Dataset Information

Label-Free High-Throughput Leukemia Detection by Holographic Microscopy.

ABSTRACT: Complete blood count and differentiation of leukocytes (DIFF) belong to the most frequently performed laboratory diagnostic tests. Here, a flow cytometry-based method for label-free DIFF of untouched leukocytes by digital holographic microscopy on the rich phase contrast of peripheral leukocyte images, using highly controlled 2D hydrodynamic focusing conditions is reported. Principal component analysis of morphological characteristics of the reconstructed images allows classification of nine leukocyte types, in addition to different types of leukemia and demonstrates disappearance of acute myeloid leukemia cells in remission. To exclude confounding effects, the classification strategy is tested by the analysis of 20 blinded clinical samples. Here, 70% of the specimens are correctly classified with further 20% classifications close to a correct diagnosis. Taken together, the findings indicate a broad clinical applicability of the cytometry method for automated and reagent-free diagnosis of hematological disorders.

SUBMITTER: Ugele M

PROVIDER: S-EPMC6299719 | biostudies-literature | 2018 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Label-Free High-Throughput Leukemia Detection by Holographic Microscopy.

Ugele Matthias M Weniger Markus M Stanzel Manfred M Bassler Michael M Krause Stefan W SW Friedrich Oliver O Hayden Oliver O Richter Lukas L

Advanced science (Weinheim, Baden-Wurttemberg, Germany) 20181011 12

Complete blood count and differentiation of leukocytes (DIFF) belong to the most frequently performed laboratory diagnostic tests. Here, a flow cytometry-based method for label-free DIFF of untouched leukocytes by digital holographic microscopy on the rich phase contrast of peripheral leukocyte images, using highly controlled 2D hydrodynamic focusing conditions is reported. Principal component analysis of morphological characteristics of the reconstructed images allows classification of nine leu ...[more]

PMID: 30581697

Similar Datasets

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.

Project description:Fluorescence spectroscopy and imaging are now used throughout the biosciences. Fluorescence microscopes, spectrofluorometers, microwell plate readers and microarray imagers all use multiple optical components to collect, redirect and focus the emission onto single point or array imaging detectors. For almost all biological samples, except those with regular nanoscale features, emission occurs in all directions. With the exception of complex microscope objectives with large collection angles (NA ? 0.5), all these instruments collect only a small fraction of the total emission. Because of the increasing knowledge base on fluorophores within near-field (<200 nm) distances from plasmonic and photonic structures we can anticipate the development of compact devices in which the sample to be detected is located directly on solid state detectors such as CCDs or CMOS cameras. Near-field interactions of fluorophores with metallic or dielectric multi-layer structures (MLSs) can capture a large fraction of the total emission. Depending on the composition and dimensions of the MLSs, the spatial distribution of the sample emission results in distinct optical patterns on the detector surface. With either plain glass slides or MLSs the most commonly used front focal plane (FFP) images reveal the x-y spatial distribution of emission from the sample. Another approach, which is often used with two or three-dimensional nanostructures, is back focal plane (BFP) imaging. The BFP images reveal the angular distribution of the emission. The FFP and BFP images occur at certain distances from the sample which is determined by the details of the optical components. Obtaining these images requires multiple optical components and distances which are too large for the compact devices. For devices described in this paper, the images will be detected at a fixed distance between the sample and some arbitrary distance below the MLS which is determined by the geometry and thicknesses of the components. We refer to measurements at these locations as out-of-focal plane (OFP) imaging. Herein we describe a method to measure the optical fields at micron and multi-micron distances below the MLS, which will represent the images seen by an optically coupled array detector. The possibility of sub-surface optical images is illustrated using five different multi-layer structures. This is accomplished using an optical configuration which allows measurement at a front focal plane (FFP), back focal plane (BFP) or any OFP locations. Our OFP imaging method provides a link between the FFP images which reveals the surface distribution of fluorophores with the BFP images that reveal the angular distribution of emission. This linkage can be useful when examining structures which have nanoscale features due to fluorescence or leakage radiation from nanostructures.

Dataset Information

Label-Free High-Throughput Leukemia Detection by Holographic Microscopy.

Publications

Label-Free High-Throughput Leukemia Detection by Holographic Microscopy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets