David Mayerich

High definition infrared spectroscopic imaging for lymph node histopathology.

Chemical imaging is a rapidly emerging field in which molecular information within samples can be used to predict biological function and recognize disease without the use of stains or manual identification. In Fourier transform infrared (FT-IR) spectroscopic imaging, molecular absorption contrast provides a large signal relative to noise. Due to the long mid-IR wavelengths and sub-optimal instrument design, however, pixel sizes have historically been much larger than cells. This limits both the accuracy of the technique in identifying small regions, as well as the ability to visualize single cells. Here we obtain data with micron-sized sampling using a tabletop FT-IR instrument, and demonstrate that the high-definition (HD) data lead to accurate identification of multiple cells in lymph nodes that was not previously possible. Highly accurate recognition of eight distinct classes - naïve and memory B cells, T cells, erythrocytes, connective tissue, fibrovascular network, smooth muscle, and light and dark zone activated B cells was achieved in healthy, reactive, and malignant lymph node biopsies using a random forest classifier. The results demonstrate that cells currently identifiable only through immunohistochemical stains and cumbersome manual recognition of optical microscopy images can now be distinguished to a similar level through a single IR spectroscopic image from a lymph node biopsy.

Comparison of rotational imaging optical coherence tomography and selective plane illumination microscopy for embryonic study

The mouse is a common model for studying developmental diseases. Different optical techniques have been developed to investigate mouse embryos, but each has its own set of limitations and restrictions. In this study, we imaged the same E9.5 mouse embryo with rotational imaging Optical Coherence Tomography (RI-OCT) and Selective Plane Illumination Microscopy (SPIM), and compared the two techniques. Results demonstrate that both methods can provide images with micrometer-scale spatial resolution. The RI-OCT technique was developed to increase imaging depth of OCT by performing traditional OCT imaging at multiple sides and co-registering the images. In SPIM, optical sectioning is achieved by illuminating the sample with a sheet of light. In this study, the images acquired from both techniques are compared with each other to evaluate the benefits and drawbacks of each technique for embryonic imaging. Since 3D stacks can be obtained by SPIM from different angles by rotating the sample, it might be possible to build a hybrid setup of two imaging modalities to combine the advantages of each technique.

Plasmonic nanoparticle-based expansion microscopy with surface-enhanced Raman and dark-field spectroscopic imaging

Fluorescence-based expansion microscopy (ExM) is a new technique which can yield nanoscale resolution of biological specimen on a conventional fluorescence microscope through physical sample expansion up to 20 times its original dimensions while preserving structural information. It however inherits known issues of fluorescence microscopy such as photostability and multiplexing capabilities, as well as an ExM-specific issue in signal intensity reduction due to a dilution effect after expansion. To address these issues, we propose using antigen-targeting plasmonic nanoparticle labels which can be imaged using surface-enhanced Raman scattering spectroscopy (SERS) and dark-field spectroscopy. We demonstrate that the nanoparticles enable multimodal imaging: bright-field, dark-field and SERS, with excellent photostability, contrast enhancement and brightness.

Selecting optimal features from Fourier transform infrared spectroscopy for discrete-frequency imaging

Tissue histology utilizing chemical and immunohistochemical labels plays an important role in biomedicine and disease diagnosis. Recent research suggests that mid-infrared (IR) spectroscopic imaging may augment histology by providing quantitative molecular information. One of the major barriers to this approach is long acquisition time using Fourier-transform infrared (FTIR) spectroscopy. Recent advances in discrete frequency sources, particularly quantum cascade lasers (QCLs), may mitigate this problem by allowing selective sampling of the absorption spectrum. However, DFIR imaging only provides a significant advantage when the number of spectral samples is minimized, requiring a priori knowledge of important spectral features. In this paper, we demonstrate the use of a GPU-based genetic algorithm (GA) using linear discriminant analysis (LDA) for DFIR feature selection. Our proposed method relies on pre-acquired broadband FTIR images for feature selection. Based on user-selected criteria for classification accuracy, our algorithm provides a minimal set of features that can be used with DFIR in a time-frame more practical for clinical diagnosis.

Imaging and Feature Selection Using GA-FDA Algorithm for the Classification of Mid-Infrared Biomedical Images

Pathologists currently rely on chemical staining of tissue samples to perform disease diagnosis. However, these techniques are highly prone to variations due to the clinical environment and staining protocol. Midinfrared spectroscopic imaging has the potential to overcome several of these problems by providing quantitative molecular information that can be used for highly specific tissue classification [1] [4]. However, there are bottlenecks that limit clinical applicability of these methods. For example, Fouriertransform infrared (FTIR) spectroscopic images are often time-consuming to acquire at a spectral resolution and SNR level that is viable for reliable classification. In addition, algorithms are computationally intensive and data sets can be terabytes in size. This research focuses on improving the clinical viability of mid-infrared spectroscopy by utilizing a QCL based imaging system, which allows feature selection from hyperspectral images (HSI) using a Genetic Algorithm and Fisher’s Discriminant Analysis as a fitness function (GA-FDA) and classification of HSI images using Random Forest classifier with the optimized feature subsets selected by GA-FDA [3]. We demonstrate that GA-FDA is very promising feature selection algorithm with performance superior to unsupervised as well as supervised feature selection algorithms while being compatible with optimized QCL imaging methods.

Robust Cell Detection for Large-Scale 3D Microscopy Using GPU-Accelerated Iterative Voting

High-throughput imaging techniques, such as Knife-Edge Scanning Microscopy (KESM),are capable of acquiring three-dimensional whole-organ images at sub-micrometer resolution. These images are challenging to segment since they can exceed several terabytes (TB) in size, requiring extremely fast and fully automated algorithms. Staining techniques are limited to contrast agents that can be applied to large samples and imaged in a single pass. This requires maximizing the number of structures labeled in a single channel, resulting in images that are densely packed with spatial features. In this paper, we propose a three-dimensional approach for locating cells based on iterative voting. Due to the computational complexity of this algorithm, a highly efficient GPU implementation is required to make it practical on large data sets. The proposed algorithm has a limited number of input parameters and is highly parallel.

TIMING 2.0: high-throughput single-cell profiling of dynamic cell–cell interactions by time-lapse imaging microscopy in nanowell grids

Motivation: Automated profiling of cell‐cell interactions from high‐throughput time‐lapse imaging microscopy data of cells in nanowell grids (TIMING) has led to fundamental insights into cell‐cell interactions in immunotherapy. This application note aims to enable widespread adoption of TIMING by (i) enabling the computations to occur on a desktop computer with a graphical processing unit instead of a server; (ii) enabling image acquisition and analysis to occur in the laboratory avoiding network data transfers to/from a server and (iii) providing a comprehensive graphical user interface. Results: On a desktop computer, TIMING 2.0 takes 5 s/block/image frame, four times faster than our previous method on the same computer, and twice as fast as our previous method (TIMING) running on a Dell PowerEdge server. The cell segmentation accuracy (f‐number = 0.993) is superior to our previous method (f‐number = 0.821). A graphical user interface provides the ability to inspect the video analysis results, make corrective edits efficiently (one‐click editing of an entire nanowell video sequence in 5‐10 s) and display a summary of the cell killing efficacy measurements. Availability and implementation: Open source Python software (GPL v3 license), instruction manual, sample data and sample results are included with the Supplement (https://github.com/RoysamLab/TIMING2). Supplementary information: Supplementary data are available at Bioinformatics online.

A dual-modality optical coherence tomography and selective plane illumination microscopy system for mouse embryonic imaging

Both optical coherence tomography (OCT) and selective plane illumination microscopy (SPIM) are frequently used in mouse embryonic research for high-resolution three-dimensional imaging. However, each of these imaging methods provide a unique and independent advantage: SPIM provides morpho-functional information through immunofluorescence and OCT provides a method for whole-embryo 3D imaging. In this study, we have combined rotational imaging OCT and SPIM into a single, dual-modality device to image E9.5 mouse embryos. The results demonstrate that the dual-modality setup is able to provide both anatomical and functional information simultaneously for more comprehensive tissue characterization.

Multimodal embryonic imaging using optical coherence tomography, selective plane illumination microscopy, and optical projection tomography

The murine model is commonly utilized for studying developmental diseases. Different optical techniques have been developed to image mouse embryos, but each has its own set of limitations and restrictions. In this study, we compare the performance of the well-established technique of optical coherence tomography (OCT) to the relatively new methods of selective plane illumination microscopy (SPIM) and optical projection tomography (OPT) to assess murine embryonic development. OCT can provide label free high resolution images of the mouse embryo, but suffers from light attenuation that limits visualization of deeper structures. SPIM is able to image shallow regions with great detail utilizing fluorescent contrast. OPT can provide superior imaging depth, and can also use fluorescence labels but, it requires samples to be fixed and cleared before imaging. OCT requires no modification of the embryo, and thus, can be used in vivo and in utero. In this study, we compare the efficacy of OCT, SPIM, and OPT for imaging murine embryonic development. The data demonstrate the superior capability of SPIM and OPT for imaging fine structures with high resolution while only OCT can provide structural and functional imaging of live embryos with micrometer scale resolution.

Fast submicrometer-scale imaging of whole zebrafish using the knife-edge scanning microscope

Advances in high-resolution 3D microscopy have enabled the investigation of subcellular microstructures in biological specimen. For a full understanding of the organisms structure and function, it is mandatory to obtain data from the whole animal, not just parts of it. In this paper, we present our work with the Knife-Edge Scanning Microscope (KESM) for imaging a Nissl-stained whole zebrafish larva. KESM combines a diamond microtome and line-scan imaging for simultaneous sectioning and imaging in 3D. We show that using the KESM, a zebrafish, less than 3 mm long and diameter less than 500 μm, can be imaged within 1 hour at a resolution of 0.6 μm × 0.7 μm × 1.0 μm. We also present new results on using a vibrating microtome to improve sectioning and imaging robustness.Advances in high-resolution 3D microscopy have enabled the investigation of subcellular microstructures in biological specimen. For a full understanding of the organisms structure and function, it is mandatory to obtain data from the whole animal, not just parts of it. In this paper, we present our work with the Knife-Edge Scanning Microscope (KESM) for imaging a Nissl-stained whole zebrafish larva. KESM combines a diamond microtome and line-scan imaging for simultaneous sectioning and imaging in 3D. We show that using the KESM, a zebrafish, less than 3 mm long and diameter less than 500 μm, can be imaged within 1 hour at a resolution of 0.6 μm × 0.7 μm × 1.0 μm. We also present new results on using a vibrating microtome to improve sectioning and imaging robustness.