Health data processing
We develop methods and computer programs to support dermatology specialists, family doctors and patients in diagnostics of dermatological diseases. We use advanced content-based image retrieval, machine learning and artificial intelligence techniques and apply them on a large database of 100000 images at the dermatology department of UMCG (University Medical Centre Groningen). When a new case is presented by a user of the system, certain characteristics of this case are automatically extracted and compared to the characteristics of the cases stored in the database. The cases that are most similar to the new case are extracted. Since their diagnoses are known, this information is used to set a diagnosis for the new case.
We developed a computer program for the evaluation of the risk of melanoma. It was used in the open door Day of the Skin in 2013 by the Dutch Association of Dermatologists in 2013.
Neuroimaging techniques such as EEG, fMRI en DTI can be used to study the structure and function of the human brain. One of the important research questions is how the activity of various brain regions is organized in brain networks. In this project we develop analysis and visualization methods to investigate the complexity of brain networks and the influence of cognitive ageing on these networks.
The GLIMPS project concerns the creation of a database of FDG-PET scans which reflect the glucose consumption of the brain in patients with neuro-degenerative diseases. The goal is to identify distinctive structural and functional brain patterns and derived quantities like network patterns of brain activity, which display statistically significant differences in healthy subjects and patients with certain types of neuro-degenerative disease. The database will be used for clinical practice as well as for research purposes. Pattern classification and machine learning methods will be developed to associate brain patterns to various types and stages of neuro-degenerative disease. To acquire a sufficient number of scans the database will be organized on a national scale.
The project is a collaboration of the following institutes at the University Medical Center Groningen/University of Groningen:
a) Department of Neurology, UMCG: prof. K.L. Leenders, neurologist (PI), drs. L. Heinen (coordinator)
b) Department of Nuclear Medicine (UMCG): scan quality control
c) Neuroimaging Center (UMCG/RUG): prof. J.B.T.M. Roerdink, dr. R. Renken (MRI acquisition, model development)
d) Johann Bernoulli Institute for Mathematics and Computer Science (RUG): prof. J.B.T.M. Roerdink (pattern recognition, visualization)
d) TARGET group (RUG): databasing, web application development.
The pilot phase of the project has formally started in January 2012.
|Laatst gewijzigd:||01 april 2015 14:03|