A systematic literature review and meta-analysis of spectral CT compared to scintigraphy in the diagnosis of acute and chronic pulmonary embolisms.

PURPOSE: To examine the diagnostic accuracy of spectral CT pulmonary angiography (S-CTPA) using ventilation-perfusions lung scintigraphy (V/Q-scan) as a reference standard in the diagnosis of acute or chronic pulmonary embolism (APE/CPE) and chronic thromboembolic pulmonary hypertension (CTEPH). METHODS: PubMed, Embase, Scopus, and Web of Science were searched for the period from 1 Jan 2006 to 7 Feb 2019; eligible studies had > 10 patients over 18 years old, a diagnostic outcome of PE or CTEPH, and used V/Q scan as a reference standard. Bias and applicability were assessed using QUADAS-2 tools. Sensitivities, specificities, and predictive values were noted or calculated from available information. Meta-analysis employed a fixed-effects model of Mantel and Haenszel. Heterogeneity was assessed with I-squared statistics. RESULTS: Four hundred ninety-three unique records were identified. Following screening by title, 53 studies were included in the abstract and full-text assessment. A total of six articles were included; four were suitable for a meta-analysis. Pooled sensitivity was 94.2% (95% CI, 88.3-100%), pooled specificity was 88.5% (95% CI, 81.3-95.6%), and positive and negative predictive values were 87.8% (95% CI, 80.3-95.4%) and 94.5% (95% CI, 89.3-99.7%), respectively. CONCLUSION: Data on S-CTPA for PE/CTEPH remains promising, but limited; only small studies with methodological issues are available. Evidence is best for CPE/CTEPH whereas no firm conclusions are possible for APE. There is a need for larger, prospective studies with a robust composite reference standard including state-of-the-art CTPA and V/Q-scans. KEY POINTS: • S-CTPA has high sensitivity and specificity for perfusion defects in patients with PE or CPETH. • Methodological issues and diversity of reference standards were found in the small number of included studies. • There is a need for larger prospective studies with more robust composite reference standards.

Convolutional neural network performance compared to radiologists in detecting intracranial hemorrhage from brain computed tomography: A systematic review and meta-analysis.

PURPOSE: To compare the diagnostic accuracy of convolutional neural networks (CNN) with radiologists as the reference standard in the diagnosis of intracranial hemorrhages (ICH) with non contrast computed tomography of the cerebrum (NCTC). METHODS: PubMed, Embase, Scopus, and Web of Science were searched for the period from 1 January 2012 to 20 July 2020; eligible studies included patients with and without ICH as the target condition undergoing NCTC, studies had deep learning algorithms based on CNNs and radiologists reports as the minimum reference standard. Pooled sensitivities, specificities and a summary receiver operating characteristics curve (SROC) were employed for meta-analysis. RESULTS: 5,119 records were identified through database searching. Title-screening left 47 studies for full-text assessment and 6 studies for meta-analysis. Comparing the CNN performance to reference standards in the retrospective studies found a pooled sensitivity of 96.00% (95% CI: 93.00% to 97.00%), pooled specificity of 97.00% (95% CI: 90.00% to 99.00%) and SROC of 98.00% (95% CI: 97.00% to 99.00%), and combining retrospective and studies with external datasets found a pooled sensitivity of 95.00% (95% CI: 91.00% to 97.00%), pooled specificity of 96.00% (95% CI: 91.00% to 98.00%) and a pooled SROC of 98.00% (95% CI: 97.00% to 99.00%). CONCLUSION: This review found the diagnostic performance of CNNs to be equivalent to that of radiologists for retrospective studies. Out-of-sample external validation studies pooled with retrospective studies found CNN performance to be slightly worse. There is a critical need for studies with a robust reference standard and external data-set validation.