ABSTRACT
BACKGROUND: Surgical site infection (SSI) is a health-threatening complication following Caesarean section (CS); however, to our knowledge, there is no worldwide estimate of the burden of post-CS SSIs. Therefore, this systematic review and meta-analysis aimed to estimate the global and regional incidence of post-CS SSI and its associated factors. METHODS: We systematically searched international scientific databases for observational studies published from January 2000 to March 2023, without language or geographical restrictions. The pooled global incidence rate was estimated using a random-effects meta-analysis (REM), and then stratified by World Health Organization (WHO)-defined regions as well as by socio-demographic and study characteristics. We also analysed causative pathogens and associated risk factors of SSIs using REM. We assessed heterogeneity with I2. RESULTS: We included 180 eligible studies (207 datasets) involving 2,188,242 participants from 58 countries. The pooled global incidence of post-CS SSI was 5.63% (95% CI, 5.18%-6.11%). The highest and lowest post-CS SSI incidences were estimated for African (11.91%, 9.67-14.34%), and North-America (3.87%, 3.02-4.83%) regions, respectively. The incidence was significantly higher in countries with lower levels of income and human development index. The pooled incidence estimates have steadily increased over time, with the highest incidence rate during the COVID-19 pandemic (2019-2023). Staphylococcus aureus and Escherichia coli were the most prevalent pathogens. Several risk factors were identified. CONCLUSION: We found an increasing and substantial burden from post-CS SSIs, especially in low-income countries. Further research, greater awareness, and the development of effective prevention and management strategies are warranted to reduce post-CS SSIs.
ABSTRACT
Medical machine learning frameworks have received much attention in recent years. The recent COVID-19 pandemic was also accompanied by a surge in proposed machine learning algorithms for tasks such as diagnosis and mortality prognosis. Machine learning frameworks can be helpful medical assistants by extracting data patterns that are otherwise hard to detect by humans. Efficient feature engineering and dimensionality reduction are major challenges in most medical machine learning frameworks. Autoencoders are novel unsupervised tools that can perform data-driven dimensionality reduction with minimum prior assumptions. This study, in a novel approach, investigated the predictive power of latent representations obtained from a hybrid autoencoder (HAE) framework combining variational autoencoder (VAE) characteristics with mean squared error (MSE) and triplet loss for forecasting COVID-19 patients with high mortality risk in a retrospective framework. Electronic laboratory and clinical data of 1474 patients were used in the study. Logistic regression with elastic net regularization (EN) and random forest (RF) models were used as final classifiers. Moreover, we also investigated the contribution of utilized features towards latent representations via mutual information analysis. HAE Latent representations model achieved decent performance with an area under ROC curve of 0.921 (±0.027) and 0.910 (±0.036) with EN and RF predictors, respectively, over the hold-out data in comparison with the raw (AUC EN: 0.913 (±0.022); RF: 0.903 (±0.020)) models. The study aims to provide an interpretable feature engineering framework for the medical environment with the potential to integrate imaging data for efficient feature engineering in rapid triage and other clinical predictive models.
Subject(s)
COVID-19 , Pandemics , Humans , Retrospective Studies , Prognosis , Machine LearningABSTRACT
Introduction: The gold standard for diagnosis of coronavirus disease 2019 (COVID-19) is detecting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) by reverse transcription polymerase chain reaction (RT-PCR), which is expensive, time-consuming and may result in false-negative results. Serological tests can be employed for RT-PCR negative patients, contact tracing, determining the probability of protection against re-infection, and seroepidemiological studies.Areas covered: The main methodologies of serology-based tests for the detection of SARS-CoV-2 including enzyme-linked immunosorbent assays (ELISAs), chemiluminescent immunoassays (CLIAs) and lateral flow immunoassays (LFIAs) were reviewed and their diagnostic performances were compared. Herein, a literature review on the databases of PubMed, Scopus and Google Scholar between January 1, 2020 and June 30, 2020 based on the main serological methods for COVID-19 detection with the focus on comparative experiments was performed. The review was updated on December 31, 2020.Expert opinion: Serology testing could be considered as a part of diagnostic panel two-week post symptom onset. Higher sensitivity for serology-based tests could be achieved by determining combined IgG/IgM titers. Furthermore, higher sensitive serological test detecting neutralization antibody could be developed by targeting spike (S) antigen. It was also demonstrated that the sensitivity of ELISA/CLIA-based methods are higher than LFIA devices.