PYSUPPOSE3D: 3D DECONVOLUTION WITH SUPERRESOLUTION IN FLUORESCENCE MICROSCOPY

Abstract

In the present work, a fluorescence microscopy system was implemented that allows the acquisition of three dimensional images for subsequent processing using the SUPPOSe 3D algorithm with the aim of obtaining a superresolution reconstruction. SUPPOSe is a superresolution deconvolution algorithm that allows reconstructing the true distorted structure in an image by incorporating a priori information. The SUPPOSe approach consists of approximating the real structure in the sample as a superposition of point sources of equal intensity, called virtual sources. This procedure simplifies the deconvolution problem and converts it into an unconstrained minimization problem, which must be solved by finding the positions of the virtual sources. These positions can take values in the whole domain and are determined by minimizing an objective function through a genetic algorithm. Throughout this work the SUPPOSe method was tested for both synthetically generated and experimentally acquired 3D images with the implemented system. Samples with known structures were used to validate the results obtained and to evaluate the performance of the algorithm as a function of the characteristics of the processed images.