Dataset Information

Convolution neural network for the diagnosis of wireless capsule endoscopy: a systematic review and meta-analysis.

ABSTRACT:

Background

Wireless capsule endoscopy (WCE) is considered to be a powerful instrument for the diagnosis of intestine diseases. Convolution neural network (CNN) is a type of artificial intelligence that has the potential to assist the detection of WCE images. We aimed to perform a systematic review of the current research progress to the CNN application in WCE.

Methods

A search in PubMed, SinoMed, and Web of Science was conducted to collect all original publications about CNN implementation in WCE. Assessment of the risk of bias was performed by Quality Assessment of Diagnostic Accuracy Studies-2 risk list. Pooled sensitivity and specificity were calculated by an exact binominal rendition of the bivariate mixed-effects regression model. I² was used for the evaluation of heterogeneity.

Results

16 articles with 23 independent studies were included. CNN application to WCE was divided into detection on erosion/ulcer, gastrointestinal bleeding (GI bleeding), and polyps/cancer. The pooled sensitivity of CNN for erosion/ulcer is 0.96 [95% CI 0.91, 0.98], for GI bleeding is 0.97 (95% CI 0.93-0.99), and for polyps/cancer is 0.97 (95% CI 0.82-0.99). The corresponding specificity of CNN for erosion/ulcer is 0.97 (95% CI 0.93-0.99), for GI bleeding is 1.00 (95% CI 0.99-1.00), and for polyps/cancer is 0.98 (95% CI 0.92-0.99).

Conclusion

Based on our meta-analysis, CNN-dependent diagnosis of erosion/ulcer, GI bleeding, and polyps/cancer approached a high-level performance because of its high sensitivity and specificity. Therefore, future perspective, CNN has the potential to become an important assistant for the diagnosis of WCE.

SUBMITTER: Qin K

PROVIDER: S-EPMC8741689 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Esophageal variceal bleeding is a severe complication of portal hypertension with significant morbidity and mortality. Although traditional screening and grading of esophageal varices has been performed by endogastroduodenoscopy (EGD), wireless video capsule endoscopy provides a minimally invasive alternative that may improve screening and surveillance compliance.The aim of the study was to perform a systematic review and structured meta-analysis of all eligible studies to evaluate the efficacy of wireless capsule endoscopy for screening and diagnosis of esophageal varices among patients with portal hypertension.Searches of PubMed, EMBASE, Web of Science, and the Cochrane Library databases were performed through December 2015. Bivariate and hierarchical models were used to compute the pooled sensitivity and specificity, and to plot the summary receiver operating characteristics curve with summary point and corresponding 95% confidence region. Bias of included studies was assessed using the quality assessment of diagnostic accuracy studies-2.Seventeen studies from 2005 to 2015 were included in this meta-analysis (n=1328). The diagnostic accuracy of wireless capsule endoscopy in the diagnosis of esophageal varices was 90% [95% confidence interval (CI), 0.88-0.93]. The diagnostic pooled sensitivity and specificity were 83% (95% CI, 0.76-0.89) and 85% (95% CI, 0.75-0.91), respectively. The diagnostic accuracy of wireless capsule endoscopy for the grading of medium to large varices was 92% (95% CI, 0.90-0.94). The pooled sensitivity and specificity were 72% (95% CI, 0.54-0.85) and 91% (95% CI, 0.86-0.94), respectively, for the grading of medium to large varices. The use of capsule demonstrated only mild adverse events. A sensitivity analysis limited to only high quality studies revealed similar results.Wireless esophageal capsule endoscopy is well tolerated and safe in patients with liver cirrhosis and suspicion of portal hypertension. The sensitivity of capsule endoscopy is not currently sufficient to replace EGD as a first exploration in these patients, but given its high accuracy, it may have a role in cases of refusal or contraindication to EGD.

Project description:BackgroundCapsule endoscopy (CE) is safe and widely accepted for small bowel (SB) investigation and an alternative to colonoscopy in specific clinical circumstances. As the capsule is orally ingested, the potential risk of aspiration is undoubtedly a constant concern among clinicians. However, it is a rare occurrence and often reported as isolated cases. Therefore, this review systematically compiles all the available data on capsule aspiration in the literature with an aim to provide an update on this complication of CE.MethodsA systematic literature search was performed on PubMed with the search terms 'capsule endoscopy' AND 'aspiration', searched as keywords and MeSH. All observational cohort studies that reported aspiration among complications/outcomes, case reports and series on capsule aspiration were included. Manual citation search was performed. Two extractors reviewed abstract and full-text and performed data extraction.ResultsWe found 95 relevant articles, and cross-checking references led to the inclusion of an additional 19 articles. We removed 57 and ended with 57 references-with 63 cases of aspirated capsules. One death was reported. The median age was 78, and there was male preponderance. The most common indication for CE was anaemia, and only aspiration of small bowel CE (SBCE) was reported. 61.9% of the aspirations were symptomatic; the most common symptom was coughing. 69.8% of capsules ended in the bronchus, but only 4 cases experienced desaturation. Thirty-two patients needed intervention for retrieval; the aspiration was self-resolved in the remaining. Only four patients had a history of dysphagia. Thirteen instances of aspiration were detected due to real-time viewing, and 24 cases from reviewing the capsule data afterwards.ConclusionsWith only 63 cases of aspirated capsules reported in the literature, this event remains rare, is safely managed, and should not discourage patients from the procedure. The importance of careful patient selection is crucial to minimize the likelihood of aspiration and capsule administration should be approached with precautions.

Project description:Several studies have evaluated the utility of double-balloon enteroscopy (DBE) and capsule endoscopy (CE) for patients with small-bowel disease showing inconsistent results. The aim of this study was to determine the sensitivity and specificity of overtube-assisted enteroscopy (OAE) as well as the diagnostic concordance between OAE and CE for small-bowel polyps and tumors.We conducted a systematic review and meta-analysis of studies in which the results of OAE were compared with the results of CE for the evaluation of small-bowel polyps and tumors. When data for surgically resected lesions were available, the histopathological results of OAE and surgical specimens were compared. The sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio for the diagnosis of small-bowel polyps and tumors were analyzed. Secondarily, the rates of diagnostic concordance and discordance between OAE and CE were calculated.There were 15 full-length studies with a total of 821 patients that met the inclusion criteria. The pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were as follows: 0.89 (95 % confidence interval [CI] 0.84 - 0.93), with heterogeneity χ(2) = 41.23 (P = 0.0002) and inconsistency (I (2)) = 66.0 %; 0.97 (95 %CI 0.95 - 0.98), with heterogeneity χ(2) = 45.27 (P = 0.07) and inconsistency (I (2)) = 69.1 %; 16.61 (95 %CI 3.74 - 73.82), with heterogeneity Cochrane's Q = 225.19 (P < 0.01) and inconsistency (I (2)) = 93.8 %; and 0.14 (95 %CI 0.05 - 0.35), with heterogeneity Cochrane's Q = 81.01 (P < .01) and inconsistency (I (2)) = 82.7 %, respectively. A summary receiver operating characteristic curve (SROC) curve was constructed, and the area under the curve (AUC) was 0.97.OAE is an accurate test for the detection of small-bowel polyps and tumors. OAE and CE have a high diagnostic concordance rate for small-bowel polyps and tumors. This study was registered in the PROSPERO international database (www.crd.york.ac.uk/prospero/) with the study number CRD42015016000.