SASC: A simple approach to synthetic cohorts for generating longitudinal observational patient cohorts from COVID-19 clinical data.
Ontology highlight
ABSTRACT:
SUBMITTER: Khorchani T
PROVIDER: S-EPMC8825316 | biostudies-literature | 2022 Apr
REPOSITORIES: biostudies-literature
Publications
SASC: A simple approach to synthetic cohorts for generating longitudinal observational patient cohorts from COVID-19 clinical data.
Khorchani Takoua T Gadiya Yojana Y Witt Gesa G Lanzillotta Delia D Claussen Carsten C Zaliani Andrea A
Patterns (New York, N.Y.) 20220209 4
One of the impacts of the coronavirus disease 2019 (COVID-19) pandemic has been a push for researchers to better exploit synthetic data and accelerate the design, analysis, and modeling of clinical trials. The unprecedented clinical efforts caused by COVID-19's emergence will certainly boost future robust and innovative approaches of statistical sciences applied to clinical fields. Here, we report the development of SASC, a simple but efficient approach to generate COVID-19-related synthetic cli ...[more]
Similar Datasets
Project description:Getting access to administrative health data for research purposes is a difficult and time-consuming process due to increasingly demanding privacy regulations. An alternative method for sharing administrative health data would be to share synthetic datasets where the records do not correspond to real individuals, but the patterns and relationships seen in the data are reproduced. This paper assesses the feasibility of generating synthetic administrative health data using a recurrent deep learning model. Our data comes from 120,000 individuals from Alberta Health's administrative health database. We assess how similar our synthetic data is to the real data using utility assessments that assess the structure and general patterns in the data as well as by recreating a specific analysis in the real data commonly applied to this type of administrative health data. We also assess the privacy risks associated with the use of this synthetic dataset. Generic utility assessments that used Hellinger distance to quantify the difference in distributions between real and synthetic datasets for event types (0.027), attributes (mean 0.0417), Markov transition matrices (order 1 mean absolute difference: 0.0896, sd: 0.159; order 2: mean Hellinger distance 0.2195, sd: 0.2724), the Hellinger distance between the joint distributions was 0.352, and the similarity of random cohorts generated from real and synthetic data had a mean Hellinger distance of 0.3 and mean Euclidean distance of 0.064, indicating small differences between the distributions in the real data and the synthetic data. By applying a realistic analysis to both real and synthetic datasets, Cox regression hazard ratios achieved a mean confidence interval overlap of 68% for adjusted hazard ratios among 5 key outcomes of interest, indicating synthetic data produces similar analytic results to real data. The privacy assessment concluded that the attribution disclosure risk associated with this synthetic dataset was substantially less than the typical 0.09 acceptable risk threshold. Based on these metrics our results show that our synthetic data is suitably similar to the real data and could be shared for research purposes thereby alleviating concerns associated with the sharing of real data in some circumstances.
| S-EPMC10034254 | biostudies-literature
Project description:There is a growing demand for the uptake of modern artificial intelligence technologies within healthcare systems. Many of these technologies exploit historical patient health data to build powerful predictive models that can be used to improve diagnosis and understanding of disease. However, there are many issues concerning patient privacy that need to be accounted for in order to enable this data to be better harnessed by all sectors. One approach that could offer a method of circumventing privacy issues is the creation of realistic synthetic data sets that capture as many of the complexities of the original data set (distributions, non-linear relationships, and noise) but that does not actually include any real patient data. While previous research has explored models for generating synthetic data sets, here we explore the integration of resampling, probabilistic graphical modelling, latent variable identification, and outlier analysis for producing realistic synthetic data based on UK primary care patient data. In particular, we focus on handling missingness, complex interactions between variables, and the resulting sensitivity analysis statistics from machine learning classifiers, while quantifying the risks of patient re-identification from synthetic datapoints. We show that, through our approach of integrating outlier analysis with graphical modelling and resampling, we can achieve synthetic data sets that are not significantly different from original ground truth data in terms of feature distributions, feature dependencies, and sensitivity analysis statistics when inferring machine learning classifiers. What is more, the risk of generating synthetic data that is identical or very similar to real patients is shown to be low.
| S-EPMC7653933 | biostudies-literature
Project description:IntroductionFederated Learning (FL) is a decentralised approach to training statistical models, where training is performed across multiple clients, producing one global model. Since the training data remains with each local client and is not shared or exchanged with other clients the use of FL may reduce privacy and security risks (compared to methods where multiple data sources are pooled) and can also address data access and heterogeneity problems. Synthetic data is artificially generated data that has the same structure and statistical properties as the original but that does not contain any of the original data records, therefore minimising disclosure risk. Using FL to produce synthetic data (which we refer to as "federated synthesis") has the potential to combine data from multiple clients without compromising privacy, allowing access to data that may otherwise be inaccessible in its raw format.ObjectivesThe objective was to review current research and practices for using FL to generate synthetic data and determine the extent to which research has been undertaken, the methods and evaluation practices used, and any research gaps.MethodsA scoping review was conducted to systematically map and describe the published literature on the use of FL to generate synthetic data. Relevant studies were identified through online databases and the findings are described, grouped, and summarised. Information extracted included article characteristics, documenting the type of data that is synthesised, the model architecture and the methods (if any) used to evaluate utility and privacy risk.ResultsA total of 69 articles were included in the scoping review; all were published between 2018 and 2023 with two thirds (46) in 2022. 30% (21) were focussed on synthetic data generation as the main model output (with 6 of these generating tabular data), whereas 59% (41) focussed on data augmentation. Of the 21 performing federated synthesis, all used deep learning methods (predominantly Generative Adversarial Networks) to generate the synthetic data.ConclusionsFederated synthesis is in its early days but shows promise as a method that can construct a global synthetic dataset without sharing any of the local client data. As a field in its infancy there are areas to explore in terms of the privacy risk associated with the various methods proposed, and more generally in how we measure those risks.
| S-EPMC10898505 | biostudies-literature
Project description:The recent availability of electronic health records (EHRs) have provided enormous opportunities to develop artificial intelligence (AI) algorithms. However, patient privacy has become a major concern that limits data sharing across hospital settings and subsequently hinders the advances in AI. Synthetic data, which benefits from the development and proliferation of generative models, has served as a promising substitute for real patient EHR data. However, the current generative models are limited as they only generate single type of clinical data for a synthetic patient, i.e., either continuous-valued or discrete-valued. To mimic the nature of clinical decision-making which encompasses various data types/sources, in this study, we propose a generative adversarial network (GAN) entitled EHR-M-GAN that simultaneously synthesizes mixed-type timeseries EHR data. EHR-M-GAN is capable of capturing the multidimensional, heterogeneous, and correlated temporal dynamics in patient trajectories. We have validated EHR-M-GAN on three publicly-available intensive care unit databases with records from a total of 141,488 unique patients, and performed privacy risk evaluation of the proposed model. EHR-M-GAN has demonstrated its superiority over state-of-the-art benchmarks for synthesizing clinical timeseries with high fidelity, while addressing the limitations regarding data types and dimensionality in the current generative models. Notably, prediction models for outcomes of intensive care performed significantly better when training data was augmented with the addition of EHR-M-GAN-generated timeseries. EHR-M-GAN may have use in developing AI algorithms in resource-limited settings, lowering the barrier for data acquisition while preserving patient privacy.
| S-EPMC10224668 | biostudies-literature
Project description:OBJECTIVES:The need for large studies and the types of large-scale data resources (LSDRs) are discussed along with their general scientific utility, role in aging research, and affordability. The diversification of approaches to large-scale data resourcing is described in order to facilitate their use in aging research. METHODS:The need for LSDRs is discussed in terms of (a) large sample size; (b) longitudinal design; (c) as platforms for additional investigator-initiated research projects; and (d) broad-based access to core genetic, biological, and phenotypic data. DISCUSSION:It is concluded that a "lite-touch, lo-tech, lo-cost" approach to LSDRs is a viable strategy for the development of LSDRs and would enhance the likelihood of LSDRs being established which are dedicated to the wide range of important aging-related issues.
| S-EPMC3165456 | biostudies-other
Project description:BackgroundPatients with COVID-19-related acute respiratory distress syndrome (ARDS) have been postulated to present with distinct respiratory subphenotypes. However, most phenotyping schema have been limited by sample size, disregard for temporal dynamics, and insufficient validation. We aimed to identify respiratory subphenotypes of COVID-19-related ARDS using unbiased data-driven approaches.MethodsPRoVENT-COVID was an investigator-initiated, national, multicentre, prospective, observational cohort study at 22 intensive care units (ICUs) in the Netherlands. Consecutive patients who had received invasive mechanical ventilation for COVID-19 (aged 18 years or older) served as the derivation cohort, and similar patients from two ICUs in the USA served as the replication cohorts. COVID-19 was confirmed by positive RT-PCR. We used latent class analysis to identify subphenotypes using clinically available respiratory data cross-sectionally at baseline, and longitudinally using 8-hourly data from the first 4 days of invasive ventilation. We used group-based trajectory modelling to evaluate trajectories of individual variables and to facilitate potential clinical translation. The PRoVENT-COVID study is registered with ClinicalTrials.gov, NCT04346342.FindingsBetween March 1, 2020, and May 15, 2020, 1007 patients were admitted to participating ICUs in the Netherlands, and included in the derivation cohort. Data for 288 patients were included in replication cohort 1 and 326 in replication cohort 2. Cross-sectional latent class analysis did not identify any underlying subphenotypes. Longitudinal latent class analysis identified two distinct subphenotypes. Subphenotype 2 was characterised by higher mechanical power, minute ventilation, and ventilatory ratio over the first 4 days of invasive mechanical ventilation than subphenotype 1, but PaO2/FiO2, pH, and compliance of the respiratory system did not differ between the two subphenotypes. 185 (28%) of 671 patients with subphenotype 1 and 109 (32%) of 336 patients with subphenotype 2 had died at day 28 (p=0·10). However, patients with subphenotype 2 had fewer ventilator-free days at day 28 (median 0, IQR 0-15 vs 5, 0-17; p=0·016) and more frequent venous thrombotic events (109 [32%] of 336 patients vs 176 [26%] of 671 patients; p=0·048) compared with subphenotype 1. Group-based trajectory modelling revealed trajectories of ventilatory ratio and mechanical power with similar dynamics to those observed in latent class analysis-derived trajectory subphenotypes. The two trajectories were: a stable value for ventilatory ratio or mechanical power over the first 4 days of invasive mechanical ventilation (trajectory A) or an upward trajectory (trajectory B). However, upward trajectories were better independent prognosticators for 28-day mortality (OR 1·64, 95% CI 1·17-2·29 for ventilatory ratio; 1·82, 1·24-2·66 for mechanical power). The association between upward ventilatory ratio trajectories (trajectory B) and 28-day mortality was confirmed in the replication cohorts (OR 4·65, 95% CI 1·87-11·6 for ventilatory ratio in replication cohort 1; 1·89, 1·05-3·37 for ventilatory ratio in replication cohort 2).InterpretationAt baseline, COVID-19-related ARDS has no consistent respiratory subphenotype. Patients diverged from a fairly homogenous to a more heterogeneous population, with trajectories of ventilatory ratio and mechanical power being the most discriminatory. Modelling these parameters alone provided prognostic value for duration of mechanical ventilation and mortality.FundingAmsterdam UMC.
| S-EPMC8510633 | biostudies-literature
Project description:Synthetic data is becoming a valuable tool for computational pathologists, aiding in tasks like data augmentation and addressing data scarcity and privacy. However, its use necessitates careful planning and evaluation to prevent the creation of clinically irrelevant artifacts.This manuscript introduces a comprehensive pipeline for generating and evaluating synthetic pathology data using a diffusion model. The pipeline features a multifaceted evaluation strategy with an integrated explainability procedure, addressing two key aspects of synthetic data use in the medical domain.The evaluation of the generated data employs an ensemble-like approach. The first step includes assessing the similarity between real and synthetic data using established metrics. The second step involves evaluating the usability of the generated images in deep learning models accompanied with explainable AI methods. The final step entails verifying their histopathological realism through questionnaires answered by professional pathologists. We show that each of these evaluation steps are necessary as they provide complementary information on the generated data's quality.The pipeline is demonstrated on the public GTEx dataset of 650 Whole Slide Images (WSIs), including five different tissues. An equal number of tiles from each tissue are generated and their reliability is assessed using the proposed evaluation pipeline, yielding promising results.In summary, the proposed workflow offers a comprehensive solution for generative AI in digital pathology, potentially aiding the community in their transition towards digitalization and data-driven modeling.
| S-EPMC11574254 | biostudies-literature
Project description:Low rates of vaccination, emergence of novel variants of SARS-CoV-2, and increasing transmission relating to seasonal changes and relaxation of mitigation measures leave many US communities at risk for surges of COVID-19 that might strain hospital capacity, as in previous waves. The trajectories of COVID-19 hospitalizations differ across communities depending on their age distributions, vaccination coverage, cumulative incidence, and adoption of risk mitigating behaviors. Yet, existing predictive models of COVID-19 hospitalizations are almost exclusively focused on national- and state-level predictions. This leaves local policymakers in urgent need of tools that can provide early warnings about the possibility that COVID-19 hospitalizations may rise to levels that exceed local capacity. In this work, we develop a framework to generate simple classification rules to predict whether COVID-19 hospitalization will exceed the local hospitalization capacity within a 4- or 8-week period if no additional mitigating strategies are implemented during this time. This framework uses a simulation model of SARS-CoV-2 transmission and COVID-19 hospitalizations in the US to train classification decision trees that are robust to changes in the data-generating process and future uncertainties. These generated classification rules use real-time data related to hospital occupancy and new hospitalizations associated with COVID-19, and when available, genomic surveillance of SARS-CoV-2. We show that these classification rules present reasonable accuracy, sensitivity, and specificity (all ≥ 80%) in predicting local surges in hospitalizations under numerous simulated scenarios, which capture substantial uncertainties over the future trajectories of COVID-19. Our proposed classification rules are simple, visual, and straightforward to use in practice by local decision makers without the need to perform numerical computations.
| S-EPMC9872755 | biostudies-literature
Project description:BackgroundThe COVID-19 pandemic prompted various containment strategies, such as work-from-home policies and reduced social contact, which significantly altered people's sleep routines. While previous studies have highlighted the negative impacts of these restrictions on sleep, they often lack a comprehensive perspective that considers other factors, such as seasonal variations and physical activity (PA), which can also influence sleep.ObjectiveThis study aims to longitudinally examine the detailed changes in sleep patterns among working adults during the COVID-19 pandemic using a combination of repeated questionnaires and high-resolution passive measurements from wearable sensors. We investigate the association between sleep and 5 sets of variables: (1) demographics; (2) sleep-related habits; (3) PA behaviors; and external factors, including (4) pandemic-specific constraints and (5) seasonal variations during the study period.MethodsWe recruited working adults in Finland for a 1-year study (June 2021-June 2022) conducted during the late stage of the COVID-19 pandemic. We collected multisensor data from fitness trackers worn by participants, as well as work and sleep-related measures through monthly questionnaires. Additionally, we used the Stringency Index for Finland at various points in time to estimate the degree of pandemic-related lockdown restrictions during the study period. We applied linear mixed models to examine changes in sleep patterns during this late stage of the pandemic and their association with the 5 sets of variables.ResultsThe sleep patterns of 27,350 nights from 112 working adults were analyzed. Stricter pandemic measures were associated with an increase in total sleep time (TST) (β=.003, 95% CI 0.001-0.005; P<.001) and a delay in midsleep (MS) (β=.02, 95% CI 0.02-0.03; P<.001). Individuals who tend to snooze exhibited greater variability in both TST (β=.15, 95% CI 0.05-0.27; P=.006) and MS (β=.17, 95% CI 0.03-0.31; P=.01). Occupational differences in sleep pattern were observed, with service staff experiencing longer TST (β=.37, 95% CI 0.14-0.61; P=.004) and lower variability in TST (β=-.15, 95% CI -0.27 to -0.05; P<.001). Engaging in PA later in the day was associated with longer TST (β=.03, 95% CI 0.02-0.04; P<.001) and less variability in TST (β=-.01, 95% CI -0.02 to 0.00; P=.02). Higher intradaily variability in rest activity rhythm was associated with shorter TST (β=-.26, 95% CI -0.29 to -0.23; P<.001), earlier MS (β=-.29, 95% CI -0.33 to -0.26; P<.001), and reduced variability in TST (β=-.16, 95% CI -0.23 to -0.09; P<.001).ConclusionsOur study provided a comprehensive view of the factors affecting sleep patterns during the late stage of the pandemic. As we navigate the future of work after the pandemic, understanding how work arrangements, lifestyle choices, and sleep quality interact will be crucial for optimizing well-being and performance in the workforce.
| S-EPMC11408889 | biostudies-literature
Project description:Objective: Biomechanical Machine Learning (ML) models, particularly deep-learning models, demonstrate the best performance when trained using extensive datasets. However, biomechanical data are frequently limited due to diverse challenges. Effective methods for augmenting data in developing ML models, specifically in the human posture domain, are scarce. Therefore, this study explored the feasibility of leveraging generative artificial intelligence (AI) to produce realistic synthetic posture data by utilizing three-dimensional posture data. Methods: Data were collected from 338 subjects through surface topography. A Variational Autoencoder (VAE) architecture was employed to generate and evaluate synthetic posture data, examining its distinguishability from real data by domain experts, ML classifiers, and Statistical Parametric Mapping (SPM). The benefits of incorporating augmented posture data into the learning process were exemplified by a deep autoencoder (AE) for automated feature representation. Results: Our findings highlight the challenge of differentiating synthetic data from real data for both experts and ML classifiers, underscoring the quality of synthetic data. This observation was also confirmed by SPM. By integrating synthetic data into AE training, the reconstruction error can be reduced compared to using only real data samples. Moreover, this study demonstrates the potential for reduced latent dimensions, while maintaining a reconstruction accuracy comparable to AEs trained exclusively on real data samples. Conclusion: This study emphasizes the prospects of harnessing generative AI to enhance ML tasks in the biomechanics domain.
| S-EPMC10899878 | biostudies-literature