Projects
Predictive clustering for structured biomedical data. KU Leuven
In many areas of science, data is generated at an increasing pace, calling for adequate data mining tools to extract useful knowledge. Challenges encountered include the scale and the complexity (structure) of the data. An increasing number of new applications, including the analysis of complex biomedical data, motivate the adaptation of existing, and the design of new methodologies. In this project, we will develop new data mining methods to ...
Automated image feature extraction and computer vision for quantitative brain tumour diagnosis Ghent University
Yearly, about 35.000 brain tumours are diagnosed. While for patient stratification, a correct diagnosis is of utmost importance, there is growing evidence that quantitative diagnosis is more accurate. This project will implement fully automatic image feature extraction for an improved diagnosis. This will eventually result in better personalized treatment and patient stratification, thereby improving patient outcome and decreasing healthcare ...
The DUCK project: Distraction from learning by Unrelated auditory events assessed by Computational modeling and Knowledge extraction from single-trial electroencephalography (EEG) Ghent University
The human brain is remarkable capable to suppress, focus on, and switch between auditory streams, but the underlying interplay of neurophysiology and acoustics remains unclear. The DUCK project aims to develop a human-mimicking computational approach applicable in real-world learning conditions. To assess and validate the included stages of auditory processing, single-trial electroencephalography (EEG) will be used in ecological valid ...
Nieuwe kernmerkextractie methoden voor de verbeterde analyse van magnetische resonantie spectroscopie en spectroscopische beeldvorming. KU Leuven
Understanding High-Dimensional Time-Series: Topological and Visual Analytics for Characterizing Sleep Apnea. Hasselt University
Geometrical data mining with applications in bioinformatics. KU Leuven
Image quality improvement of emission tomography Ghent University
The focus of this project will be to further enhance the image quality of emission tomography. To achieve this we will use (and evantually modify) prototype or recent clinical systems to maximize the sensitivity. On the processing side we will try to minimize the image quality loss in the reconstruction process by accurate modeling of the imaging process. In most ET imaging situations there is a high fraction of contamination in the measured ...
Detection and identification of nonlinear properties in the respiratory impedance Ghent University
This research proposal aims to correlate the nonlinear contributions present in systems, with the class of models known as generalized-order models (fractional-order models). The proposal enables the possibilities to extract information from nonlinear systems by means of detection and quantification of the nonlinear contributions in a generalized context. The envisaged methods are based on specific signal design and specific signal processing ...