Grado Department of Industrial and Systems Engineering

Permanent URI for this community

https://hdl.handle.net/10919/24217

Browse

Browsing Grado Department of Industrial and Systems Engineering by Author "Abedi, Vida"

Now showing 1 - 3 of 3
  • Thumbnail Image
    High-resolution computational modeling of immune responses in the gut
    Verma, Meghna; Bassaganya-Riera, Josep; Leber, Andrew; Tubau-Juni, Nuria; Hoops, Stefan; Abedi, Vida; Chen, Xi; Hontecillas, Raquel (Oxford University Press, 2019-06-01)
    Background: Helicobacter pylori causes gastric cancer in 1-2% of cases but is also beneficial for protection against allergies and gastroesophageal diseases. An estimated 85% of H. pylori-colonized individuals experience no detrimental effects. To study the mechanisms promoting host tolerance to the bacterium in the gastrointestinal mucosa and systemic regulatory effects, we investigated the dynamics of immunoregulatory mechanisms triggered by H. pylori using a high-performance computing-driven ENteric Immunity SImulator multiscale model. Immune responses were simulated by integrating an agent-based model, ordinary, and partial differential equations. Results: The outputs were analyzed using 2 sequential stages: The first used a partial rank correlation coefficient regression-based and the second a metamodel-based global sensitivity analysis. The influential parameters screened from the first stage were selected to be varied for the second stage. The outputs from both stages were combined as a training dataset to build a spatiotemporal metamodel. The Sobol indices measured time-varying impact of input parameters during initiation, peak, and chronic phases of infection. The study identified epithelial cell proliferation and epithelial cell death as key parameters that control infection outcomes. In silico validation showed that colonization with H. pylori decreased with a decrease in epithelial cell proliferation, which was linked to regulatory macrophages and tolerogenic dendritic cells. Conclusions: The hybrid model of H. pylori infection identified epithelial cell proliferation as a key factor for successful colonization of the gastric niche and highlighted the role of tolerogenic dendritic cells and regulatory macrophages in modulating the host responses and shaping infection outcomes.
  • Thumbnail Image
    Machine Learning-Enabled 30-Day Readmission Model for Stroke Patients
    Darabi, Negar; Hosseinichimeh, Niyousha; Noto, Anthony; Zand, Ramin; Abedi, Vida (Frontiers, 2021-03-31)
    Background and Purpose: Hospital readmissions impose a substantial burden on the healthcare system. Reducing readmissions after stroke could lead to improved quality of care especially since stroke is associated with a high rate of readmission. The goal of this study is to enhance our understanding of the predictors of 30-day readmission after ischemic stroke and develop models to identify high-risk individuals for targeted interventions. Methods: We used patient-level data from electronic health records (EHR), five machine learning algorithms (random forest, gradient boosting machine, extreme gradient boosting–XGBoost, support vector machine, and logistic regression-LR), data-driven feature selection strategy, and adaptive sampling to develop 15 models of 30-day readmission after ischemic stroke. We further identified important clinical variables. Results: We included 3,184 patients with ischemic stroke (mean age: 71 ± 13.90 years, men: 51.06%). Among the 61 clinical variables included in the model, the National Institutes of Health Stroke Scale score above 24, insert indwelling urinary catheter, hypercoagulable state, and percutaneous gastrostomy had the highest importance score. The Model’s AUC (area under the curve) for predicting 30-day readmission was 0.74 (95%CI: 0.64–0.78) with PPV of 0.43 when the XGBoost algorithm was used with ROSE-sampling. The balance between specificity and sensitivity improved through the sampling strategy. The best sensitivity was achieved with LR when optimized with feature selection and ROSE-sampling (AUC: 0.64, sensitivity: 0.53, specificity: 0.69). Conclusions: Machine learning-based models can be designed to predict 30-day readmission after stroke using structured data from EHR. Among the algorithms analyzed, XGBoost with ROSE-sampling had the best performance in terms of AUC while LR with ROSE-sampling and feature selection had the best sensitivity. Clinical variables highly associated with 30-day readmission could be targeted for personalized interventions. Depending on healthcare systems’ resources and criteria, models with optimized performance metrics can be implemented to improve outcomes.
  • Thumbnail Image
    Multi-Resolution Sensitivity Analysis of Model of Immune Response to Helicobacter pylori Infection via Spatio-Temporal Metamodeling
    Chen, Xi; Wang, Wenjing; Xie, Guangrui; Hontecillas, Raquel; Verma, Meghna; Leber, Andrew; Bassaganya-Riera, Josep; Abedi, Vida (Frontiers, 2019-02-05)
    Computational immunology studies the interactions between the components of the immune system that includes the interplay between regulatory and inflammatory elements. It provides a solid framework that aids the conversion of pre-clinical and clinical data into mathematical equations to enable modeling and in silico experimentation. The modeling-driven insights shed lights on some of the most pressing immunological questions and aid the design of fruitful validation experiments. A typical system of equations, mapping the interaction among various immunological entities and a pathogen, consists of a high-dimensional input parameter space that could drive the stochastic system outputs in unpredictable directions. In this paper, we perform spatio-temporal metamodel-based sensitivity analysis of immune response to Helicobacter pylori infection using the computational model developed by the ENteric Immune SImulator (ENISI). We propose a two-stage metamodel-based procedure to obtain the estimates of the Sobol’ total and first-order indices for each input parameter, for quantifying their time-varying impacts on each output of interest. In particular, we fully reuse and exploit information from an existing simulated dataset, develop a novel sampling design for constructing the two-stage metamodels, and perform metamodel-based sensitivity analysis. The proposed procedure is scalable, easily interpretable, and adaptable to any multi-input multi-output complex systems of equations with a high-dimensional input parameter space.