abstract |
In this project with Altana Pharma AG, we are dealing with a large amount of image data which is obtained by a fluorescence microscope camera. Normally, an expert classifies some of the cells in the images and generates rules for classifying the others. The amount of data is so large that it would take a long time to classify all cells by hand. We applied a new approach for automatic cell classification which uses the fuzzy c-means algorithm to find the underlying structure of the data. Afterwards, the expert evaluates and possibly adjusts these clusters. The aim of the project is to reduce the amount of training data and time significantly while keeping the misclassification rate at a low level.
|