An Automated Assay for the Evaluation of Mortality in Fish Embryo

• Main Authors: Puybareau, Élodie, Léonard, Marc, Talbot, Hugues
• Other Authors: Laboratoire d'Informatique Gaspard-Monge (LIGM), Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout (BEZOUT), Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), L’Oréal Recherche & Innovation, L'OREAL
• Format: Conference Object
• Language: English
• Published: HAL CCSD 2015
• Subjects: Toxicology; medaka; image stabilisation; change detection; connected filtering; [INFO]Computer Science [cs]; Iceland
• Online Access: https://hal.science/hal-01154543; https://hal.science/hal-01154543/document; https://hal.science/hal-01154543/file/paper40.pdf; https://doi.org/10.1007/978-3-319-18720-4_10

International audience Waterways are often first and severely affected by pollution. In this context, fish embryos – which constitute a good model for sensitivity to chemicals – are widely used in environmental toxicology studies. Such studies are devoted to the analysis of a wide spectrum of physiological parameters, for instance mortality ratio. In this article, we develop an assay to determine the mortality rate of Medaka embryo. Based on video sequences, our purpose is to obtain reliable, repeatable results in a fully automated fashion. To reach that challenging goal, we develop an efficient morphological pipeline that analyses image sequences in a multiscale paradigm, from the global scene to the embryo, and then to its heart, finally analysing its putative motion, characterized by intensity variations. Our pipeline, based on robust morphological operators, has a low computational cost, and was experimentally assessed on a dataset consisting of 660 images, providing a success ratio higher than 99%.