Computer-Aided Diagnosis for Leaving Colorectal Polyps In Situ : A Systematic Review and Meta-analysis.

BACKGROUND: Computer-aided diagnosis (CADx) allows prediction of polyp histology during colonoscopy, which may reduce unnecessary removal of nonneoplastic polyps. However, the potential benefits and harms of CADx are still unclear. PURPOSE: To quantify the benefit and harm of using CADx in colonoscopy for the optical diagnosis of small (≤5-mm) rectosigmoid polyps. DATA SOURCES: Medline, Embase, and Scopus were searched for articles published before 22 December 2023. STUDY SELECTION: Histologically verified diagnostic accuracy studies that evaluated the real-time performance of physicians in predicting neoplastic change of small rectosigmoid polyps without or with CADx assistance during colonoscopy. DATA EXTRACTION: The clinical benefit and harm were estimated on the basis of accuracy values of the endoscopist before and after CADx assistance. The certainty of evidence was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework. The outcome measure for benefit was the proportion of polyps predicted to be nonneoplastic that would avoid removal with the use of CADx. The outcome measure for harm was the proportion of neoplastic polyps that would be not resected and left in situ due to an incorrect diagnosis with the use of CADx. Histology served as the reference standard for both outcomes. DATA SYNTHESIS: Ten studies, including 3620 patients with 4103 small rectosigmoid polyps, were analyzed. The studies that assessed the performance of CADx alone (9 studies; 3237 polyps) showed a sensitivity of 87.3% (95% CI, 79.2% to 92.5%) and specificity of 88.9% (CI, 81.7% to 93.5%) in predicting neoplastic change. In the studies that compared histology prediction performance before versus after CADx assistance (4 studies; 2503 polyps), there was no difference in the proportion of polyps predicted to be nonneoplastic that would avoid removal (55.4% vs. 58.4%; risk ratio [RR], 1.06 [CI, 0.96 to 1.17]; moderate-certainty evidence) or in the proportion of neoplastic polyps that would be erroneously left in situ (8.2% vs. 7.5%; RR, 0.95 [CI, 0.69 to 1.33]; moderate-certainty evidence). LIMITATION: The application of optical diagnosis was only simulated, potentially altering the decision-making process of the operator. CONCLUSION: Computer-aided diagnosis provided no incremental benefit or harm in the management of small rectosigmoid polyps during colonoscopy. PRIMARY FUNDING SOURCE: European Commission. (PROSPERO: CRD42023402197).

Accuracy of Computer-aided Diagnosis in Colonoscopy Varies according to Polyp Location. A Systematic Review and Meta-analysis.

BACKGROUND & AIMS: Computer-Aided Diagnosis (CADx) assists endoscopists in differentiating between neoplastic and non-neoplastic polyps during colonoscopy. This study aimed to evaluate the impact of polyp location (proximal vs. distal colon) on the diagnostic performance of CADx for ≤5mm polyps. METHODS: We searched for studies evaluating the performance of real-time CADx alone (i.e., independently of endoscopist judgement) for predicting the histology of colorectal polyps ≤5mm. The primary endpoints were CADx sensitivity and specificity in the proximal and distal colon. Secondary outcomes were the negative predictive value (NPV), positive predictive value (PPV), and the accuracy of the CADx alone. Distal colon was limited to the rectum and sigmoid. RESULTS: We included 11 studies for analysis with a total of 7,782 <5mm polyps. CADx specificity was significantly lower in the proximal colon compared to the distal colon (62% versus 85%; Risk ratio (RR): 0.74 [95% CI: 0.72-0.84]). Conversely, sensitivity was similar (89% vs 87% (EC-1); RR: 1.00 [95% CI: 0.97-1.03]. The NPV (64% versus 93%; RR: 0.71 [95% CI: 0.64-0.79]) and accuracy (81% vs 86%; RR: 0.95 [95% CI: 0.91-0.99]) were significantly lower in the proximal than distal colon, while PPV was higher in the proximal colon (87% vs 76%; RR: 1.11 [95% CI: 1.06-1.17]). CONCLUSION: The diagnostic performance of CADx for polyps in the proximal colon is inadequate, exhibiting significantly lower specificity compared to its performance for distal polyps. While current CADx systems are suitable for use in the distal colon, they should not be employed for proximal polyps until more performant systems are developed specifically for these lesions.

Precision endoscopy in the era of climate change and sustainability.

Global warming caused by increased greenhouse gas (GHG) emissions has a direct impact on human health. Gastrointestinal (GI) endoscopy contributes significantly to GHG emissions due to energy consumption, reprocessing of endoscopes and accessories, production of equipment, safe disposal of biohazardous waste, and travel by patients. Moreover, GHGs are also generated in histopathology through tissue processing and the production of biopsy specimen bottles. The reduction in unnecessary surveillance endoscopies and biopsies is a practical approach to decrease GHG emissions without affecting disease outcomes. This narrative review explores the role of precision medicine in GI endoscopy, such as image-enhanced endoscopy and artificial intelligence, with a focus on decreasing unnecessary endoscopic procedures and biopsies in the surveillance and diagnosis of premalignant lesions in the esophagus, stomach, and colon. This review offers strategies to minimize unnecessary endoscopic procedures and biopsies, decrease GHG emissions, and maintain high-quality patient care, thereby contributing to sustainable healthcare practices.

Predictors of technical difficulty for trainees in esophageal endoscopic submucosal dissection.

BACKGROUND: Esophageal endoscopic submucosal dissection (ESD) is technically challenging, especially for trainees, and requires a safe training system. This study aimed to identify predictors of technical difficulty facing trainees performing esophageal ESD to establish such system. METHODS: This was a single-center retrospective study of patients with esophageal cancer who underwent ESD performed by trainees between January 2010 and August 2022. Technical difficulties were defined as muscularis propria exposure and long procedure time (≥ 90 min). Factors associated with these technical difficulties were investigated. RESULTS: A total of 798 lesions in 721 patients were evaluated. Muscularis propria exposure occurred in 298 lesions (37.3%), including 10 perforations (1.3%). The procedure time was ≥ 90 min in 134 lesions (16.8%). In the multivariate analysis, tumor size ≥ 20 mm, tumors ≥ 1/2 of the circumference, and those close to previous treatment scars significantly increased the incidence of both difficulties, whereas tumors in the upper esophagus significantly decreased this incidence. Furthermore, female sex and tumors in the left wall were independent predictors of muscularis propria exposure, and elevated morphology was an independent predictor of long procedure time. Muscularis propria exposure and long procedure time occurred in more than half of the cases with three or more predictors of each difficulty. CONCLUSIONS: Large tumors and tumors close to previous treatment scars increase technical difficulties for trainees in esophageal ESD. Conversely, tumors in the upper esophagus reduce these difficulties. These results enable us to predict the difficulty level preoperatively and select appropriate cases in stepwise training.

Real-World Validation of a Computer-Aided Diagnosis System for Prediction of Polyp Histology in Colonoscopy: A Prospective Multicenter Study.

INTRODUCTION: Computer-aided diagnosis (CADx) of polyp histology could support endoscopists in clinical decision-making. However, this has not been validated in a real-world setting. METHODS: We performed a prospective, multicenter study comparing CADx and endoscopist predictions of polyp histology in real-time colonoscopy. Optical diagnosis based on visual inspection of polyps was made by experienced endoscopists. After this, the automated output from the CADx support tool was recorded. All imaged polyps were resected for histological assessment. Primary outcome was difference in diagnostic performance between CADx and endoscopist prediction of polyp histology. Subgroup analysis was performed for polyp size, bowel preparation, difficulty of location of the polyps, and endoscopist experience. RESULTS: A total of 661 eligible polyps were resected in 320 patients aged ≥40 years between March 2021 and July 2022. CADx had an overall accuracy of 71.6% (95% confidence interval [CI] 68.0-75.0), compared with 75.2% (95% CI 71.7-78.4) for endoscopists ( P = 0.023). The sensitivity of CADx for neoplastic polyps was 61.8% (95% CI 56.9-66.5), compared with 70.3% (95% CI 65.7-74.7) for endoscopists ( P < 0.001). The interobserver agreement between CADx and endoscopist predictions of polyp histology was moderate (83.1% agreement, κ 0.661). When there was concordance between CADx and endoscopist predictions, the accuracy increased to 78.1%. DISCUSSION: The overall diagnostic accuracy and sensitivity for neoplastic polyps was higher in experienced endoscopists compared with CADx predictions, with moderate interobserver agreement. Concordance in predictions increased this diagnostic accuracy. Further research is required to improve the performance of CADx and to establish its role in clinical practice.

Diet and cancer of the esophagus and stomach.

PURPOSE OF REVIEW: Dietary factors have been linked to cancers. This review focuses on several nutrients, which have strong evidence showing increase in cancer risks in the esophagus and stomach. RECENT FINDINGS: Obesity is an important risk factor in upper gastrointestinal cancers. High sugar content in food and sugary drinks are the main drivers of obesity. Proinflammatory diet is another dietary factor, which is increasingly recognized as being associated with esophageal and gastric cancer. SUMMARY: Cancer has been predicted to be the leading cause of death in this century. Cancers of the esophagus and stomach are the six and third most common cause of death worldwide. Although Helicobacter pylori infection is a known cause of gastric cancer, obesity is a leading contributor to esophageal adenocarcinoma. Epidemiological data have shown that dietary factors are associated with the two cancers. Observational, case control, animal and recent large cohort studies have identified associations between dietary factors and upper gastrointestinal cancer. Data are also emerging from studies, which look at dietary patterns, such as reduction in the Dietary Inflammatory Index, as well as adherence to a modified Mediterranean diet, and its association with the incidence of esophageal and gastric cancer.

Artificial intelligence and polyp detection in colonoscopy: Use of a single neural network to achieve rapid polyp localization for clinical use.

BACKGROUND AND AIM: Artificial intelligence has been extensively studied to assist clinicians in polyp detection, but such systems usually require expansive processing power, making them prohibitively expensive and hindering wide adaption. The current study used a fast object detection algorithm, known as the YOLOv3 algorithm, to achieve real-time polyp detection on a laptop. In addition, we evaluated and classified the causes of false detections to further improve accuracy. METHODS: The YOLOv3 algorithm was trained and validated with 6038 and 2571 polyp images, respectively. Videos from live colonoscopies in a tertiary center and those obtained from public databases were used for the training and validation sets. The algorithm was tested on 10 unseen videos from the CVC-Video ClinicDB dataset. Only bounding boxes with an intersection over union area of > 0.3 were considered positive predictions. RESULTS: Polyp detection rate in our study was 100%, with the algorithm able to detect every polyp in each video. Sensitivity, specificity, and F1 score were 74.1%, 85.1%, and 83.3, respectively. The algorithm achieved a speed of 61.2 frames per second (fps) on a desktop RTX2070 GPU and 27.2 fps on a laptop GTX2060 GPU. Nearly a quarter of false negatives happened when the polyps were at the corner of an image. Image blurriness accounted for approximately 3% and 9% of false positive and false negative detections, respectively. CONCLUSION: The YOLOv3 algorithm can achieve real-time poly detection with high accuracy and speed on a desktop GPU, making it low cost and accessible to most endoscopy centers worldwide.

A prospective randomized tandem gastroscopy pilot study of linked color imaging versus white light imaging for detection of upper gastrointestinal lesions.

BACKGROUND AND AIM: Gastrointestinal (GI) lesions may have subtle morphological changes. Linked color imaging (LCI) combines narrow-band wavelength light and white light imaging (WLI) in appropriate balance to enhance lesion detection. We compared the detection rates of upper GI lesions using LCI and WLI. METHOD: Patients were randomized in a 1:1 ratio to receive tandem gastroscopy with WLI inspection followed by LCI, or vice versa. Endoscopic examination was performed using the EG-L590ZW gastroscope and the LASEREO endoscope system (Fujifilm Co., Tokyo, Japan). Histology was reported by a specialist GI pathologist blinded to the technique of lesion detection and was used as the gold standard for diagnosis. RESULTS: Ninety patients (mean age 66.8 years, 51.5% male patients) were randomized to either LCI examination first followed by WLI (LCI-WLI), or vice versa (WLI-LCI). An 18.9% of gastroscopies in the study were for surveillance of previously known gastric cancer precursors. Ten patients (11.1%) had a history of Helicobacter pylori infection. There was no significant difference in the time taken for examination under LCI (311 ± 96 s) and WLI (342 ± 86 s) (P = 0.700). LCI detection rates were higher than WLI detection rates for gastric cancer precursors such as atrophic gastritis (2.19% vs 0.55%) (P < 0.01) and intestinal metaplasia (19.73% vs 7.67%) (P < 0.01). Both sensitivity (82.74% vs 50.96%) and specificity (98.71% vs 96.10%) of LCI were higher than WLI for detection of upper GI lesions. CONCLUSIONS: Linked color imaging had better detection rates, sensitivity, and specificity for detection of upper GI lesions compared with WLI.

A systematic review and meta-analysis of the COVID-19 associated liver injury.

INTRODUCTION AND OBJECTIVES: The novel coronavirus disease 2019 (COVID-19) has affected more than 5 million people globally. Data on the prevalence and degree of COVID-19 associated liver injury among patients with COVID-19 remain limited. We conducted a systematic review and meta-analysis to assess the prevalence and degree of liver injury between patients with severe and non-severe COVID-19. METHODS: We performed a systematic search of three electronic databases (PubMed/MEDLINE, EMBASE and Cochrane Library), from inception to 24th April 2020. We included all adult human studies (>20 subjects) regardless of language, region or publication date or status. We assessed the pooled odds ratio (OR), mean difference (MD) and 95% confidence interval (95%CI) using the random-effects model. RESULTS: Among 1543 citations, there were 24 studies (5961 subjects) which fulfilled our inclusion criteria. The pooled odds ratio for elevated ALT (OR = 2.5, 95%CI: 1.6-3.7, I2 = 57%), AST (OR = 3.4, 95%CI: 2.3-5.0, I2 = 56%), hyperbilirubinemia (OR = 1.7, 95%CI: 1.2-2.5, I2 = 0%) and hypoalbuminemia (OR = 7.1, 95%CI: 2.1-24.1, I2 = 71%) were higher subjects in critical COVID-19. CONCLUSION: COVID-19 associated liver injury is more common in severe COVID-19 than non-severe COVID-19. Physicians should be aware of possible progression to severe disease in subjects with COVID-19-associated liver injury.

Adaptation of endoscopic submucosal dissection techniques for endoscopic full-thickness resection: a review of key steps and technical aspects.

Background and Aims: GI stromal tumors (GISTs) represent the most common mesenchymal tumors of the GI tract. Guidelines recommend the removal of histologically proven gastric GISTs >2 cm. While the conventional treatment of a gastric GIST involves surgical excision, endoscopic full-thickness resection (EFTR) has been described as an acceptable alternative. We aim to outline how the key steps used in endoscopic submucosal dissection (ESD) can be adapted to the performance of exposed EFTR and discuss the variations in technical aspects between the 2 procedures. Methods: We use a video case illustration with a comprehensive narrative to highlight the similarities and differences in equipment used and techniques in EFTR and ESD. Images and graphical illustrations are also used to describe these techniques. Results: ESD techniques and equipment can be adapted for use in EFTR of gastric GISTs. Principles such as deep mucosal incision, the appropriate use of traction, and identification of vessels for prophylactic coagulation help to ensure a safe and efficient procedure. The main difference in EFTR is the need for general anesthesia, starting the mucosal incision as close to the tumor margin as possible, submucosal dissection around the surface of the tumor capsule, and a strong closure method for the muscle defect. Conclusions: The equipment and techniques in ESD can be adapted to EFTR for gastric GISTs by endoscopists who are familiar with ESD techniques.

Multimodal management of foregut neuroendocrine neoplasms.

The foregut, which includes the esophagus, stomach and duodenum, represents one of the most common sites for neuroendocrine neoplasms. These are highly heterogenous with different risk of progression depending on location, cell-type of origin, size, grade and other factors. Various endoscopic and imaging modalities exist to inform therapeutic decision-making, which may be in the form of surgical or endoscopic resection and medical therapy depending on the extent of the disease after diagnostic evaluation. This narrative review aims to explore the literature on the multimodal management of such foregut neuroendocrine neoplasms.

Use of artificial intelligence in the management of T1 colorectal cancer: a new tool in the arsenal or is deep learning out of its depth?

The field of artificial intelligence is rapidly evolving, and there has been an interest in its use to predict the risk of lymph node metastasis in T1 colorectal cancer. Accurately predicting lymph node invasion may result in fewer patients undergoing unnecessary surgeries; conversely, inadequate assessments will result in suboptimal oncological outcomes. This narrative review aims to summarize the current literature on deep learning for predicting the probability of lymph node metastasis in T1 colorectal cancer, highlighting areas of potential application and barriers that may limit its generalizability and clinical utility.

Computer-aided diagnosis for the resect-and-discard strategy for colorectal polyps: a systematic review and meta-analysis.

BACKGROUND: The resect-and-discard strategy allows endoscopists to replace post-polypectomy pathology with real-time prediction of polyp histology during colonoscopy (optical diagnosis). We aimed to investigate the benefits and harms of implementing computer-aided diagnosis (CADx) for polyp pathology into the resect-and-discard strategy. METHODS: In this systematic review and meta-analysis, we searched MEDLINE, Embase, and Scopus from database inception to June 5, 2024, without language restrictions, for diagnostic accuracy studies that assessed the performance of real-time CADx systems, compared with histology, for the optical diagnosis of diminutive polyps (≤5 mm) in the entire colon. We synthesised data for three strategies: CADx-alone, CADx-unassisted, and CADx-assisted; when the endoscopist was involved in the optical diagnosis, we synthesised data exclusively from diagnoses for which confidence in the prediction was reported as high. The primary outcomes were the proportion of polyps that would have avoided pathological assessment (ie, the proportion optically diagnosed with high confidence; main benefit) and the proportion of polyps incorrectly predicted due to false positives and false negatives (main harm), directly compared between CADx-assisted and CADx-unassisted strategies. We used DerSimonian and Laird's random-effects model to calculate all outcomes. We used Higgins I2 to assess heterogeneity, the Grading of Recommendations, Assessment, Development, and Evaluation approach to rate certainty, and funnel plots and Egger's test to examine publication bias. This study is registered with PROSPERO, CRD42024508440. FINDINGS: We found 1019 studies, of which 11 (7400 diminutive polyps, 3769 patients, and 185 endoscopists) were included in the final meta-analysis. Three studies (1817 patients and 4086 polyps [2148 neoplastic and 1938 non-neoplastic]) provided data to directly compare the primary outcome measures between the CADx-unassisted and CADx-assisted strategies. We found no significant difference between the CADx-assisted and CADx-unassisted strategies for the proportion of polyps that would have avoided pathological assessment (90% [88-93], 3653 [89·4%] of 4086 polyps diagnosed with high confidence vs 90% [95% CI 85-94], 3588 [87·8%] of 4086 polyps diagnosed with high confidence; risk ratio 1·01 [95% CI 0·99-1·04; I2=53·49%; low-certainty evidence; Egger's test p=0·18). The proportion of incorrectly predicted polyps was lower with the CADx-assisted strategy than with the CADx-unassisted strategy (12% [95% CI 7-17], 523 [14·3%] of 3653 polyps incorrectly predicted with a CADx-assisted strategy vs 13% [6-20], 582 [16·2%] of 3588 polyps incorrectly diagnosed with a CADx-unassisted strategy; risk ratio 0·88 [95% CI 0·79-0·98]; I2=0·00%; low-certainty evidence; Egger's test p=0·18). INTERPRETATION: CADx did not produce benefit nor harm for the resect-and-discard strategy, questioning its value in clinical practice. Improving the accuracy and explainability of CADx is desired. FUNDING: European Commission (Horizon Europe), the Japan Society of Promotion of Science, and Associazione Italiana per la Ricerca sul Cancro.

Resection of large terminal ileum polyp: usefulness of underwater EMR.

Video 1Underwater endoscopic mucosal resection for a large polyp at the terminal ileum.

Patient Acceptability of Symptom Screening and Patient Education Using a Chatbot for Autoimmune Inflammatory Diseases: Survey Study.

BACKGROUND: Chatbots have the potential to enhance health care interaction, satisfaction, and service delivery. However, data regarding their acceptance across diverse patient populations are limited. In-depth studies on the reception of chatbots by patients with chronic autoimmune inflammatory diseases are lacking, although such studies are vital for facilitating the effective integration of chatbots in rheumatology care. OBJECTIVE: We aim to assess patient perceptions and acceptance of a chatbot designed for autoimmune inflammatory rheumatic diseases (AIIRDs). METHODS: We administered a comprehensive survey in an outpatient setting at a top-tier rheumatology referral center. The target cohort included patients who interacted with a chatbot explicitly tailored to facilitate diagnosis and obtain information on AIIRDs. Following the RE-AIM (Reach, Effectiveness, Adoption, Implementation and Maintenance) framework, the survey was designed to gauge the effectiveness, user acceptability, and implementation of the chatbot. RESULTS: Between June and October 2022, we received survey responses from 200 patients, with an equal number of 100 initial consultations and 100 follow-up (FU) visits. The mean scores on a 5-point acceptability scale ranged from 4.01 (SD 0.63) to 4.41 (SD 0.54), indicating consistently high ratings across the different aspects of chatbot performance. Multivariate regression analysis indicated that having a FU visit was significantly associated with a greater willingness to reuse the chatbot for symptom determination (P=.01). Further, patients' comfort with chatbot diagnosis increased significantly after meeting physicians (P<.001). We observed no significant differences in chatbot acceptance according to sex, education level, or diagnosis category. CONCLUSIONS: This study underscores that chatbots tailored to AIIRDs have a favorable reception. The inclination of FU patients to engage with the chatbot signifies the possible influence of past clinical encounters and physician affirmation on its use. Although further exploration is required to refine their integration, the prevalent positive perceptions suggest that chatbots have the potential to strengthen the bridge between patients and health care providers, thus enhancing the delivery of rheumatology care to various cohorts.