the Netherlands. The scientific question was how the treatment and survival of early oesophageal and ...gastro-oesophageal junction cancer has changed over time in the Netherlands.
Methods
Data were obtained from the nationwide population-based Netherlands Cancer Registry. All patients diagnosed with clinical in situ or T1 oesophageal or GOJ cancer without lymph node or distance metastasis during the study period (2000–2014) were extracted. Primary outcome parameters were the trends in treatment modalities over time and relative survival of each treatment regime.
Results
A total of 1020 patients were diagnosed with a clinical in situ or T1 oesophageal or gastro-oesophageal junction cancer without lymph node or distance metastasis. The proportion of patients who received endoscopic treatment increased from 2.5% in 2000 to 58.1% in 2014. During the same period the proportion of patients who received surgery decreased from 57.5 to 23.1%. Five-year relative survival of all patients was 69%. Five-year relative survival after endoscopic therapy was 83% and after surgery 80%. Relative excess risk analyses showed no significant difference in survival between patients in the endoscopic therapy group and patients in the surgery group after adjustment for age, sex, clinical TNM classification, morphology and tumour location (RER 1.15; CI 0.76–1.75;
p
0.76).
Conclusion
Our results demonstrate an increase in endoscopic treatment and a decrease of surgical treatment for in situ and T1 oesophageal/GOJ cancer between 2000–2014 in the Netherlands. The relative 5-year survival after endoscopic treatment is high (83%) and comparable with surgery (80%).