Bayesian methods have a big potential for immediate application in areas outside science. There is a long-standing and quite unique tradition in the SNN group to build such application together with her spin-off companies Smart Research and Promedas. Here are a few examples:

Oil exploration

For Shell, we built a petrophysical expert system. It estimates the type of soil and the probability that it contains oil, gas or other valuable minerals, based on drilling measurements. The system is based on a Bayesian network where the probability computation is done using a Monte Carlo sampling method. See Smart Research for further details and other products.

Victim identification

For the Netherlands Forensic Institute, we are building a victim identification system by matching of their DNA profiles against the Pedigrees of Relatives from Missing Person's DNA profiles in large databases, using a Bayesian network. See Bonaparte for further details.

Promedas

We have built the world largest and most up-to-date medical expert system for diagnostic advice in internal medicine. The system is being commercialized by Promedas bv. The system is operational since end 2008 at the Utrecht academic hospital. See Promedas for further details.

Wine and food

We have built a system that selects the most appropriate wines to combine with your food Wine wine wine.

Neural networks for Parkinson′s Disease

Levodopa induced dyskinesia (LID)is a common problem for patients with Parkinson's disease (PD) and is usually related to chronic Levodopa therapy and progression of PD. An automatic and quantitative method which could be used in daily life to assess LID is not available yet and would be important to reduce LID. Accelerometers are used to measure movements of limb segments of PD patients with various degrees of severity of LID. Neural networks used parameters obtained from these accelerometer signals to classify LID corresponding to a rating system used by physicians. The advantage of neural networks is that they can be trained to distinguish LID from voluntary movements and to assess the severity of LID, even when no explicit rules are available for proper classification, as long as data are available to train these neural networks. In this project we have demonstrated the use of neural networks to detect and classify LID in daily life.

Neural Networks for the Paper Industry

Papermaking is a difficult and only partially understood process. Traditionally, the production process is controlled and optimized by human experts that have gathered insights and rules of thumb through years of experience. The increase in the number of sensors and the amount of process data stored facilitates a more quantitative analysis. The goal is to obtain insight in the production process and use this insight to help the machine operator improve the production.

In this project, we have developed methods to find new relations between quality parameters and machine settings. Narrow margins for paper qualities result in a considerable amount of out-of-specification production. Quality parameters like curl and internal bond can only be measured off-line. Hence feedback for the operators comes only once every hour or less.Often, the precise causes of quality-related problems are not known. We have developed data mining and data visualization approaches can help the technologists to find these causes and to improve and adapt the process accordingly.