We present an extreme-throughput (>1 million cells per second) imaging flow cytometer with deep learning to achieve a highly simple, rapid, and cost-effective liquid biopsy for ex-vivo drug-susceptibility testing of leukemia. The drug resistance of leukemia cells was detected in whole blood with only 24-hour drug treatment without hemolysis or dilution, making the sample preparation extremely simple, rapid and cost-effective. Our method also accurately evaluated the drug susceptibility of white blood cells from untreated patients with acute lymphoblastic leukemia, holding great promise for affordable precision medicine.
Acquired drug resistance is a fundamental predicament in cancer therapy. Early detection of drug-resistant cancer cells during or after treatment is expected to benefit patients from unnecessary drug administration and thus play a significant role in the development of a therapeutic strategy. However, the development of an effective method of detecting drug-resistant cancer cells is still in its infancy due to their complex mechanism in drug resistance. To address this problem, we propose and experimentally demonstrate label-free image-based drug resistance detection with optofluidic time-stretch microscopy using leukemia cells (K562 and K562/ADM). By adding adriamycin (ADM) to both K562 and K562/ADM (ADM-resistant K562 cells) cells, both types of cells express unique morphological changes, which are subsequently captured by an optofluidic time-stretch microscope. These unique morphological changes are extracted as image features and are subjected to supervised machine learning for cell classification. We hereby have successfully differentiated K562 and K562/ADM solely with label-free images, which suggests that our technique is capable of detecting drug-resistant cancer cells. Our optofluidic time-stretch microscope consists of a time-stretch microscope with a high spatial resolution of 780 nm at a 1D frame rate of 75 MHz and a microfluidic device that focuses and orders cells. We compare various machine learning algorithms as well as various concentrations of ADM for cell classification. Owing to its unprecedented versatility of using label-free image and its independency from specific molecules, our technique holds great promise for detecting drug resistance of cancer cells for which its underlying mechanism is still unknown or chemical probes are still unavailable.
KEYWORDS: In vivo imaging, Visualization, Digital image processing, Statistical analysis, Microscopy, Blood, Prostate cancer, Breast cancer, Spatial resolution, Microfluidics
According to WHO, approximately 10 million new cases of thrombotic disorders are diagnosed worldwide every year. In the U.S. and Europe, their related diseases kill more people than those from AIDS, prostate cancer, breast cancer and motor vehicle accidents combined. Although thrombotic disorders, especially arterial ones, mainly result from enhanced platelet aggregability in the vascular system, visual detection of platelet aggregates in vivo is not employed in clinical settings. Here we present a high-throughput label-free platelet aggregate detection method, aiming at the diagnosis and monitoring of thrombotic disorders in clinical settings. With optofluidic time-stretch microscopy with a spatial resolution of 780 nm and an ultrahigh linear scanning rate of 75 MHz, it is capable of detecting aggregated platelets in lysed blood which flows through a hydrodynamic-focusing microfluidic device at a high throughput of 10,000 particles/s. With digital image processing and statistical analysis, we are able to distinguish them from single platelets and other blood cells via morphological features. The detection results are compared with results of fluorescence-based detection (which is slow and inaccurate, but established). Our results indicate that the method holds promise for real-time, low-cost, label-free, and minimally invasive detection of platelet aggregates, which is potentially applicable to detection of platelet aggregates in vivo and to the diagnosis and monitoring of thrombotic disorders in clinical settings. This technique, if introduced clinically, may provide important clinical information in addition to that obtained by conventional techniques for thrombotic disorder diagnosis, including ex vivo platelet aggregation tests.
The ability to sift through a large heterogeneous population of cells is of paramount importance in a diverse range of biomedical and green applications. Furthermore, the capability of identifying various features of cells in a label-free manner is useful for high-throughput screening. Here we present optofluidic time-stretch quantitative phase microscopy for high-throughput label-free single-cell screening. This method is based on an integration of a hydrodynamic-focusing microfluidic chip, an optical time-stretch microscope for high-speed imaging with a spatial resolution of ~800 nm at a frame rate of ~10 million frames per second, and a digital image processor for image-based characterization, classification, and statistical analysis of biological cells such as blood cells and microalgae. It provides both the opacity (amplitude) and thickness (phase) content of every cell at a high throughput of ~10,000 cells per second. This method is expected to be effective for a diverse range of applications such as cancer detection and biofuel production.
The development of reliable, sustainable, and economical sources of alternative fuels is an important, but challenging goal for the world. As an alternative to liquid fossil fuels, microalgal biofuel is expected to play a key role in reducing the detrimental effects of global warming since microalgae absorb atmospheric CO2 via photosynthesis. Unfortunately, conventional analytical methods only provide population-averaged lipid contents and fail to characterize a diverse population of microalgal cells with single-cell resolution in a noninvasive and interference-free manner. Here we demonstrate high-throughput label-free single-cell screening of lipid-producing microalgal cells with optofluidic time-stretch quantitative phase microscopy. In particular, we use Euglena gracilis – an attractive microalgal species that produces wax esters (suitable for biodiesel and aviation fuel after refinement) within lipid droplets. Our optofluidic time-stretch quantitative phase microscope is based on an integration of a hydrodynamic-focusing microfluidic chip, an optical time-stretch phase-contrast microscope, and a digital image processor equipped with machine learning. As a result, it provides both the opacity and phase contents of every single cell at a high throughput of 10,000 cells/s. We characterize heterogeneous populations of E. gracilis cells under two different culture conditions to evaluate their lipid production efficiency. Our method holds promise as an effective analytical tool for microalgaebased biofuel production.
The world is faced with environmental problems and the energy crisis due to the combustion and depletion of fossil
fuels. The development of reliable, sustainable, and economical sources of alternative fuels is an important, but
challenging goal for the world. As an alternative to liquid fossil fuels, algal biofuel is expected to play a key role in
alleviating global warming since algae absorb atmospheric CO2 via photosynthesis. Among various algae for fuel
production, Euglena gracilis is an attractive microalgal species as it is known to produce wax ester (good for biodiesel
and aviation fuel) within lipid droplets. To date, while there exist many techniques for inducing microalgal cells to
produce and accumulate lipid with high efficiency, few analytical methods are available for characterizing a population
of such lipid-accumulated microalgae including E. gracilis with high throughout, high accuracy, and single-cell
resolution simultaneously. Here we demonstrate a high-throughput optofluidic Euglena gracilis profiler which consists
of an optical time-stretch microscope and a fluorescence analyzer on top of an inertial-focusing microfluidic device that
can detect fluorescence from lipid droplets in their cell body and provide images of E. gracilis cells simultaneously at a
high throughput of 10,000 cells/s. With the multi-dimensional information acquired by the system, we classify nitrogen-sufficient
(ordinary) and nitrogen-deficient (lipid-accumulated) E. gracilis cells with a low false positive rate of 1.0%.
This method provides a promise for evaluating the efficiency of lipid-inducing techniques for biofuel production, which
is also applicable for identifying biomedical samples such as blood cells and cancer cells.
Routine procedures in standard histopathology involve laborious steps of tissue processing and staining for final examination. New techniques which can bypass these procedures and thus minimize the tissue handling error would be of great clinical value. Coherent anti-Stokes Raman scattering (CARS) microscopy is an attractive tool for label-free biochemical-specific characterization of biological specimen. However, a vast majority of prior works on CARS (or stimulated Raman scattering (SRS)) bioimaging restricted analyses on a narrowband or well-distinctive Raman spectral signatures. Although hyperspectral SRS/CARS imaging has recently emerged as a better solution to access wider-band spectral information in the image, studies mostly focused on a limited spectral range, e.g. CH-stretching vibration of lipids, or non-biological samples. Hyperspectral image information in the congested fingerprint spectrum generally remains untapped for biological samples. In this regard, we further explore ultrabroadband hyperspectral multiplex (HM-CARS) to perform chemoselective histological imaging with the goal of exploring its utility in stain-free clinical histopathology. Using the supercontinuum Stokes, our system can access the CARS spectral window as wide as >2000cm-1. In order to unravel the congested CARS spectra particularly in the fingerprint region, we first employ a spectral phase-retrieval algorithm based on Kramers–Kronig (KK) transform to minimize the non-resonant background in the CARS spectrum. We then apply principal component analysis (PCA) to identify and map the spatial distribution of different biochemical components in the tissues. We demonstrate chemoselective HM-CARS imaging of a colon tissue section which displays the key cellular structures that correspond well with standard stained-tissue observation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.