Detection of live breast cancer cells in bright-field microscopy images containing white blood cells by image analysis and deep learning

An enhanced multimode phase imaging method based on the transport of intensity equation

Label-free biological cell imaging relies on rapid multimode phase imaging of biological samples in natural settings. To improve image contrast, phase is encoded into intensity information using the differential interference contrast (DIC) and Zernike phase contrast (ZPC) techniques. To enable multimode contrast-enhanced observation of unstained specimens, this paper proposes an improved multimode phase imaging method based on the transport of intensity equation (TIE), which combines conventional microscopy with computational imaging. The ZPC imaging module based on adaptive aperture adjustment is applied when the quantitative phase results of biological samples have been obtained by solving the TIE. Simultaneously, a rotationally symmetric shear-based technique is used that can yield isotropic DIC. In this paper, we describe numerical simulation and optical experiments carried out to validate the accuracy and viability of this technology. The calculated Michelson contrast of the ZPC image in the resolution plate experiment increased from 0.196 to 0.394.

High-Fidelity Sensitive Tracing Circulating Tumor Cell Telomerase Activity

Dynamic tracing of intracellular telomerase activity plays a crucial role in cancer cell recognition and correspondingly in earlier cancer diagnosis and personalized precision therapy. However, due to the complexity of the required reaction system and insufficient loading of reaction components into cells, achieving a high-fidelity determination of telomerase activity is still a challenge. Herein, an Aptamer-Liposome mediated Telomerase activated poly-Molecular beacon Arborescent Nanoassembly(ALTMAN) approach was described for direct high-fidelity visualization of telomerase activity. Briefly, intracellular telomerase activates molecular beacons, causing their hairpin structures to unfold and produce fluorescent signals. Furthermore, multiple molecular beacons can self-assemble, forming arborescent nanostructures and leading to exponential amplification of fluorescent signals. Integrating the enzyme-free isothermal signal amplification successfully increased the sensitivity and reduced interference by leveraging the skillful design of the molecular beacon and the extension of the telomerase-activated TTAGGG repeat sequence. The proposed approach enabled ultrasensitive visualization of activated telomerase exclusively with a prominent detection limit of 2 cells·μL-1 and realized real-time imaging of telomerase activity in living cancer cells including blood samples from breast cancer patients and urine samples from bladder cancer patients. This approach opens an avenue for establishing a telomerase activity determination and in situ monitoring technique that can facilitate both telomerase fundamental biological studies and cancer diagnostics.

Morphology-based deep learning enables accurate detection of senescence in mesenchymal stem cell cultures

BACKGROUND

Cell senescence is a sign of aging and plays a significant role in the pathogenesis of age-related disorders. For cell therapy, senescence may compromise the quality and efficacy of cells, posing potential safety risks. Mesenchymal stem cells (MSCs) are currently undergoing extensive research for cell therapy, thus necessitating the development of effective methods to evaluate senescence. Senescent MSCs exhibit distinctive morphology that can be used for detection. However, morphological assessment during MSC production is often subjective and uncertain. New tools are required for the reliable evaluation of senescent single cells on a large scale in live imaging of MSCs.

RESULTS

We have developed a successful morphology-based Cascade region-based convolution neural network (Cascade R-CNN) system for detecting senescent MSCs, which can automatically locate single cells of different sizes and shapes in multicellular images and assess their senescence state. Additionally, we tested the applicability of the Cascade R-CNN system for MSC senescence and examined the correlation between morphological changes with other senescence indicators.

CONCLUSIONS

This deep learning has been applied for the first time to detect senescent MSCs, showing promising performance in both chronic and acute MSC senescence. The system can be a labor-saving and cost-effective option for screening MSC culture conditions and anti-aging drugs, as well as providing a powerful tool for non-invasive and real-time morphological image analysis integrated into cell production.

Single-Cell Proliferation Microfluidic Device for High Throughput Investigation of Replicative Potential and Drug Resistance of Cancer Cells

Introduction

Cell proliferation represents a major hallmark of cancer biology, and manifests itself in the assessment of tumor growth, drug resistance and metastasis. Tracking cell proliferation or cell fate at the single-cell level can reveal phenotypic heterogeneity. However, characterization of cell proliferation is typically done in bulk assays which does not inform on cells that can proliferate under given environmental perturbations. Thus, there is a need for single-cell approaches that allow longitudinal tracking of the fate of a large number of individual cells to reveal diverse phenotypes.

Methods

We fabricated a new microfluidic architecture for high efficiency capture of single tumor cells, with the capacity to monitor cell divisions across multiple daughter cells. This single-cell proliferation (SCP) device enabled the quantification of the fate of more than 1000 individual cancer cells longitudinally, allowing comprehensive profiling of the phenotypic heterogeneity that would be otherwise masked in standard cell proliferation assays. We characterized the efficiency of single cell capture and demonstrated the utility of the SCP device by exposing MCF-7 breast tumor cells to different doses of the chemotherapeutic agent doxorubicin.

Results

The single cell trapping efficiency of the SCP device was found to be ~ 85%. At the low doses of doxorubicin (0.01 µM, 0.001 µM, 0.0001 µM), we observed that 50-80% of the drug-treated cells had undergone proliferation, and less than 10% of the cells do not proliferate. Additionally, we demonstrated the potential of the SCP device in circulating tumor cell applications where minimizing target cell loss is critical. We showed selective capture of breast tumor cells from a binary mixture of cells (tumor cells and white blood cells) that was isolated from blood processing. We successfully characterized the proliferation statistics of these captured cells despite their extremely low counts in the original binary suspension.

Conclusions

The SCP device has significant potential for cancer research with the ability to quantify proliferation statistics of individual tumor cells, opening new avenues of investigation ranging from evaluating drug resistance of anti-cancer compounds to monitoring the replicative potential of patient-derived cells.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12195-023-00773-z.

Blur-specific image quality assessment of microscopic hyperspectral images

Hyperspectral (HS) imaging (HSI) expands the number of channels captured within the electromagnetic spectrum with respect to regular imaging. Thus, microscopic HSI can improve cancer diagnosis by automatic classification of cells. However, homogeneous focus is difficult to achieve in such images, being the aim of this work to automatically quantify their focus for further image correction. A HS image database for focus assessment was captured. Subjective scores of image focus were obtained from 24 subjects and then correlated to state-of-the-art methods. Maximum Local Variation, Fast Image Sharpness block-based Method and Local Phase Coherence algorithms provided the best correlation results. With respect to execution time, LPC was the fastest.

Deep learning assisted holography microscopy for in-flow enumeration of tumor cells in blood

Currently, detection of circulating tumor cells (CTCs) in cancer patient blood samples relies on immunostaining, which does not provide access to live CTCs, limiting the breadth of CTC-based applications. Here, we take the first steps to address this limitation, by demonstrating staining-free enumeration of tumor cells spiked into lysed blood samples using digital holographic microscopy (DHM), microfluidics and machine learning (ML). A 3D-printed module for laser assembly was developed to simplify the optical set up for holographic imaging of cells flowing through a sheath-based microfluidic device. Computational reconstruction of the holograms was performed to localize the cells in 3D and obtain the plane of best focus images to train deep learning models. We developed a custom-designed light-weight shallow Network dubbed s-Net and compared its performance against off-the-shelf CNN models including ResNet-50. The accuracy, sensitivity and specificity of the s-Net model was found to be higher than the off-the-shelf ML models. By applying an optimized decision threshold to mixed samples prepared in silico, the false positive rate was reduced from 1 × 10^-2 to 2.77 × 10^-4. Finally, the developed DHM-ML framework was successfully applied to enumerate spiked MCF-7 breast cancer cells and SkOV3 ovarian cancer cells from lysed blood samples containing white blood cells (WBCs) at concentrations typical of label-free enrichment techniques. We conclude by discussing the advances that need to be made to translate the DHM-ML approach to staining-free enumeration of actual CTCs in cancer patient blood samples.

Synthesis and In Vitro Testing of YVO4:Eu3+@silica-NH-GDA-IgG Bio-Nano Complexes for Labelling MCF-7 Breast Cancer Cells

We present a visual tool and facile method to detect MCF-7 breast cancer cells by using YVO₄:Eu³⁺@silica-NH-GDA-IgG bio-nanocomplexes. To obtain these complexes, YVO₄:Eu³⁺ nanoparticles with a uniform size of 10-25 nm have been prepared firstly by the hydrothermal process, followed by surface functionalization to be bio-compatible and conjugated with cancer cells. The YVO₄:Eu³⁺@silica-NH-GDA-IgG nanoparticles exhibited an enhanced red emission at 618 nm under an excitation wavelength of 355 nm and were strongly coupled with MCF-7 breast cancer cells via biological conjugation. These bio-nanocomplexes showed a superior sensitiveness for MCF-7 cancer cell labelling with a detection percentage as high as 82%, while no HEK-293A healthy cells were probed under the same conditions of in vitro experiments. In addition, the detection percentage of MCF-7 breast cancer cells increased significantly via the functionalization and conjugation of YVO₄:Eu³⁺ nanoparticles. The experimental results demonstrated that the YVO₄:Eu³⁺@silica-NH-GDA-IgG bio-nanocomplexes can be used as a promising labelling agent for biomedical imaging and diagnostics.

Inertial focusing of particles and cells in the microfluidic labyrinth device: Role of sharp turns

Inertial, size-based focusing was investigated in the microfluidic labyrinth device consisting of several U-shaped turns along with circular loops. Turns are associated with tight curvature and, therefore, induce strong Dean forces for separating particles; however, systematic studies exploring this possibility do not exist. We characterized the focusing dynamics of different-sized rigid particles, cancer cells, and white blood cells over a range of fluid Reynolds numbers R e f . Streak widths of the focused particle streams at all the turns showed intermittent fluctuations that were substantial for smaller particles and at higher R e f . In contrast, cell streaks were less prone to fluctuations. Computational fluid dynamics simulations revealed the existence of strong turn-induced Dean vortices, which help explain the intermittent fluctuations seen in particle focusing. Next, we developed a measure of pairwise separability to evaluate the quality of separation between focused streams of two different particle sizes. Using this, we assessed the impact of a single sharp turn on separation. In general, the separability was found to vary significantly as particles traversed the tight-curvature U-turn. Comparing the separability at the entry and exit sections, we found that turns either improved or reduced separation between different-sized particles depending on R e f . Finally, we evaluated the separability at the downstream expansion section to quantify the performance of the labyrinth device in terms of achieving size-based enrichment of particles and cells. Overall, our results show that turns are better for cell focusing and separation given that they are more immune to curvature-driven fluctuations in comparison to rigid particles.