Wish-IT ®, Wish Innovation Technologies ®
In this project we are developing a new business framework for
customer driven innovation. The technology behind this
is patent applied and based upon advanced data mining,
machine learning and pattern recognition techniques.
Basically it is a framework to
"provide the customer with what the consumer wish ©".
The project and the services will later be presented at
The Wish-IT ® portal.
|
Knowledge Discovery in a Drug Safety Database
Within this joint project together with the WHO Collaborating Centre for International Drug Monitoring in Uppsala (WHO-UMC), Sweden, NeuroLogic is responsible for the development of methods for automated discovery of dependencies and complex patterns in large databases. The methods developed in this on-going project have been in routine use, since 1998, to signal possible causal relationships between drug substances and adverse events, in the world's largest collection of spontaneous reports of suspected adverse drug reactions.
|
Orre et al. (2000), Bate et al. (1998)
|
|
Knowledge Discovery in a Patient Safety Database
In 2002, NeuroLogic carried out a pilot study of the application of automated knowledge discovery methods for pattern recognition to a database of adverse incidents within the British National Health Service (NHS). The sponsor of this project was the UK National Patient Safety Agency (NPSA).
|
|
Automated Risk Assessment for Cases of Organo-Phosphate Poisoning
On behalf of the International Programme on Chemical Safety (IPCS), NeuroLogic implemented a Bayesian Neural Network for predicting the outcome of organo-phosphate poisoning. Classification was based on a set of approximately 50 relevant variables including for example gender, age, heart rate and blood pressure.
|
|
Pulp Quality Modeling
In a project sponsored by the Swedish Pulp and Paper Research Institute (STFI) and the Swedish Research Council for Engineering Sciences (TFR), NeuroLogic developed a Bayesian Mixture Density Neural Network for prediction of pulp quality based on process parameters.
|
Orre and Lansner (1995)
|



|