Intelligent Platelet Morphometry

Revolutionizing microfluidics with artificial intelligence: a new dawn for lab-on-a-chip technologies

Keisuke Goda, Hang Lu, Peng Fei, and Jochen Guck introduce the AI in Microfluidics themed collection, on revolutionizing microfluidics with artificial intelligence: a new dawn for lab-on-a-chip technologies.

Real-time intelligent classification of COVID-19 and thrombosis via massive image-based analysis of platelet aggregates

Microvascular thrombosis is a typical symptom of COVID-19 and shows similarities to thrombosis. Using a microfluidic imaging flow cytometer, we measured the blood of 181 COVID-19 samples and 101 non-COVID-19 thrombosis samples, resulting in a total of 6.3 million bright-field images. We trained a convolutional neural network to distinguish single platelets, platelet aggregates, and white blood cells and performed classical image analysis for each subpopulation individually. Based on derived single-cell features for each population, we trained machine learning models for classification between COVID-19 and non-COVID-19 thrombosis, resulting in a patient testing accuracy of 75%. This result indicates that platelet formation differs between COVID-19 and non-COVID-19 thrombosis. All analysis steps were optimized for efficiency and implemented in an easy-to-use plugin for the image viewer napari, allowing the entire analysis to be performed within seconds on mid-range computers, which could be used for real-time diagnosis.

Studying the efficacy of antiplatelet drugs on atherosclerosis by optofluidic imaging on a chip

Vascular stenosis caused by atherosclerosis instigates activation and aggregation of platelets, eventually resulting in thrombus formation. Although antiplatelet drugs are commonly used to inhibit platelet activation and aggregation, they unfortunately cannot prevent recurrent thrombotic events in patients with atherosclerosis. This is partially due to the limited understanding of the efficacy of antiplatelet drugs in the complex hemodynamic environment of vascular stenosis. Conventional methods for evaluating the efficacy of antiplatelet drugs under stenosis either fail to simulate the hemodynamic environment of vascular stenosis characterized by high shear stress and recirculatory flow or lack spatial resolution in their analytical techniques to statistically identify and characterize platelet aggregates. Here we propose and experimentally demonstrate a method comprising an in vitro 3D stenosis microfluidic chip and an optical time-stretch quantitative phase imaging system for studying the efficacy of antiplatelet drugs under stenosis. Our method simulates the atherogenic flow environment of vascular stenosis while enabling high-resolution and statistical analysis of platelet aggregates. Using our method, we distinguished the efficacy of three antiplatelet drugs, acetylsalicylic acid (ASA), cangrelor, and eptifibatide, for inhibiting platelet aggregation induced by stenosis. Specifically, ASA failed to inhibit stenosis-induced platelet aggregation, while eptifibatide and cangrelor showed high and moderate efficacy, respectively. Furthermore, we demonstrated that the drugs tested also differed in their efficacy for inhibiting platelet aggregation synergistically induced by stenosis and agonists (e.g., adenosine diphosphate, and collagen). Taken together, our method is an effective tool for investigating the efficacy of antiplatelet drugs under vascular stenosis, which could assist the development of optimal pharmacologic strategies for patients with atherosclerosis.

Intelligent sort-timing prediction for image-activated cell sorting

Intelligent image-activated cell sorting (iIACS) has enabled high-throughput image-based sorting of single cells with artificial intelligence (AI) algorithms. This AI-on-a-chip technology combines fluorescence microscopy, AI-based image processing, sort-timing prediction, and cell sorting. Sort-timing prediction is particularly essential due to the latency on the order of milliseconds between image acquisition and sort actuation, during which image processing is performed. The long latency amplifies the effects of the fluctuations in the flow speed of cells, leading to fluctuation and uncertainty in the arrival time of cells at the sort point on the microfluidic chip. To compensate for this fluctuation, iIACS measures the flow speed of each cell upstream, predicts the arrival timing of the cell at the sort point, and activates the actuation of the cell sorter appropriately. Here, we propose and demonstrate a machine learning technique to increase the accuracy of the sort-timing prediction that would allow for the improvement of sort event rate, yield, and purity. Specifically, we trained an algorithm to predict the sort timing for morphologically heterogeneous budding yeast cells. The algorithm we developed used cell morphology, position, and flow speed as inputs for prediction and achieved 41.5% lower prediction error compared to the previously employed method based solely on flow speed. As a result, our technique would allow for an increase in the sort event rate of iIACS by a factor of ~2.

Targeting biophysical cues to address platelet storage lesions

Platelets play vital roles in vascular repair, especially in primary hemostasis, and have been widely used in transfusion to prevent bleeding or manage active bleeding. Recently, platelets have been used in tissue repair (e.g., bone, skin, and dental alveolar tissue) and cell engineering as drug delivery carriers. However, the biomedical applications of platelets have been associated with platelet storage lesions (PSLs), resulting in poor clinical outcomes with reduced recovery, survival, and hemostatic function after transfusion. Accumulating evidence has shown that biophysical cues play important roles in platelet lesions, such as granule secretion caused by shear stress, adhesion affected by substrate stiffness, and apoptosis caused by low temperature. This review summarizes four major biophysical cues (i.e., shear stress, substrate stiffness, hydrostatic pressure, and thermal microenvironment) involved in the platelet preparation and storage processes, and discusses how they may synergistically induce PSLs such as platelet shape change, activation, apoptosis and clearance. We also review emerging methods for studying these biophysical cues in vitro and existing strategies targeting biophysical cues for mitigating PSLs. We conclude with a perspective on the future direction of biophysics-based strategies for inhibiting PSLs. STATEMENT OF SIGNIFICANCE: Platelet storage lesions (PSLs) involve a series of structural and functional changes. It has long been accepted that PSLs are initiated by biochemical cues. Our manuscript is the first to propose four major biophysical cues (shear stress, substrate stiffness, hydrostatic pressure, and thermal microenvironment) that platelets experience in each operation step during platelet preparation and storage processes in vitro, which may synergistically contribute to PSLs. We first clarify these biophysical cues and how they induce PSLs. Strategies targeting each biophysical cue to improve PSLs are also summarized. Our review is designed to draw the attention from a broad range of audience, including clinical doctors, biologists, physical scientists, engineers and materials scientists, and immunologist, who study on platelets physiology and pathology.

Fast, streamlined fluorescence nanoscopy resolves rearrangements of SNARE and cargo proteins in platelets co-incubated with cancer cells

BACKGROUND

Increasing evidence suggests that platelets play a central role in cancer progression, with altered storage and selective release from platelets of specific tumor-promoting proteins as a major mechanism. Fluorescence-based super-resolution microscopy (SRM) can resolve nanoscale spatial distribution patterns of such proteins, and how they are altered in platelets upon different activations. Analysing such alterations by SRM thus represents a promising, minimally invasive strategy for platelet-based diagnosis and monitoring of cancer progression. However, broader applicability beyond specialized research labs will require objective, more automated imaging procedures. Moreover, for statistically significant analyses many SRM platelet images are needed, of several different platelet proteins. Such proteins, showing alterations in their distributions upon cancer progression additionally need to be identified.

RESULTS

A fast, streamlined and objective procedure for SRM platelet image acquisition, analysis and classification was developed to overcome these limitations. By stimulated emission depletion SRM we imaged nanoscale patterns of six different platelet proteins; four different SNAREs (soluble N-ethylmaleimide factor attachment protein receptors) mediating protein secretion by membrane fusion of storage granules, and two angiogenesis regulating proteins, representing cargo proteins within these granules coupled to tumor progression. By a streamlined procedure, we recorded about 100 SRM images of platelets, for each of these six proteins, and for five different categories of platelets; incubated with cancer cells (MCF-7, MDA-MB-231, EFO-21), non-cancer cells (MCF-10A), or no cells at all. From these images, structural similarity and protein cluster parameters were determined, and probability functions of these parameters were generated for the different platelet categories. By comparing these probability functions between the categories, we could identify nanoscale alterations in the protein distributions, allowing us to classify the platelets into their correct categories, if they were co-incubated with cancer cells, non-cancer cells, or no cells at all.

CONCLUSIONS

The fast, streamlined and objective acquisition and analysis procedure established in this work confirms the role of SNAREs and angiogenesis-regulating proteins in platelet-mediated cancer progression, provides additional fundamental knowledge on the interplay between tumor cells and platelets, and represent an important step towards using tumor-platelet interactions and redistribution of nanoscale protein patterns in platelets as a basis for cancer diagnostics.

Deep imaging flow cytometry

Imaging flow cytometry (IFC) has become a powerful tool for diverse biomedical applications by virtue of its ability to image single cells in a high-throughput manner. However, there remains a challenge posed by the fundamental trade-off between throughput, sensitivity, and spatial resolution. Here we present deep-learning-enhanced imaging flow cytometry (dIFC) that circumvents this trade-off by implementing an image restoration algorithm on a virtual-freezing fluorescence imaging (VIFFI) flow cytometry platform, enabling higher throughput without sacrificing sensitivity and spatial resolution. A key component of dIFC is a high-resolution (HR) image generator that synthesizes "virtual" HR images from the corresponding low-resolution (LR) images acquired with a low-magnification lens (10×/0.4-NA). For IFC, a low-magnification lens is favorable because of reduced image blur of cells flowing at a higher speed, which allows higher throughput. We trained and developed the HR image generator with an architecture containing two generative adversarial networks (GANs). Furthermore, we developed dIFC as a method by combining the trained generator and IFC. We characterized dIFC using Chlamydomonas reinhardtii cell images, fluorescence in situ hybridization (FISH) images of Jurkat cells, and Saccharomyces cerevisiae (budding yeast) cell images, showing high similarities of dIFC images to images obtained with a high-magnification lens (40×/0.95-NA), at a high flow speed of 2 m s^-1. We lastly employed dIFC to show enhancements in the accuracy of FISH-spot counting and neck-width measurement of budding yeast cells. These results pave the way for statistical analysis of cells with high-dimensional spatial information.

Three-dimensional analysis of blood platelet spreading using digital holographic microscopy: a statistical study of the differential effect of coatings in healthy volunteers and dialyzed patients

In cardiovascular disorders, the study of thrombocytes, commonly known as platelets, is highly important since they are involved in blood clotting, essential in hemostasis, and they can in pathological situations affect the blood circulation. In this paper, single deposited platelets are measured using interferometric digital holographic microscopy. We have shown that the average optical height of platelets is significantly lower in healthy volunteers than in dialyzed patients, meaning a better spreading. It demonstrates the great interest for assessing this parameter in any patients, and therefore the high potential of analyzing single spread platelets using digital holographic microscopy in fundamental research as well as a diagnostic tool in routine laboratories, for usual blood tests.

Massive image-based single-cell profiling reveals high levels of circulating platelet aggregates in patients with COVID-19

A characteristic clinical feature of COVID-19 is the frequent incidence of microvascular thrombosis. In fact, COVID-19 autopsy reports have shown widespread thrombotic microangiopathy characterized by extensive diffuse microthrombi within peripheral capillaries and arterioles in lungs, hearts, and other organs, resulting in multiorgan failure. However, the underlying process of COVID-19-associated microvascular thrombosis remains elusive due to the lack of tools to statistically examine platelet aggregation (i.e., the initiation of microthrombus formation) in detail. Here we report the landscape of circulating platelet aggregates in COVID-19 obtained by massive single-cell image-based profiling and temporal monitoring of the blood of COVID-19 patients (n = 110). Surprisingly, our analysis of the big image data shows the anomalous presence of excessive platelet aggregates in nearly 90% of all COVID-19 patients. Furthermore, results indicate strong links between the concentration of platelet aggregates and the severity, mortality, respiratory condition, and vascular endothelial dysfunction level of COVID-19 patients.

Droplet flow cytometry for single-cell analysis

The interrogation of single cells has revolutionised biology and medicine by providing crucial unparalleled insights into cell-to-cell heterogeneity. Flow cytometry (including fluorescence-activated cell sorting) is one of the most versatile and high-throughput approaches for single-cell analysis by detecting multiple fluorescence parameters of individual cells in aqueous suspension as they flow past through a focus of excitation lasers. However, this approach relies on the expression of cell surface and intracellular biomarkers, which inevitably lacks spatial and temporal phenotypes and activities of cells, such as secreted proteins, extracellular metabolite production, and proliferation. Droplet microfluidics has recently emerged as a powerful tool for the encapsulation and manipulation of thousands to millions of individual cells within pico-litre microdroplets. Integrating flow cytometry with microdroplet architectures surrounded by aqueous solutions (e.g., water-in-oil-in-water (W/O/W) double emulsion and hydrogel droplets) opens avenues for new cellular assays linking cell phenotypes to genotypes at the single-cell level. In this review, we discuss the capabilities and applications of droplet flow cytometry (DFC). This unique technique uses standard commercially available flow cytometry instruments to characterise or select individual microdroplets containing single cells of interest. We explore current challenges associated with DFC and present our visions for future development.