Biological imaging has recently received a strong impulse by the development of fluorescent probes and new high-resolution microscopes. It is presently having a profound impact on the way research is being conducted in the life sciences. Biologists can now visualize subcellular components and processes in vivo, both structurally and functionally. Observations can be made in two or three dimensions, at different wavelengths (spectroscopy), possibly with time-lapse imaging to investigate cellular dynamics. The observation of many biological processes relies on the ability to identify and locate specific proteins within their cellular environment. High content screening is the combination of modern cell biology, with all its molecular tools, with automated high resolution microscopy and robotic handling. Image analysis is then used to measure changes in properties of the cells caused by external treatment such as chemical inhibitors or RNA interference. Therefore the methods for automatically determine and measure the number and shape of cells within a given multichannel image constitute a basic step in much of the applications of high throughput screening. Here the problem of a fully automatic segmentation of 3D high throughput images is approached. The adopted method is an extension to the three dimensional domain of a recently proposed energy based method. This is a two step process, first there is the detection of nuclei and then an energy based surface evolution approach tries to separate the surfaces of the various cells in the image. The adopted energy model is an adaptation of the repulsion competition approach, already proposed in a two-dimensional framework by Yan, Zhou and Shah Here we use a different nucleus identification method suited for three dimensional images and an efficient extension of the level set surface evolution algorithm. Experimental results on samples of the HeLa cell image database are reported.