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.

Distinct Sources of a Bovine Blastocyst Digital Image do not Produce the Same Classification by a Previously Trained Software using Artificial Neural Network

“…To analyze the acquired images, was used the desktop version of Blasto3Q software. First, the image was uploaded on the program, so the image was processed and evaluated by the already trained algorithm Blasto3Q, which is detailed described in [42]. In the software, the uploaded images underwent an automatized segmentation process, whereby the images were properly isolated from the background.…”

“…Supported by the GA, three ANNs (with the best accuracy) were developed after to have been trained to classify bovine blastocyst images based on the IETS standards [41,42]. In that study, 482…”

“…The work mentioned above, resulted in two different interfaces, one using the MATLAB ® platform and an interface using multiplatform approach for online purpose. Both of them, we called "Blasto3Q" program (3Q is refer to the three qualities of blastocysts evaluation) [41,42]. In the first case, the interface created (in accordance with the recommendations for the commercial use of the program) allows the user interact with the program in a fast and intuitive way.…”

“…To the image segmentation they were properly isolated from the background and were then subjected to the information extraction techniques. The segmentation algorithm is composed for 4 steps describe below (more details in our previous works) [41,42]. The first step was calculating the magnitude of the image gradient and the edges were highlighted in all directions, it is important to next steps and essential to characterize the circular shape of the embryo.…”

“…After image standardization and segmentation, 36 variables were extracted, sufficient to extract all the relevant information for the representation of the analyzed bovine blastocyst image. For the complete description of variables, see previous work [42]. The collinearity analysis was performed to eliminate those variables that were correlated with one another (i.e., redundant).…”