Mining of Cell Assay Images Using Active Semi-Supervised Clustering
Nicolas Cebron, Michael R. Berthold · KOPS (University of Konstanz) · 2005
Classifying large datasets without any a-priori information poses a problem especially in the field of bioinformatics. In this work, we explore the problem of classifying hundreds of thousands of cell assay images obtained by a highthroughput screening camera. The goal is to label a few selected examples by hand and to automatically label the rest of the images afterwards. We deal with three major requirements: first, the model should be easy to understand, second it should offer the possibility to be adjusted by a domain expert, and third the interaction with the user should be kept to a minimum. We propose a new active clustering scheme, based on an initial Fuzzy c-means clustering and Learning Vector Quantization. This scheme can initially cluster large datasets unsupervised and then allows for adjustment of the classification by the user. Furthermore, we introduce a framework for the classification of cell assay images based on this technique. Early experiments show promising results. 1.