Dataset Information

Protocol for a systematic review and meta-analysis of the diagnostic accuracy of artificial intelligence for grading of ophthalmology imaging modalities.

ABSTRACT:

Background

With the rise of artificial intelligence (AI) in ophthalmology, the need to define its diagnostic accuracy is increasingly important. The review aims to elucidate the diagnostic accuracy of AI algorithms in screening for all ophthalmic conditions in patient care settings that involve digital imaging modalities, using the reference standard of human graders.

Methods

This is a systematic review and meta-analysis. A literature search will be conducted on Ovid MEDLINE, Ovid EMBASE, and Wiley Cochrane CENTRAL from January 1, 2000, to December 20, 2021. Studies will be selected via screening the titles and abstracts, followed by full-text screening. Articles that compare the results of AI-graded ophthalmic images with results from human graders as a reference standard will be included; articles that do not will be excluded. The systematic review software DistillerSR will be used to automate part of the screening process as an adjunct to human reviewers. After the full-text screening, data will be extracted from each study via the categories of study characteristics, patient information, AI methods, intervention, and outcomes. Risk of bias will be scored using Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) by two trained independent reviewers. Disagreements at any step will be addressed by a third adjudicator. The study results will include summary receiver operating characteristic (sROC) curve plots as well as pooled sensitivity and specificity of artificial intelligence for detection of any ophthalmic conditions based on imaging modalities compared to the reference standard. Statistics will be calculated in the R statistical software.

Discussion

This study will provide novel insights into the diagnostic accuracy of AI in new domains of ophthalmology that have not been previously studied. The protocol also outlines the use of an AI-based software to assist in article screening, which may serve as a reference for improving the efficiency and accuracy of future large systematic reviews.

Trial registration

PROSPERO, CRD42021274441.

SUBMITTER: Cao J

PROVIDER: S-EPMC9281030 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:IntroductionStandards for Reporting of Diagnostic Accuracy Study (STARD) was developed to improve the completeness and transparency of reporting in studies investigating diagnostic test accuracy. However, its current form, STARD 2015 does not address the issues and challenges raised by artificial intelligence (AI)-centred interventions. As such, we propose an AI-specific version of the STARD checklist (STARD-AI), which focuses on the reporting of AI diagnostic test accuracy studies. This paper describes the methods that will be used to develop STARD-AI.Methods and analysisThe development of the STARD-AI checklist can be distilled into six stages. (1) A project organisation phase has been undertaken, during which a Project Team and a Steering Committee were established; (2) An item generation process has been completed following a literature review, a patient and public involvement and engagement exercise and an online scoping survey of international experts; (3) A three-round modified Delphi consensus methodology is underway, which will culminate in a teleconference consensus meeting of experts; (4) Thereafter, the Project Team will draft the initial STARD-AI checklist and the accompanying documents; (5) A piloting phase among expert users will be undertaken to identify items which are either unclear or missing. This process, consisting of surveys and semistructured interviews, will contribute towards the explanation and elaboration document and (6) On finalisation of the manuscripts, the group's efforts turn towards an organised dissemination and implementation strategy to maximise end-user adoption.Ethics and disseminationEthical approval has been granted by the Joint Research Compliance Office at Imperial College London (reference number: 19IC5679). A dissemination strategy will be aimed towards five groups of stakeholders: (1) academia, (2) policy, (3) guidelines and regulation, (4) industry and (5) public and non-specific stakeholders. We anticipate that dissemination will take place in Q3 of 2021.

Project description:BackgroundEndocytoscopy (EC) is a nuclei and micro-vessels visualization in real-time and can facilitate "optical biopsy" and "virtual histology" of colorectal lesions. This study aimed to investigate the significance of employing artificial intelligence (AI) in the field of endoscopy, specifically in diagnosing colorectal lesions. The research was conducted under the supervision of experienced professionals and trainees.MethodsEMBASE, PubMed, Cochrane Library, Web of Science, Chinese National Knowledge Infrastructure (CNKI) database, and other potential databases were surveyed for articles related to the EC with AI published before September 2023. RevMan (5.40), Stata (14.0), and R software (4.1.0) were used for statistical assessment. Studies that measured the accuracy of EC using AI for colorectal lesions were included. Two authors independently assessed the selected studies and their extracted data. This included information such as the country, literature, total study population, study design, characteristics of the fundamental study and control groups, sensitivity, number of samples, assay methodology, specificity, true positives or negatives, and false positives or negatives. The diagnostic accuracy of EC by AI was determined by a bivariate random-effects model, avoiding a high heterogeneity effect. The ANOVA model was employed to determine the more effective approach.ResultsA total of 223 studies were reviewed; 8 articles were selected that included 2984 patients (4241 lesions) for systematic review and meta-analysis. AI assessed 4069 lesions; experts diagnosed 3165 and 5014 by trainees. AI demonstrated high accuracy, sensitivity, and specificity levels in detecting colorectal lesions, with values of 0.93 (95% CI: 0.90, 0.95) and 0.94 (95% CI: 0.73, 0.99). Expert diagnosis was 0.90 (95% CI: 0.85, 0.94), 0.87 (95% CI: 0.78, 0.93), and trainee diagnosis was 0.74 (95% CI: 0.67, 0.79), 0.72 (95% CI: 0.62, 0.80). With the EC by AI, the AUC from SROC was 0.95 (95% CI: 0.93, 0.97), therefore classified as excellent category, expert showed 0.95 (95% CI: 0.93, 0.97), and the trainee had 0.79 (95% CI: 0.75, 0.82). The superior index from the ANOVA model was 4.00 (1.15,5.00), 2.00 (1.15,5.00), and 0.20 (0.20,0.20), respectively. The examiners conducted meta-regression and subgroup analyses to evaluate the presence of heterogeneity. The findings of these investigations suggest that the utilization of NBI technology was correlated with variability in sensitivity and specificity. There was a lack of solid evidence indicating the presence of publishing bias.ConclusionsThe present findings indicate that using AI in EC can potentially enhance the efficiency of diagnosing colorectal abnormalities. As a valuable instrument, it can enhance prognostic outcomes in ordinary EC procedures, exhibiting superior diagnostic accuracy compared to trainee-level endoscopists and demonstrating comparability to expert endoscopists. The research is subject to certain constraints, namely a limited number of clinical investigations and variations in the methodologies used for identification. Consequently, it is imperative to conduct comprehensive and extensive research to enhance the precision of diagnostic procedures.

Project description:BackgroundArtificial Intelligence (AI) holds considerable promise for diagnostics in the field of gastroenterology. This systematic review and meta-analysis aims to assess the diagnostic accuracy of AI models compared with the gold standard of experts and histopathology for the diagnosis of various gastrointestinal (GI) luminal pathologies including polyps, neoplasms, and inflammatory bowel disease.MethodsWe searched PubMed, CINAHL, Wiley Cochrane Library, and Web of Science electronic databases to identify studies assessing the diagnostic performance of AI models for GI luminal pathologies. We extracted binary diagnostic accuracy data and constructed contingency tables to derive the outcomes of interest: sensitivity and specificity. We performed a meta-analysis and hierarchical summary receiver operating characteristic curves (HSROC). The risk of bias was assessed using Quality Assessment for Diagnostic Accuracy Studies-2 (QUADAS-2) tool. Subgroup analyses were conducted based on the type of GI luminal disease, AI model, reference standard, and type of data used for analysis. This study is registered with PROSPERO (CRD42021288360).FindingsWe included 73 studies, of which 31 were externally validated and provided sufficient information for inclusion in the meta-analysis. The overall sensitivity of AI for detecting GI luminal pathologies was 91.9% (95% CI: 89.0-94.1) and specificity was 91.7% (95% CI: 87.4-94.7). Deep learning models (sensitivity: 89.8%, specificity: 91.9%) and ensemble methods (sensitivity: 95.4%, specificity: 90.9%) were the most commonly used models in the included studies. Majority of studies (n = 56, 76.7%) had a high risk of selection bias while 74% (n = 54) studies were low risk on reference standard and 67% (n = 49) were low risk for flow and timing bias.InterpretationThe review suggests high sensitivity and specificity of AI models for the detection of GI luminal pathologies. There is a need for large, multi-center trials in both high income countries and low- and middle- income countries to assess the performance of these AI models in real clinical settings and its impact on diagnosis and prognosis.Systematic review registration[https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=288360], identifier [CRD42021288360].