Applications

Predictive Models in Biomedical Engineering and Epidemiology

SLL researchers have developed predictive models for heart volumes which are used to perform virtual heart transplants, spatial statistical models for the growth of thrombus in abdominal aortic aneurysms, nonparametric functional classifiers for disease indentification and spatio-temporal models for epidemics.  

Spatio-temporal models in Geo-Sciences

SLL members carry out innovative research on spatio-temporal models with spatially varying coefficients, by combining Moran’s eigenvector filtering with fast algorithms for dimension reduction and penalized estimators. Such models have been applied in the analysis of remotely sensed data and the analysis of outputs from regional climate models. 

Vehicular Traffic Forecasting and Incident Detection

The SLL team develops real-time systems for incident detection and short-term forecasting of traffic variables (e.g. speeds, volumes and occupancies) in urban networks and freeways. For that purpose, alternative modeling approaches, including linear and nonlinear, parametric and nonparametric time series models and quantile regressions, are synthesized. 

Space-time Econometrics

SLL researchers developed space-time models which aim to explain the dynamics of regional productivity in the EU and the dynamics of adoption of innovations in the agricultural sector in the US. Currently, our group develops spatial and spatiotemporal models for the adoption of agricultural innovation  in the US. 

Statistical Analysis of Transportation Emissions

SLL analysts conduct cutting edge research on modeling high-frequency emissions rates from real-world experiments, emphasizing on pollutants which are hard to measure, such as particle numbers (PN). The undertaken analyses combine parametric nonlinear time series models with robust estimators.  

Parametric Sensitivity Analysis in Biochemical Reaction Networks

SLL members have developed parametric sensitivity analysis tools as an approach to mathematically and computationally understand and evaluate the behavior of complex phenomena in biochemical reaction networks.

Deep Learning in Speech Processing

Deep Neural Networks have taken the engineering community by storm. Data-rich areas such as image processing and speech processing have been transformed during the last years. SLL team combines its expertise on speech processing and applies deep learning techniques to applications such as voice conversion, speech synthesis and speech enhancement.

Various Predictive Models

SLL members participate in several peripheral projects that require the design, implementation, validation and deployment of a predictive model. Applications are diverse including energy production forecasting for wind farms, outbreak forecasting, Raman spectroscopy, materials' properties prediction and more.