T.W. Nattkemper, W. Schubert, T. Hermann, and H. Ritter (Germany)
Molecular and Cellular Engineering, Optical Imaging, Fluorescence Microscopy, Medical Image Processing,Artificial Neural Networks
To analyze large sets of digital micrographs from high-throughput screening studies with constant accuracy, advanced image processing algorithms are necessary. In the literature, systems have been proposed applying model based fitting algorithms, morphological operators and arti ficial neural networks (ANN). Because single approaches show limited performance, we propose a hybrid system that combines the Hough transform with a multi-layer percep tron (MLP) network. Our results show, that the combina tion of both approaches improves the performance and the positions of cell bodies are obtained with increased sensi tivity and positive predictive value.
Important Links:
Go Back
