We developed a large-scale morphological profiling approach that reads out a multitude of high-resolution biophysical fractal properties of single cells, based on our high-throughput quantitative phase imaging (QPI) platform (at 10,000 cells/sec). We showed that the single-cell morphological profile constructed by the sub-cellular fractal characteristics could be harnessed to implicate cell types and states in the context of identifying lung cancer cell line subtypes, assessing cell morphology in response to drug treatment, and tracking cell cycle progression.
We report the use of conditional generative adversarial network (cGAN) for restoring undersampled images captured in free-space angular-chirp-enhanced delay (FACED) microscopy. We show that this deep-learning approach allows the wider imaging field of view (FOV) along FACED axis, without substantially sacrificing the imaging resolution, photon-budget and speed even with lower density of scanning foci. This study could show the potential of further extending the applicability of FACED imaging to a wider range of biological applications that require extended FOV imaging.
We present an unprecedented, generative deep learning model (named beGAN) in reconstructing batch-effect-free quantitative phase image (QPI). By employing the high-throughput microfluidic multimodal imaging flow cytometry platform (i.e. multi-ATOM), our model demonstrated a robust QPI prediction from brightfield on various lung cancer cell lines (>800,000 cells). With batch-free QPI, biophysical phenotypes of cells are unified across batches and a significant improvement from 33.61% to 91.34% is achieved on the cross-batches cancer cell lines classification. This work unveil an avenue on overcoming batch effect with deep learning at single-cell imaging level.
We present a quantitative phase image (QPI) reconstruction method using generative deep learning (with high similarity of 91% and low error rate of < 1%), and its ability to integrate with a high-throughput microfluidic multimodal imaging flow cytometry platform (called multi-ATOM) that can consistently classify cancer cells in heterogeneous tumors from human non-small cell lung cancer patients at large scale (~200,000 cells) and high accuracy (~98%); and can reveal biophysical heterogeneity of tumors. This work represents another groundwork of synergizing high-throughput QPI and deep learning for future label-free intelligent clinical cancer diagnosis.
We report the use of high-throughput quantitative phase imaging (QPI) flow cytometry (based on multiplexed asymmetric-detection time-stretch optical microscopy (multi-ATOM)) to investigate biophysical profiles of single cells infected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This technique reveals the subtle biophysical heterogeneity of SARS-CoV-2 infection under the same multiplicity of infection. Furthermore, analyzing the label-free high-dimensional single-cell biophysical profiles (derived from multi-ATOM images) based on an unsupervised trajectory inference algorithm accurately recovers the infection progression over time. This study could offer biophysical insight of cellular morphogenesis of SARS-CoV-2 and shows the potential of label-free morphological profiling as a complementary drug discovery strategy for SARS-CoV-2.
Using a high-throughput imaging flow cytometer (10,000 cells/sec) multi-ATOM, we established a hierarchical biophysical phenotyping approach for label-free single-cell analysis. We demonstrate that the label-free multi-ATOM contrasts can be derived into a set of spatially hierarchical biophysical features that reflect optical density and dry mass density distributions in local and global scales. This phenotypic profile enables us to delineate subtle cellular response of molecularly targeted drug even at an early time point after the drug administration (6 hours). Based on fluorescence image analysis, we further interpreted how these biophysical phenotypes correlate with specific intracellular organelles alteration upon drug treatment.
We report a robust method based on generative deep learning to reconstruct quantitative phase image (QPI). By employing multiplexed asymmetric-detection time-stretch optical microscopy (multi-ATOM), we simultaneously captured multiple intensity image contrasts of the same cell in microfluidic flow, revealing different phase-gradient orientations at high throughput (10,000 cells/sec). Using conditional generative adversarial networks (cGAN), we performed a systematic analysis of how different orientations of the phase-gradient contrasts and their combinations influence the QPI prediction performance, which overall general achieves a high similarity (structural similarity index > 0.91) and low error rate (mean squared error < 0.01).
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.