RESUMEN
The purpose of the study was to assess the performance of readers in diagnosing thoracic anomalies on standard chest radiographs (CXRs) with and without a deep-learning-based AI tool (Rayvolve) and to evaluate the standalone performance of Rayvolve in detecting thoracic pathologies on CXRs. This retrospective multicentric study was conducted in two phases. In phase 1, nine readers independently reviewed 900 CXRs from imaging group A and identified thoracic abnormalities with and without AI assistance. A consensus from three radiologists served as the ground truth. In phase 2, the standalone performance of Rayvolve was evaluated on 1500 CXRs from imaging group B. The average values of AUC across the readers significantly increased by 15.94%, with AI-assisted reading compared to unaided reading (0.88 ± 0.01 vs. 0.759 ± 0.07, p < 0.001). The time taken to read the CXRs decreased significantly, by 35.81% with AI assistance. The average values of sensitivity and specificity across the readers increased significantly by 11.44% and 2.95% with AI-assisted reading compared to unaided reading (0.857 ± 0.02 vs. 0.769 ± 0.02 and 0.974 ± 0.01 vs. 0.946 ± 0.01, p < 0.001). From the standalone perspective, the AI model achieved an average sensitivity, specificity, PPV, and NPV of 0.964, 0.844, 0.757, and 0.9798. The speed and performance of the readers improved significantly with AI assistance.
RESUMEN
Pregnancy-related complications (PRC) impact maternal and fetal morbidity and mortality and place a huge burden on healthcare systems. Thus, effective diagnostic screening strategies are crucial. Currently, national and international guidelines define patients at low risk of PRC exclusively based on their history, thus excluding the possibility of identifying patients with de novo risk (patients without a history of disease), which represents most women. In this setting, previous studies have underlined the potential contribution of non-coding RNAs (ncRNAs) to detect patients at risk of PRC. However, placenta biopsies or cord blood samples are required, which are not simple procedures. Our review explores the potential of ncRNAs in biofluids (fluids that are excreted, secreted, or developed because of a physiological or pathological process) as biomarkers for identifying patients with low-risk pregnancies. Beyond the regulatory roles of ncRNAs in placental development and vascular remodeling, we investigated their specific expressions in biofluids to determine favorable pregnancy outcomes as well as the most frequent pathologies of pregnant women. We report distinct ncRNA panels associated with PRC based on omics technologies and subsequently define patients at low risk. We present a comprehensive analysis of ncRNA expression in biofluids, including those using next-generation sequencing, shedding light on their predictive value in clinical practice. In conclusion, this paper underscores the emerging significance of ncRNAs in biofluids as promising biomarkers for risk stratification in PRC. The investigation of ncRNA expression patterns and their potential clinical applications is of diagnostic, prognostic, and theragnostic value and paves the way for innovative approaches to improve prenatal care and maternal and fetal outcomes.