20 February 2020 in JMIR Medical Informatics
Background: Privacy restrictions limit access to protected patient-derived health information for research purposes. Consequently, data anonymization is required to allow researchers data access for initial analysis before granting institutional review board approval. A system installed and activated at our institution enables synthetic data generation that mimics data from real electronic medical records, wherein only fictitious patients are listed. Objective: This paper aimed to validate the results obtained when analyzing synthetic structured data for medical research. A comprehensive validation process concerning meaningful clinical questions and various types of data was conducted to assess the accuracy and precision of statistical estimates derived from synthetic patient data. Methods: A cross-hospital project was conducted to validate results obtained from synthetic data produced for five contemporary studies on various topics. For each study, results derived from synthetic data were compared with those based on real data. In addition, repeatedly generated synthetic datasets were used to estimate the bias and stability of results obtained from synthetic data. Results: This study demonstrated that results derived from synthetic data were predictive of results from real data. When the number of patients was large relative to the number of variables used, highly accurate and strongly consistent results were observed between synthetic and real data. For studies based on smaller populations that accounted for confounders and modifiers by multivariate models, predictions were of moderate accuracy, yet clear trends were correctly observed. Conclusions: The use of synthetic structured data provides a close estimate to real data results and is thus a powerful tool in shaping research hypotheses and accessing estimated analyses, without risking patient privacy. Synthetic data enable broad access to data (eg, for out-of-organization researchers), and rapid, safe, and repeatable analysis of data in hospitals or other health organizations where patient privacy is a primary value.
1 March 1994 in Infection
12 January 2022 in PLOS Pathogens
In-depth analysis of SARS-CoV-2 quasispecies is pivotal for a thorough understating of its evolution during infection. The recent deployment of COVID-19 vaccines, which elicit protective anti-spike neutralizing antibodies, has stressed the importance of uncovering and characterizing SARS-CoV-2 variants with mutated spike proteins. Sequencing databases have allowed to follow the spread of SARS-CoV-2 variants that are circulating in the human population, and several experimental platforms were developed to study these variants. However, less is known about the SARS-CoV-2 variants that are developed in the respiratory system of the infected individual. To gain further insight on SARS-CoV-2 mutagenesis during natural infection, we preformed single-genome sequencing of SARS-CoV-2 isolated from nose-throat swabs of infected individuals. Interestingly, intra-host SARS-CoV-2 variants with mutated S genes or N genes were detected in all individuals who were analyzed. These intra-host variants were present in low frequencies in the swab samples and were rarely documented in current sequencing databases. Further examination of representative spike variants identified by our analysis showed that these variants have impaired infectivity capacity and that the mutated variants showed varied sensitivity to neutralization by convalescent plasma and to plasma from vaccinated individuals. Notably, analysis of the plasma neutralization activity against these variants showed that the L1197I mutation at the S2 subunit of the spike can affect the plasma neutralization activity. Together, these results suggest that SARS-CoV-2 intra-host variants should be further analyzed for a more thorough characterization of potential circulating variants.
2 July 2015 in The Journal of Maternal-Fetal & Neonatal Medicine
Our aim was to evaluate postpartum glycemic control in women with pregestational diabetes and to assess contributing factors. Retrospective data collection from an electronic database on a cohort of Israeli women at Clalit Healthcare Services with pregestational diabetes who gave birth in 2008-2011, including data on HbA(1C), prescription fillings, and socio-demographics. HbA(1C) level was assessed during a 2 year time-period, from periconception until one-year postpartum. There were 180 deliveries to 166 women. Compared with the periconception period, the HbA(1C) level improved significantly during the last 6 months of pregnancy (6.7% versus 6.0%, p < 0.05). However, there was rapid continuous deterioration in glycemic control in the postpartum period with median HbA(1C) = 6.9% in the first 6 months postpartum and 7.2% in the late postpartum period (p < 0.05). One-year postpartum 107 women (59.4%) had suboptimal care (defined as HbA(1C) > 7.0% or no test). In the multiple logistic regression analysis, the only significant predictor of 1-year postpartum suboptimal care was suboptimal periconception care, OR = 6.1 (95% CI 3.15-11.84, p = 0.001). Postpartum glycemic control deteriorated rapidly despite excellent control in most women in the last 6 month of pregnancy. More intensive and targeted intervention is needed in order to optimize postpartum care of diabetic patients.
13 January 2026 in Machine Learning: Health
While circadian phenotypes of paroxysmal atrial fibrillation (AF) have been observed in patient cohorts, differences in parameter definition and methodological limitations have left the existence and clinical relevance of distinct AF circadian phenotypes unclear. We hypothesize that paroxysmal AF comprises multiple circadian phenotypes (‘chronophenotypes’), each associated with different survival outcomes and population characteristics. We analyzed 24 h Holter recordings and clinical data from 58 995 examinations collected in 20 primary care facilities in Israel. AF episodes were detected using ArNet2, a deep learning model for AF detection trained on over 51 000 h of Holter recordings. Unsupervised hierarchical clustering was then applied to identify different chronophenotypes of AF based on the short-term AF burden. Each chronophenotype was further characterized by demographic, clinical, and treatment differences, including survival outcomes. The analysis resulted in the following distinct chronophenotypes: Nocturnal-to-Morning (12 AM–10 AM), Evening-to-Early Morning (3 PM-4 AM), and Daytime (9 AM-7 PM). These chronophenotypes were associated with different AF burden and clinical outcomes. The Evening-to-Early Morning chronophenotype was associated with a higher AF burden (
0.001) than the other paroxysmal AF chronophenotypes. However, the Nocturnal-to-Morning and Daytime chronophenotypes were associated with a higher risk of mortality; Daytime was also associated with a higher risk of heart failure. The study introduces a new approach to discovering chronophenotypes from Holter recordings. The results support the existence of distinct AF chronophenotypes, which are associated with significantly different average AF burden (
0.001) and clinical outcomes. The open-source implementation is publicly available at https://github.com/aim-lab/AFtoolkit.
27 April 2023 in Population Medicine
Background and Objective: In recent years working from home (WFH) has become more common worldwide. In spite of its advantages, its prevalence worldwide before the outbreak of the COVID-19 pandemic was relatively low. The outbreak of COVID-19 pandemic enforced millions of employees to switch to...