DeepFinder is an original deep learning approach to localize macromolecules in cryo electron tomography images. The method is based on image segmentation using a 3D convolutional neural network.
This method has been developed in the frame of Emmanuel Moebel’s PhD thesis at Inria Rennes, under Charles Kervrann’s supervision (team Serpico).
If you are curious, you can access his PhD manuscript here , entitled “New strategies for the identification and enumeration of macromolecules in 3D images of cryo electron tomography”.
What DeepFinder is¶
DeepFinder is a python3 package that allows analysing 3D images with a deep learning method. It possesses a graphical user interface to allow non-computer scientists to work with this tool. It is based on the Keras package.