Rapid and accurate identification of marine microbes with single-cell Raman spectroscopy

The Analyst ◽  
2020 ◽  
Vol 145 (9) ◽  
pp. 3297-3305 ◽  
Author(s):  
Yaoyao Liu ◽  
Jingjing Xu ◽  
Yi Tao ◽  
Teng Fang ◽  
Wenbin Du ◽  
...  
Keyword(s):  
Neural Networks ◽  
Raman Spectroscopy ◽  
Single Cell ◽  
Raman Spectra ◽  
Convolutional Neural Networks ◽  
Marine Microbes ◽  
Accurate Identification ◽  
One Dimensional

Rapid and accurate identification of individual microorganisms using single-cell Raman spectra combining with one-dimensional convolutional neural networks.

scholarly journals One-dimensional convolutional neural networks for low/high arousal classification from electrodermal activity

2022 ◽  
Vol 71 ◽  
pp. 103203
Author(s):  
Roberto Sánchez-Reolid ◽  
Francisco López de la Rosa ◽  
María T. López ◽  
Antonio Fernández-Caballero
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Electrodermal Activity ◽  
High Arousal ◽  
One Dimensional

scholarly journals Cow Rump Identification Based on Lightweight Convolutional Neural Networks

Information ◽  
2021 ◽  
Vol 12 (9) ◽  
pp. 361
Author(s):  
Handan Hou ◽  
Wei Shi ◽  
Jinyan Guo ◽  
Zhe Zhang ◽  
Weizheng Shen ◽  
...  
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Dairy Cow ◽  
Individual Recognition ◽  
Animal Husbandry ◽  
Identification Accuracy ◽  
Individual Identification ◽  
Production Environment ◽  
Accurate Identification ◽  
Computing Device

Individual identification of dairy cows based on computer vision technology shows strong performance and practicality. Accurate identification of each dairy cow is the prerequisite of artificial intelligence technology applied in smart animal husbandry. While the rump of each dairy cow also has lots of important features, so do the back and head, which are also important for individual recognition. In this paper, we propose a non-contact cow rump identification method based on convolutional neural networks. First, the rump image sequences of the cows while feeding were collected. Then, an object detection model was applied to detect the cow rump object in each frame of image. Finally, a fine-tuned convolutional neural network model was trained to identify cow rumps. An image dataset containing 195 different cows was created to validate the proposed method. The method achieved an identification accuracy of 99.76%, which showed a better performance compared to other related methods and a good potential in the actual production environment of cow husbandry, and the model is light enough to be deployed in an edge-computing device.

scholarly journals Raman Spectroscopy-Based Measurements of Single-Cell Phenotypic Diversity in Microbial Populations

mSphere ◽  
2020 ◽  
Vol 5 (5) ◽  
Author(s):  
Cristina García-Timermans ◽  
Ruben Props ◽  
Boris Zacchetti ◽  
Myrsini Sakarika ◽  
Frank Delvigne ◽  
...  
Keyword(s):  
Raman Spectroscopy ◽  
Single Cell ◽  
Raman Spectra ◽  
Phenotypic Diversity ◽  
Cell Permeability ◽  
Microbial Populations ◽  
Molecular Profile ◽  
Microbial Cells ◽  
Content Type ◽  
Stressed Cells

ABSTRACT Microbial cells experience physiological changes due to environmental change, such as pH and temperature, the release of bactericidal agents, or nutrient limitation. This has been shown to affect community assembly and physiological processes (e.g., stress tolerance, virulence, or cellular metabolic activity). Metabolic stress is typically quantified by measuring community phenotypic properties such as biomass growth, reactive oxygen species, or cell permeability. However, bulk community measurements do not take into account single-cell phenotypic diversity, which is important for a better understanding and the subsequent management of microbial populations. Raman spectroscopy is a nondestructive alternative that provides detailed information on the biochemical makeup of each individual cell. Here, we introduce a method for describing single-cell phenotypic diversity using the Hill diversity framework of Raman spectra. Using the biomolecular profile of individual cells, we obtained a metric to compare cellular states and used it to study stress-induced changes. First, in two Escherichia coli populations either treated with ethanol or nontreated and then in two Saccharomyces cerevisiae subpopulations with either high or low expression of a stress reporter. In both cases, we were able to quantify single-cell phenotypic diversity and to discriminate metabolically stressed cells using a clustering algorithm. We also described how the lipid, protein, and nucleic acid compositions changed after the exposure to the stressor using information from the Raman spectra. Our results show that Raman spectroscopy delivers the necessary resolution to quantify phenotypic diversity within individual cells and that this information can be used to study stress-driven metabolic diversity in microbial populations. IMPORTANCE Microbial cells that live in the same community can exist in different physiological and morphological states that change as a function of spatiotemporal variations in environmental conditions. This phenomenon is commonly known as phenotypic heterogeneity and/or diversity. Measuring this plethora of cellular expressions is needed to better understand and manage microbial processes. However, most tools to study phenotypic diversity only average the behavior of the sampled community. In this work, we present a way to quantify the phenotypic diversity of microbial samples by inferring the (bio)molecular profile of its constituent cells using Raman spectroscopy. We demonstrate how this tool can be used to quantify the phenotypic diversity that arises after the exposure of microbes to stress. Raman spectroscopy holds potential for the detection of stressed cells in bioproduction.

scholarly journals A Spectral Feature Based Convolutional Neural Network for Classification of Sea Surface Oil Spill

2019 ◽  
Vol 8 (4) ◽  
pp. 160 ◽  
Author(s):  
Bingxin Liu ◽  
Ying Li ◽  
Guannan Li ◽  
Anling Liu
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Convolutional Neural Network ◽  
Convolutional Neural Networks ◽  
Machine Learning Algorithms ◽  
Classification Model ◽  
Support Vector ◽  
Oil Film ◽  
One Dimensional

Spectral characteristics play an important role in the classification of oil film, but the presence of too many bands can lead to information redundancy and reduced classification accuracy. In this study, a classification model that combines spectral indices-based band selection (SIs) and one-dimensional convolutional neural networks was proposed to realize automatic oil films classification using hyperspectral remote sensing images. Additionally, for comparison, the minimum Redundancy Maximum Relevance (mRMR) was tested for reducing the number of bands. The support vector machine (SVM), random forest (RF), and Hu’s convolutional neural networks (CNN) were trained and tested. The results show that the accuracy of classifications through the one dimensional convolutional neural network (1D CNN) models surpassed the accuracy of other machine learning algorithms such as SVM and RF. The model of SIs+1D CNN could produce a relatively higher accuracy oil film distribution map within less time than other models.

scholarly journals Single-Cell Phenotype Classification Using Deep Convolutional Neural Networks

2016 ◽  
Vol 21 (9) ◽  
pp. 998-1003 ◽  
Author(s):  
Oliver Dürr ◽  
Beate Sick
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
Single Cell ◽  
Convolutional Neural Networks ◽  
State Of The Art ◽  
Misclassification Rate ◽  
Support Vector ◽  
Learning Methods ◽  
Phenotype Classification

Deep learning methods are currently outperforming traditional state-of-the-art computer vision algorithms in diverse applications and recently even surpassed human performance in object recognition. Here we demonstrate the potential of deep learning methods to high-content screening–based phenotype classification. We trained a deep learning classifier in the form of convolutional neural networks with approximately 40,000 publicly available single-cell images from samples treated with compounds from four classes known to lead to different phenotypes. The input data consisted of multichannel images. The construction of appropriate feature definitions was part of the training and carried out by the convolutional network, without the need for expert knowledge or handcrafted features. We compare our results against the recent state-of-the-art pipeline in which predefined features are extracted from each cell using specialized software and then fed into various machine learning algorithms (support vector machine, Fisher linear discriminant, random forest) for classification. The performance of all classification approaches is evaluated on an untouched test image set with known phenotype classes. Compared to the best reference machine learning algorithm, the misclassification rate is reduced from 8.9% to 6.6%.

Sign in / Sign up

Export Citation Format

Share Document