Alte Version der Fakultätshomepage! Zur neuen hier klicken.
Statistik: mehr als Erbsen zählen

You are here:

areas of Research in Biostatistics


Molecular Epidemiology

Genetic variation in combination with epidemiological factors can lead to the development of complex diseases such as cancer. The SFB 475 project "Statistical Complexity Reduction in Molecular Epidemiology" is concerned with finding such risk factors, with the main focus of attention on gene-gene and gene-environment-interactions.



Gene Expression Data

Micro-arrays make the simulaneous measurement of tens of thousands of genes possible. Not just this huge number of variables combined with a much lower number of observations, but also the very noisy signals some micro-arrays deliver present large problems for the analysis of such data, which can only be negotiated through interdisciplinary collaboration. One such collaboration takes place in the SFB 475 project "Statistical Complexity Reduction in Molecular Epidemiology".




Proteins are responsible for almost every biological process in an organism and even for the development of diseases like cancer. This is why the research of the proteome and also particularly the indentifcation of proteins, whose expression differs between different groups (e.g. diseased and healthy tissue), and the location of protein-protein interaction, is of great interest. Again, the interdiciplinary collaberation is of great use, like it takes place in the Centre for Applied Proteomics.



Diagnostic Processes

Diagnostic Processes are needed for the development, optimisation and validation of test systems and analysis machines. New methods and approaches are needed in this area in oder to optimise existing processes and in order to appropriately deal with new problems. The sustained accuracy of diagnostic tests in routine diagnostics, e.g., can only be garanteed through optimal calibtraion methods. In the field of Genomics, Proteomics, Peptidomics, etc., new diagnostics are always saught after, and statistics are essential for planinng, implementation, evaluation and approval of diagnostical methods. The statistics department and the Roch Diagnostik GmbH in Penzberg, Bavaria work together to develop research in these areas.



Controlled Clinical Trials

In controlled clinical trials a treatment will be evaluated against another. The department "Statistics with Application in the Field of Engineering Sciences" is concerned with Meta-Analysis of combinations of such controlled clinical trials and the planning and evaluation of such studies with flexible or adaptable designs.



Intensive Care

In intensive care vital statistics, laboratory results, administered medication and treatments are all recorded online. The SFB 475 project "Analytical Time Series Methods for use with Online Monitoring Data in Intensive Care" is developing intelligent analytical and alarm systems based on these parameters for use in bedside observation of patients, which the SFB 475 project "Intelligent Analytical and Alarm Systems for use in Intensive Care" will lead to a new monitor prototypes complying with the requirements of clinical practise. The current approach is based particularly on robust realtime singal extraction and pattern recognition.



Toxicological Applications

Statistical methods are necessary for the planning and evaluation of toxicological studies, in order to be able to analyse the possible harmful effects of substances. The department "Mathematical Statistcs with Applications in Biometrics" is working together with the Leibniz Research Centre for Working Environment and Human Factors to analsye such studies. A big problem, which has been observed by the "Mathematical Statistics and Applications in Science" is the tansferrability of animal testing to how a substance will affect humans.



Spatial Statistics

The risk of getting a disease can also be dependent on the area in which you live. Spatial statistics and geographical information is recorded in order to discover and categorise such diseases.




Sensometrics is the application and development of statistical methods to measure sensations, e.g. the taste of food. Such methods are developed at the department of "Mathematical Statistics and Applications in Science".



Linguistic Information in Neuronal Response

In cooperation with the Fraunhofer Institute for Digital Media Technology (IDMT) in Ilmenau, Thüringen, information is being sought after in the neuronal responses of the human acoustic nerves. Based on a simulation model of the inner ear, speech is automatically recognised.