Dataset Information

A High Throughput and Unbiased Machine Learning Approach for Classification of Graphene Dispersions.

ABSTRACT: Significant research to define and standardize terminologies for describing stacks of atomic layers in bulk graphene materials has been undertaken. Most methods to measure the stacking characteristics are time consuming and are not suited for obtaining information by directly imaging dispersions. Conventional optical microscopy has difficulty in identifying the size and thickness of a few layers of graphene stacks due to their low photon absorption capacity. Utilizing a contrast based on anisotropic refractive index in 2D materials, it is shown that localized thickness-specific information can be captured in birefringence images of graphene dispersions. Coupling pixel-by-pixel information from brightfield and birefringence images and using unsupervised statistical learning algorithms, three unique data clusters representing flakes (unexfoliated), nanoplatelets (partially exfoliated), and 2D sheets (well-exfoliated) species in various laboratory-based and commercial dispersions of graphene and graphene oxide are identified. The high-throughput, multitasking capability of the approach to classify stacking at sub-nanometer to micrometer scale and measure the size, thickness, and concentration of exfoliated-species in generic dispersions of graphene/graphene oxide are demonstrated. The method, at its current stage, requires less than half an hour to quantitatively assess one sample of graphene/graphene oxide dispersion.

SUBMITTER: Abedin MJ

PROVIDER: S-EPMC7578897 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A High Throughput and Unbiased Machine Learning Approach for Classification of Graphene Dispersions.

Abedin Md Joynul MJ Barua Titon T Shaibani Mahdokht M Majumder Mainak M

Advanced science (Weinheim, Baden-Wurttemberg, Germany) 20200825 20

Significant research to define and standardize terminologies for describing stacks of atomic layers in bulk graphene materials has been undertaken. Most methods to measure the stacking characteristics are time consuming and are not suited for obtaining information by directly imaging dispersions. Conventional optical microscopy has difficulty in identifying the size and thickness of a few layers of graphene stacks due to their low photon absorption capacity. Utilizing a contrast based on anisotr ...[more]

PMID: 33101862

Similar Datasets

Project description:BackgroundRacial disparities in health care are well documented in the United States. As machine learning methods become more common in health care settings, it is important to ensure that these methods do not contribute to racial disparities through biased predictions or differential accuracy across racial groups.ObjectiveThe goal of the research was to assess a machine learning algorithm intentionally developed to minimize bias in in-hospital mortality predictions between white and nonwhite patient groups.MethodsBias was minimized through preprocessing of algorithm training data. We performed a retrospective analysis of electronic health record data from patients admitted to the intensive care unit (ICU) at a large academic health center between 2001 and 2012, drawing data from the Medical Information Mart for Intensive Care-III database. Patients were included if they had at least 10 hours of available measurements after ICU admission, had at least one of every measurement used for model prediction, and had recorded race/ethnicity data. Bias was assessed through the equal opportunity difference. Model performance in terms of bias and accuracy was compared with the Modified Early Warning Score (MEWS), the Simplified Acute Physiology Score II (SAPS II), and the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE).ResultsThe machine learning algorithm was found to be more accurate than all comparators, with a higher sensitivity, specificity, and area under the receiver operating characteristic. The machine learning algorithm was found to be unbiased (equal opportunity difference 0.016, P=.20). APACHE was also found to be unbiased (equal opportunity difference 0.019, P=.11), while SAPS II and MEWS were found to have significant bias (equal opportunity difference 0.038, P=.006 and equal opportunity difference 0.074, P<.001, respectively).ConclusionsThis study indicates there may be significant racial bias in commonly used severity scoring systems and that machine learning algorithms may reduce bias while improving on the accuracy of these methods.

Project description:BackgroundRhodopsin is a seven-transmembrane protein covalently linked with retinal chromophore that absorbs photons for energy conversion and intracellular signaling in eukaryotes, bacteria, and archaea. Haloarchaeal rhodopsins are Type-I microbial rhodopsin that elicits various light-driven functions like proton pumping, chloride pumping and Phototaxis behaviour. The industrial application of Ion-pumping Haloarchaeal rhodopsins is limited by the lack of full-length rhodopsin sequence-based classifications, which play an important role in Ion-pumping activity. The well-studied Haloarchaeal rhodopsin is a proton-pumping bacteriorhodopsin that shows promising applications in optogenetics, biosensitized solar cells, security ink, data storage, artificial retinal implant and biohydrogen generation. As a result, a low-cost computational approach is required to identify Ion-pumping Haloarchaeal rhodopsin sequences and its subtype.ResultsThis study uses a support vector machine (SVM) technique to identify these ion-pumping Haloarchaeal rhodopsin proteins. The haloarchaeal ion pumping rhodopsins viz., bacteriorhodopsin, halorhodopsin, xanthorhodopsin, sensoryrhodopsin and marine prokaryotic Ion-pumping rhodopsins like actinorhodopsin, proteorhodopsin have been utilized to develop the methods that accurately identified the ion pumping haloarchaeal and other type I microbial rhodopsins. We achieved overall maximum accuracy of 97.78%, 97.84% and 97.60%, respectively, for amino acid composition, dipeptide composition and hybrid approach on tenfold cross validation using SVM. Predictive models for each class of rhodopsin performed equally well on an independent data set. In addition to this, similar results were achieved using another machine learning technique namely random forest. Simultaneously predictive models performed equally well during five-fold cross validation. Apart from this study, we also tested the own, blank, BLAST dataset and annotated whole-genome rhodopsin sequences of PWS haloarchaeal isolates in the developed methods. The developed web server ( https://bioinfo.imtech.res.in/servers/rhodopred ) can identify the Ion Pumping Haloarchaeal rhodopsin proteins and their subtypes. We expect this web tool would be useful for rhodopsin researchers.ConclusionThe overall performance of the developed method results show that it accurately identifies the Ionpumping Haloarchaeal rhodopsin and their subtypes using known and unknown microbial rhodopsin sequences. We expect that this study would be useful for optogenetics, molecular biologists and rhodopsin researchers.

Project description:BackgroundPlankton are foundational to marine food webs and an important feature for characterizing ocean health. Recent developments in quantitative imaging devices provide in-flow high-throughput sampling from bulk volumes-opening new ecological challenges exploring microbial eukaryotic variation and diversity, alongside technical hurdles to automate classification from large datasets. However, a limited number of deployable imaging instruments have been coupled with the most prominent classification algorithms-effectively limiting the extraction of curated observations from field deployments. Holography offers relatively simple coherent microscopy designs with non-intrusive 3-D image information, and rapid frame rates that support data-driven plankton imaging tasks. Classification benchmarks across different domains have been set with transfer learning approaches, focused on repurposing pre-trained, state-of-the-art deep learning models as classifiers to learn new image features without protracted model training times. Combining the data production of holography, digital image processing, and computer vision could improve in-situ monitoring of plankton communities and contribute to sampling the diversity of microbial eukaryotes.ResultsHere we use a light and portable digital in-line holographic microscope (The HoloSea) with maximum optical resolution of 1.5 μm, intensity-based object detection through a volume, and four different pre-trained convolutional neural networks to classify > 3800 micro-mesoplankton (> 20 μm) images across 19 classes. The maximum classifier performance was quickly achieved for each convolutional neural network during training and reached F1-scores > 89%. Taking classification further, we show that off-the-shelf classifiers perform strongly across every decision threshold for ranking a majority of the plankton classes.ConclusionThese results show compelling baselines for classifying holographic plankton images, both rare and plentiful, including several dinoflagellate and diatom groups. These results also support a broader potential for deployable holographic microscopes to sample diverse microbial eukaryotic communities, and its use for high-throughput plankton monitoring.

Dataset Information

A High Throughput and Unbiased Machine Learning Approach for Classification of Graphene Dispersions.

Publications

A High Throughput and Unbiased Machine Learning Approach for Classification of Graphene Dispersions.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets