Data Descriptor: Automatized image processing of bovine blastocysts produced in vitro for quantitative variable determination

Abstract

There is currently no objective, real-time and non-invasive method for evaluating the quality of mammalian embryos. In this study, we processed images of in vitro produced bovine blastocysts to obtain a deeper comprehension of the embryonic morphological aspects that are related to the standard evaluation of blastocysts. Information was extracted from 482 digital images of blastocysts. The resulting imaging data were individually evaluated by three experienced embryologists who graded their quality. To avoid evaluation bias, each image was related to the modal value of the evaluations. Automated image processing produced 36 quantitative variables for each image. The images, the modal and individual quality grades, and the variables extracted could potentially be used in the development of artificial intelligence techniques (e.g., evolutionary algorithms and artificial neural networks), multivariate modelling and the study of defined structures of the whole blastocyst.