Raman spectroscopy, combined with machine learning techniques, holds great promise for many applications as a rapid, sensitive, and label-free identification method. Such approaches perform well when classifying spectra of chemical species that were encountered during the training phase. That is, species that are known to the neural network. However, in real-world settings, such as in clinical applications, there will always be substances whose spectra have not yet been taken. When the neural network encounters these new species during the testing phase, the number of false positives becomes uncontrollable, limiting the usefulness of these techniques, especially in public safety applications. To overcome these barriers, we implemented the recently introduced Entropic Open Set and Objectosphere loss functions. To demonstrate the efficacy and efficiency of this approach, we compiled a database of hyperspectral Raman images of 40 chemical species separating them into three class categorizations. The known class consisted of 20 biologically relevant species comprising amino acids, the ignored class was 10 “irrelevant” species comprising bio-related chemicals, and the never seen before class was 10 various chemical species that the neural network had not seen before. We show that this approach not only enables the network to effectively separate the unknown species while preserving high accuracy on the known ones and reducing false positives but also performs better than the current gold standards in machine learning techniques. This opens the door to using Raman spectroscopy, combined with our novel machine learning algorithm, in a variety of practical applications. Availability and implementation: freely available on the web at .
Total internal reflection fluorescence microscopy with polarized excitation (P-TIRF) can be used to image nanoscale curvature phenomena in live cells. We used P-TIRF to visualize rat basophilic leukemia cells (RBL-2H3 cells) primed with fluorescent anti-dinitrophenyl (anti-DNP) immunoglobulin E (IgE) coming into contact with a supported lipid bilayer containing mobile, monovalent DNP, modeling an immunological synapse. The spatial relationship of the IgE-bound high affinity IgE receptor (FcεRI) to the ratio image of P-polarized excitation and S-polarized excitation was analyzed. These studies help correlate the dynamics of cell surface molecules with the mechanical properties of the plasma membrane during synapse formation.
Background Structured illumination microscopy (SIM) is a family of methods in optical fluorescence microscopy that can achieve both optical sectioning and super-resolution effects. SIM is a valuable method for high resolution imaging of fixed cells or tissues labeled with conventional fluorophores, as well as for imaging the dynamics of live cells expressing fluorescent protein constructs. In SIM, one acquires a set of images with shifting illumination patterns. This set of images is subsequently treated with image analysis algorithms to produce an image with reduced out-of-focus light (optical sectioning) and/or with improved resolution (super-resolution).Findings Five complete and freely available SIM datasets are presented including raw and analyzed data. We report methods for image acquisition and analysis using open source software along with examples of the resulting images when processed with different methods. We processed the data using established optical sectioning SIM and superresolution SIM methods, and with newer Bayesian restoration approaches which we are developing. ConclusionVarious methods for SIM data acquisition and processing are actively being developed, but complete raw data from SIM experiments is not typically published. Publicly available, high quality raw data with examples of processed results will aid researchers when developing new methods in SIM. Biologists will also find interest in the high-resolution images of animal tissues and cells we acquired. All of the data was processed with SIMToolbox, an open source and freely available software solution for SIM.
BackgroundStructured illumination microscopy (SIM) is a family of methods in optical fluorescence microscopy that can achieve both optical sectioning and super-resolution effects. SIM is a valuable method for high-resolution imaging of fixed cells or tissues labeled with conventional fluorophores, as well as for imaging the dynamics of live cells expressing fluorescent protein constructs. In SIM, one acquires a set of images with shifting illumination patterns. This set of images is subsequently treated with image analysis algorithms to produce an image with reduced out-of-focus light (optical sectioning) and/or with improved resolution (super-resolution).FindingsFive complete, freely available SIM datasets are presented including raw and analyzed data. We report methods for image acquisition and analysis using open-source software along with examples of the resulting images when processed with different methods. We processed the data using established optical sectioning SIM and super-resolution SIM methods and with newer Bayesian restoration approaches that we are developing.ConclusionsVarious methods for SIM data acquisition and processing are actively being developed, but complete raw data from SIM experiments are not typically published. Publically available, high-quality raw data with examples of processed results will aid researchers when developing new methods in SIM. Biologists will also find interest in the high-resolution images of animal tissues and cells we acquired. All of the data were processed with SIMToolbox, an open-source and freely available software solution for SIM.
Background Fluorescence microscopy is an important technique in many areas of biological research. Two factors that limit the usefulness and performance of fluorescence microscopy are photobleaching of fluorescent probes during imaging and, when imaging live cells, phototoxicity caused by light exposure. Recently developed methods in machine learning are able to greatly improve the signal-to-noise ratio of acquired images. This allows researchers to record images with much shorter exposure times, which in turn minimizes photobleaching and phototoxicity by reducing the dose of light reaching the sample. Findings To use deep learning methods, a large amount of data is needed to train the underlying convolutional neural network. One way to do this involves use of pairs of fluorescence microscopy images acquired with long and short exposure times. We provide high-quality datasets that can be used to train and evaluate deep learning methods under development. Conclusion The availability of high-quality data is vital for training convolutional neural networks that are used in current machine learning approaches.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.