High throughput automated detection of axial malformations in Medaka embryo. (February 2019)

• Record Type: Journal Article
• Authors: Genest, Diane; Puybareau, Elodie; Léonard, Marc; Cousty, Jean; De Crozé, Noémie; Talbot, Hugues
• Is Part Of: Computers in biology and medicine. Volume 105(2019)
• Page Start: 157
• Page End: 168
• Publication Date: 2019-02
• Subjects: Toxicology screening -- Discrete mathematical morphology -- Segmentation -- Features screening -- Machine learning -- Random forest; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2018.12.016
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Fish embryo models are widely used as screening tools to assess the efficacy and/or toxicity of chemicals. This assessment involves the analysis of embryo morphological abnormalities. In this article, we propose a multi-scale pipeline to allow automated classification of fish embryos (Medaka: Oryzias latipes ) based on the presence or absence of spine malformations. The proposed pipeline relies on the acquisition of fish embryo 2D images, on feature extraction based on mathematical morphology operators and on machine learning classification. After image acquisition, segmentation tools are used to detect the embryo before analysing several morphological features. An approach based on machine learning is then applied to these features to automatically classify embryos according to the presence of axial malformations. We built and validated our learning model on 1459 images with a 10-fold cross-validation by comparison with the gold standard of 3D observations performed under a microscope by a trained operator. Our pipeline results in correct classification in 85% of the cases included in the database. This percentage is similar to the percentage of success of a trained human operator working on 2D images. The key benefit of our approach is the low computational cost of our image analysis pipeline, which guarantees optimal throughput analysis. Graphical abstract: Image 1 Highlights: An automated method to detect the presence or absence of spine malformations is fish embryos 2D images is proposed. The method is based on features extraction due to mathematical morphology and on machine learning classification. We assess the inter-expert subjectivity and the rate of information loss between microscope 3D observations and 2D images. We achieve a success rate of 85% on 1459 images, which is similar to the one of a human operator working on 2D images.