Extracting information about the structure of biological tissue from static image data is a complex task which requires a series of computationally intensive operations. Here we present how the power of multi-core CPUs and massively parallel GPUs have been utilised to extract information about the shape, size and path followed by the mammalian oviduct, called the fallopian tube in humans, from histology images, to create a realistic 3D virtual organ for use in predictive computational models. Histology images from a mouse oviduct were processed, using a combination of GPU and multi-core CPU techniques, to identify the individual cross-sections and determine the 3D path that the tube follows through the tissue. This information was then related back to the histology images, linking the 2D cross-sections with their corresponding 3D position along the oviduct. Measurements were then taken from the images and used to computationally generate a series of linear 2D spline cross-sections for the length of the oviduct, which were bound to the 3D path of the tube using a novel particle system based technique that provides smooth resolution of self intersections and crossovers from adjacent sections. This results in a unique 3D model of the oviduct, which is based on measurements of histology slides and therefore grounded in reality. The GPU is used for the processor intensive operations of image processing and particle physics based simulations, significantly reducing the time required to generate a complete model. A set of models created using this technique is being used to investigate the influence that the 3D structure of the oviductal environment has on sperm transport and navigation.
Paper available at ACM.