DeepFRAP: Fast fluorescence recovery after photobleaching data analysis using deep neural networks

Dataset and code used in V Wåhlstrand-Skärström, et al, "DeepFRAP: Fast fluorescence recovery after photobleaching data analysis using deep neural networks", published in Journal of Microscopy. In this work, we develop a new approach for FRAP analysis based on deep neural networks. From a numerical FRAP model developed in previous work, we generate a very large set of realistic, simulated recovery curve data. The data is used for training deep neural network regression models for prediction of e.g. the diffusion coefficient. We compare the performance of the neural network estimation framework to conventional least squares estimation on simulated and 
experimental data. Herein, the simulated FRAP data used for the training, validation, and test data sets, the experimental data, and the Matlab and Python/Tensorflow code are supplied.