Best performance award for new machine learning technique

The Team Admire-LVQ headed by Prof. Michael Biehl (Computer Science) has achieved best performance in the DREAM6/FlowCAP challenge 2011.

 
The goal in the Molecular Classification of Acute Myeloid Leukaemia (AML) Challenge was to identify cases of AML based on flow cytometry data. Example data of diagnosed patients were provided; the teams had to hand in predictions with respect to a set of 180 patients whose diagnosis was unknown to the participants.

The University of Groningen team applied a machine learning technique which has been developed within the Intelligent Systems group as part of an NWO supported research project: Adaptive Distance Measures in Relevance Learning Vector Quantization (Admire-LVQ). The team achieved the best possible performance with 100% correct prediction of all AML cases in the test set.

The challenge was organized jointly by the DREAM project (Dialogue for Reverse Engineering Assessments and Methods) and the FlowCAP project (Flow Cytometry: Critical Assessment of Population Identification).

The team was composed of Dr. Petra Schneider, Kerstin Bunte and Prof. Michael Biehl.

More information: Prof. Michael Biehl