University of Houston

Presently, the zebrafish has become a viable model for various research including vertebrate development, gene expression analysis, human diseases modelling, drug screening and toxicology analysis. Compared to other models, the zebrafish embryo is an attractive model due its fast development, large transparent embryos that develop outside the mother, and the availability of a large selection of mutant strains. An important factor facilitating the application of zebrafish in research is high throughput screening. However, the lack of tools for automated analysis of complex images is a huge bottleneck in utilizing the zebrafish to its full potential. Zebrafish intersegmental vessel (ISV) growth is widely used for drug discovery, and to study toxins impact on growth. Most of the current research based on ISV observe the presence or absence of ISV or perturbation of ISV morphology but do not measure growth dynamics. Moreover, these analysis are done manually hence it is tedious and expensive. All of these factors facilitate the need for an automated image processing methods to quantitatively analyse the image dataset.

In this work, we have focused on developing an image processing algorithm to automatically segment and quantify ISVs of zebrafish embryos that have been treated by various toxins. The challenge in this type of application includes aligning embryos of different orientations and automatically extracting ISVs. The image processing pipeline consists of Segmentation, Region Detection, ISV Extraction, ISV refinement and quantification of ISVs. We tested the algorithm using images of zebrafish embryos obtained from screening compounds that may act as an ISV disruptor. The efficiency of segmentation approach is demonstrated by our experiments of the entire zebrafish vasculature recorded from the fluorescence confocal microscope. The experiments also demonstrate that automated segmentation of ISV is comparable to that of manual segmentation.