Use of artificial intelligence in obstetric and gynaecological diagnostics: a protocol for a systematic review and meta-analysis.

INTRODUCTION: Emerging developments in applications of artificial intelligence (AI) in healthcare offer the opportunity to improve diagnostic capabilities in obstetrics and gynaecology (O&G), ensuring early detection of pathology, optimal management and improving survival. Consensus on a robust AI healthcare framework is crucial for standardising protocols that promote data privacy and transparency, minimise bias, and ensure patient safety. Here, we describe the study protocol for a systematic review and meta-analysis to evaluate current applications of AI in O&G diagnostics with consideration of reporting standards used and their ethical implications. This protocol is written following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 checklist. METHODS AND ANALYSIS: The study objective is to explore the current application of AI in O&G diagnostics and assess the reporting standards used in these studies. Electronic bibliographic databases MEDLINE, EMBASE and Cochrane will be searched. Study selection, data extraction and subsequent narrative synthesis and meta-analyses will be carried out following the PRISMA-P guidelines. Included papers will be English-language full-text articles from May 2015 to March 2024, which provide original data, as AI has been redefined in recent literature. Papers must use AI as the predictive method, focusing on improving O&G diagnostic outcomes.We will evaluate the reporting standards including the risk of bias, lack of transparency and consider the ethical implications and potential harm to patients. Outcome measures will involve assessing the included studies against gold-standard criteria for robustness of model development (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis, model predictive performance, model risk of bias and applicability (Prediction model Risk Of Bias Assessment Tool and study reporting (Consolidated Standards of Reporting Trials-AI) guidance. ETHICS AND DISSEMINATION: Ethical approval is not required for this systematic review. Findings will be shared through peer-reviewed publications. There will be no patient or public involvement in this study. PROSPERO REGISTRATION NUMBER: CRD42022357024 .

Evaluating Different Quantitative Shear Wave Parameters of Ultrasound Elastography in the Diagnosis of Lymph Node Malignancies: A Systematic Review and Meta-Analysis.

Shear wave elastography (SWE) has shown promise in distinguishing lymph node malignancies. However, the diagnostic accuracies of various SWE parameters that quantify tissue stiffness are yet to be demonstrated. To evaluate the pooled diagnostic accuracy of different SWE parameters for differentiating lymph node malignancies, we conducted a systematic screening of four databases using the PRISMA guidelines. Lymph node biopsy was adopted as the reference standard. Emax (maximum stiffness), Emean (mean stiffness), Emin (minimum stiffness), and Esd (standard deviation) SWE parameters were subjected to separate meta-analyses. A sub-group analysis comparing the use of Emax in cervical (including thyroid) and axillary lymph node malignancies was also conducted. Sixteen studies were included in this meta-analysis. Emax and Esd demonstrated the highest pooled sensitivity (0.78 (95% CI: 0.69-0.87); 0.78 (95% CI: 0.68-0.87)), while Emean demonstrated the highest pooled specificity (0.93 (95% CI: 0.88-0.98)). From the sub-group analysis, the diagnostic performance did not differ significantly in cervical and axillary LN malignancies. In conclusion, SWE is a promising adjunct imaging technique to conventional ultrasonography in the diagnosis of lymph node malignancy. SWE parameters of Emax and Esd have been identified as better choices of parameters for screening clinical purposes.