A multi-scale convolutional neural network for phenotyping high-content cellular images
Open Access
- 15 February 2017
- journal article
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 33 (13), 2010-2019
- https://doi.org/10.1093/bioinformatics/btx069
Abstract
Motivation: Identifying phenotypes based on high-content cellular images is challenging. Conventional image analysis pipelines for phenotype identification comprise multiple independent steps, with each step requiring method customization and adjustment of multiple parameters. Results: Here we present an approach based on a multi-scale convolutional neural network (M-CNN) that classifies, in a single cohesive step, cellular images into phenotypes by using directly and solely the images’ pixel intensity values. The only parameters in the approach are the weights of the neural network, which are automatically optimized based on training images. The approach requires no a priori knowledge or manual customization, and is applicable to single- or multi-channel images displaying single or multiple cells. We evaluated the classification performance of the approach on eight diverse benchmark datasets. The approach yielded overall a higher classification accuracy compared to state-ofthe- art results, including those of other deep CNN architectures. In addition to using the network to simply obtain a yes-or-no prediction for a given phenotype, we use the probability outputs calculated by the network to quantitatively describe the phenotypes. Our study shows that these probability values correlate with chemical treatment concentrations. This finding validates further our approach and enables chemical treatment potency estimation via convolutional neural networks. Availability: The network specifications and solver definitions are provided in Supplementary Software 1. Contact:william_jose.godinez_navarro@novartis.com, xian-1.zhang@novartis.com Supplementary information:Keywords
This publication has 47 references indexed in Scilit:
- BIOCAT: a pattern recognition platform for customizable biological image classification and annotationBMC Bioinformatics, 2013
- Determining the subcellular location of new proteins from microscope images using local featuresBioinformatics, 2013
- Automated Analysis and Reannotation of Subcellular Locations in Confocal Images from the Human Protein AtlasPLOS ONE, 2012
- Annotated high-throughput microscopy image sets for validationNature Methods, 2012
- An imaging assay to analyze primary neurons for cellular neurotoxicityJournal of Neuroscience Methods, 2010
- Clustering phenotype populations by genome‐wide RNAi and multiparametric imagingMolecular Systems Biology, 2010
- Single‐cell‐based image analysis of high‐throughput cell array screens for quantification of viral infectionCytometry Part A, 2008
- A multiresolution approach to automated classification of protein subcellular location imagesBMC Bioinformatics, 2007
- Alsterpaullone, a novel cyclin‐dependent kinase inhibitor, induces apoptosis by activation of caspase‐9 due to perturbation in mitochondrial membrane potentialMolecular Carcinogenesis, 2003
- Synthesis and adrenergic activity of benzimidazole bioisosteres of norepinephrine and isoproterenolJournal of Medicinal Chemistry, 1977