Search | Research Portal

1 - 10 of 1833 results

BOF Beleidsreserve: mandate of Thijs Becker: Explainable machine learning, predictive modelling, and causal inference, applied to medical data Hasselt University

This research consists of applying machine learning to medical data, with a focus on trustworthy and explainable machine learning methods. We are investigating how to improve longitudinal methods for optimizing the care of people with multiple sclerosis. We investigate predictive models, which can tell the patient whether he or she will progress in disability after a certain time period. This helps the patients plan their life. We focus on how ...

Predictive mOdelling Tools to evaluate the Effects of Climate change on food safeTy and spoilage KU Leuven

Jan Van Impe

Chemical & Biochemical Process Technology & Control (BioTec+)

Climate change and food safety have become interdependent worldwide research priorities. In order to meet the EU challenge of doubling food production by 2050 (to meet population demands) while dealing with the impact of climate change on food safety, investment in research to address this issue is required. The overarching aim of this Innovative Training Network (ITN) is to provide high-level training in Predictive mOdelling Tools to ...

A rigorous modelling cycle for reliable predictive microbiology: Application to the specific microbial growth rate KU Leuven

Jan Van Impe

Sustainable Chemical Process Technology, Ghent and Aalst Technology Campuses

Predictive microbiology is the science of developing mathematical models that describe the behaviour of microorganisms in food products. This behaviour is described as a function of the food properties. However, there are several difficulties in describing the lag phase of microbial growth, i.e., the time it takes until microorganisms can start growing when placed in a new environment. Therefore, the current research will work on developing ...

Dynamic causal modelling of preparatory cortical activity and its relation to the theory of predictive coding. KU Leuven

Johan Wagemans

Brain and Cognition, Laboratory for Experimental Psychology

A general theory of cortical interactions, known as predictive coding, has recently gained a lot of interest among neuroscientists. The theory claims that the brain actively anticipates upon the upcoming sensory input by generating predictions against which the input in sensory cortex is evaluated. From this, one would expect modifications of neural activity in sensory cortex before the actual appearance of a stimulus. In the current proposal we ...

Validating predictive models of radiotherapy toxicity to improve quality-of-life and reduce side-effects in cancer survivors. KU Leuven

Dirk De Ruysscher

Laboratory of Experimental Radiotherapy

Long-term side-effects of radiotherapy impact on the quality-of-life (QoL) of cancer survivors. These side-effects could be reduced if predicted in advance. Previous work identified clinical and biological predictors but a major, coordinated approach is needed to validate them so they can be used clinically. The EU has ~17.8 million people living with a prior diagnosis of cancer of whom ~7 million received radiotherapy. In the long-term, ...

Sparse predictive modeling with applications in insurance pricing and mortality forecasting KU Leuven

Katrien Antonio

Research Centre Insurance, Leuven

The insurance sector relies heavily on data for a variety of their operational processes such as product pricing, marketing and estimating future expenses. As today's society generates data more rapidly than ever before, the demand for new algorithms, able to infer meaningful information from this data, is rising. A modern issue in insurance is that data sets not only contain a lot of observations, but also many variables of different types. ...

M3Strength: Efficient predictive modeling for composites strength Ghent University

Joris Degrieck

Department of Materials Science and Engineering

SIM SBO M3 Strength Efficient predictive modeling for the strength of composite materials through the first SBO Project, "M3Strength", the "M3" program focus in a first step on the strength of composite materials; The strength is in fact the main driving factor in the development of structural applications in automotive, aerospace and wind energy. The strength will be studied for all relevant stins regimes: (quasi-) static, dynamic (fatigue) ...

Development of predictive models for the growth of L. monocytogenes and for the compliance with L. monocytogenes food safety criteria in foods covered by the baseline survey prevalence of L. monocytogenes contaminated RTE foods. Hasselt University

Marc AERTS

Centre for Statistics

Predictive models for microbial growth of L. monocytogenes will be developed under the various storage conditions. These models will use enumeration data from the surveys in packaged fish at two different time points (at sampling and at end of shelf-life). The effect of temperature, pH and water activity will be addressed. The development of predictive models for the compliance with the L. monocytogenes criteria is laid down in Regulation ...

Development of predictive models for critically ill patients with acute kidney injury KU Leuven

Celine Vens

Public Health and Primary Care, Kulak Kortrijk Campus

Predictive models are widely used in intensive care units (ICU), due to the widespread implementation of electronic systems to collect patient data: demographic information, continuous monitoring of vital functions, information on medication administered, results from labatory analyses, etc. In this project, we will develop predictive modelling techniques to tackle a number of machine learning challenges related to data analysis within an ICU ...

Predictive modeling of spatiotemporal phenomena in Geographic Information Systems using Machine Learning Ghent University

Nico Van de Weghe

Department of Geography

The growing wealth of data on spatiotemporal phenomena allows for new modeling approaches in Geographical
Information Science. In this context, this doctoral mandate will examine the usefulness of machine learning techniques --U+2010 a data--U+2010driven approach for predictive modeling. The usefull techniques for this challenge will be identified, ported to the GIS--U+2010framework and tested using data from a mass--U+2010event. ...

Projects