ABSTRACT
The diagnosis of Moyamoya disease (MMD) relies heavily on imaging, which could benefit from standardized machine learning tools. This study aims to evaluate the diagnostic efficacy of deep learning (DL) algorithms for MMD by analyzing sensitivity, specificity, and the area under the curve (AUC) compared to expert consensus. We conducted a systematic search of PubMed, Embase, and Web of Science for articles published from inception to February 2024. Eligible studies were required to report diagnostic accuracy metrics such as sensitivity, specificity, and AUC, excluding those not in English or using traditional machine learning methods. Seven studies were included, comprising a sample of 4,416 patients, of whom 1,358 had MMD. The pooled sensitivity for common and random effects models was 0.89 (95% CI: 0.85 to 0.92) and 0.92 (95% CI: 0.85 to 0.96), respectively. The pooled specificity was 0.89 (95% CI: 0.86 to 0.91) in the common effects model and 0.91 (95% CI: 0.75 to 0.97) in the random effects model. Two studies reported the AUC alongside their confidence intervals. A meta-analysis synthesizing these findings aggregated a mean AUC of 0.94 (95% CI: 0.92 to 0.96) for common effects and 0.89 (95% CI: 0.76 to 1.02) for random effects models. Deep learning models significantly enhance the diagnosis of MMD by efficiently extracting and identifying complex image patterns with high sensitivity and specificity. Trial registration: CRD42024524998 https://www.crd.york.ac.uk/prospero/displayrecord.php?RecordID=524998.
Subject(s)
Deep Learning , Moyamoya Disease , Moyamoya Disease/diagnosis , Humans , Algorithms , Sensitivity and SpecificityABSTRACT
INTRODUCTION: Several observational studies have evaluated the effects of pre-operative steroids (STER) for transsphenoidal pituitary removal in patients with an intact hypothalamus-pituitary-adrenal axis. However, a consensus built upon randomized studies has not been previously performed. PURPOSE: To comprehensively evaluate the clinical effects of patients receiving STER when compared to those not receiving steroids (NOSTER) in transsphenoidal pituitary resection, a meta-analysis of randomized clinical trials (RCT) was conducted. METHODS: A systematic review of the literature of RCTs comparing STER and NOSTER was performed in accordance with the PRISMA guidelines. Databases, including PUBMED, Cochrane Library, Web of Science, and Embase were queried. The primary outcomes were adrenal insufficiency (AI) and diabetes insipidus (DI) post-operatively. RESULTS: A total of 4 final studies were included, in which 530 total patients were analyzed. The meta-analysis suggested that there was no significant difference between STER and NOSTER groups post-operatively related to transient AI (RR= 0.83, 95% CI [0.51-1.35], p = 0.45; I² = 52%), permanent AI (RR= 0.97, 95% CI [0.41-2.31], p = 0.95; I² = 0%), and permanent DI (RR= 0.62, 95% CI [0.16-2.33], p = 0.48; I² = 0%). Nevertheless, STER group had significantly lower rates of transient DI (RR= 0.60, 95% CI [0.38-0.95], p = 0.03; I² = 5%), and post-op hyponatremia (RR = 0.49, 95% CI [0.28-0.87], p = 0.02; I² = 0%). CONCLUSION: This study demonstrates evidence from RCTs that patients receiving pre-operative STER are both safe and effective pre-operatively for resection of pituitary adenomas with an intact hypothalamus-pituitary-adrenal axis.