A computer vision approach to rare cell in vivo fluorescence flow cytometry

Effect of pulsed laser parameters on photoacoustic flow cytometry efficiency in vitro and in vivo

Photoacoustic flow cytometry is one of the most effective approaches to detect "alien" objects in the bloodstream, including circulating tumor cells, blood clots, parasites, and emboli. However, the possibility of detecting high-amplitude signals from these objects against the background of blood depends on the parameters of the laser pulse. So, the dependencies of photoacoustic signals amplitude and number on laser pulse energy (5-150 μJ), pulse length (1, 2, 5 ns), and pulse repetition rate (2, 5, 10 kHz) for the melanoma cells were investigated. First, the PA responses of a melanoma cell suspension in vitro were measured to directly assess the efficiency of converting laser light into an acoustic signal. After it, the same dependence with the developed murine model based on constant rate melanoma cell injection into the animal blood flow was tested. Both in vivo and in vitro experiments show that signal generation efficiency increases with laser pulse energy above 15 μJ. Shorter pulses, especially 1 ns, provide more efficient signal generation as well as higher pulse rates. A higher pulse rate also provides more efficient signal generation, but also leads to overheating of the skin. The results show the limits where the photoacoustic flow cytometry system can be effectively used for the detection of circulating tumor cells in undiluted blood both for in vitro experiments and for in vivo murine models.

Advanced flow cytometry for biomedical applications

Flow cytometry (FC) is a versatile tool with excellent capabilities to detect and measure multiple characteristics of a population of cells or particles. Notable advancements in in vivo photoacoustic FC, coherent Raman FC, microfluidic FC, and so on, have been achieved in the last two decades, which endows FC with new functions and expands its applications in basic research and clinical practice. Advanced FC broadens the tools available to researchers to conduct research involving cancer detection, microbiology (COVID-19, HIV, bacteria, etc.), and nucleic acid analysis. This review presents an overall picture of advanced flow cytometers and provides not only a clear understanding of their mechanisms but also new insights into their practical applications. We identify the latest trends in this area and aim to raise awareness of advanced techniques of FC. We hope this review expands the applications of FC and accelerates its clinical translation.

In Vivo Flow Cytometry

In vivo flow cytometry (IVFC) was first designed to detect circulating cells in a mouse ear. It allows real-time monitoring of cells in peripheral blood with no need to draw blood. The IVFC field has made great progress during the last decade with the development of fluorescence, photoacoustic, and multiphoton microscopy. Moreover, the application of IVFC is no longer restricted to circulating cells. IVFC based on fluorescence and photoacoustic are most widely applied in biomedical research. Methods based on fluorescence are often used for object monitoring in superficial vessels, while methods based on photoacoustics have an advantage of label-free monitoring in deep vessels. In this chapter, we introduce technical points and key applications of IVFC. We focus on the principles, labeling strategies, sensitivity, and biomedical applications of the technology. In addition, we summarize this chapter and discuss important research directions of IVFC in the future.

Image Analysis Using Machine Learning for Automated Detection of Hemoglobin H Inclusions in Blood Smears - A Method for Morphologic Detection of Rare Cells

Background

Morphologic rare cell detection is a laborious, operator-dependent process which has the potential to be improved by the use of image analysis using artificial intelligence. Detection of rare hemoglobin H (HbH) inclusions in red cells in the peripheral blood is a common screening method for alpha-thalassemia. This study aims to develop a convolutional neural network-based algorithm for the detection of HbH inclusions.

Methods

Digital images of HbH-positive and HbH-negative blood smears were used to train and test the software. The software performance was tested on images obtained at various magnifications and on different scanning platforms. Another model was developed for total red cell counting and was used to confirm HbH cell frequency in alpha-thalassemia trait. The threshold minimum red cells to image for analysis was determined by Poisson modeling and validated on image sets.

Results

The sensitivity and specificity of the software for HbH+ cells on images obtained at ×100, ×60, and ×40 objectives were close to 91% and 99%, respectively. When an AI-aided diagnostic model was tested on a pilot of 40 whole slide images (WSIs), good inter-rater reliability and high sensitivity and specificity of slide-level classification were obtained. Using the lowest frequency of HbH+ cells (1 in 100,000) observed in our study, we estimated that a minimum of 2.4 × 10⁶ red cells would need to be analyzed to reduce misclassification at the slide level. The minimum required smear size was validated on 78 image sets which confirmed its validity.

Conclusions

WSI image analysis can be utilized effectively for morphologic rare cell detection. The software can be further developed on WISs and evaluated in future clinical validation studies comparing AI-aided diagnosis with the routine diagnostic method.

Biomedical Applications of Tissue Clearing and Three-Dimensional Imaging in Health and Disease

Three-dimensional (3D) optical imaging techniques can expand our knowledge about physiological and pathological processes that cannot be fully understood with 2D approaches. Standard diagnostic tests frequently are not sufficient to unequivocally determine the presence of a pathological condition. Whole-organ optical imaging requires tissue transparency, which can be achieved by using tissue clearing procedures enabling deeper image acquisition and therefore making possible the analysis of large-scale biological tissue samples. Here, we review currently available clearing agents, methods, and their application in imaging of physiological or pathological conditions in different animal and human organs. We also compare different optical tissue clearing methods discussing their advantages and disadvantages and review the use of different 3D imaging techniques for the visualization and image acquisition of cleared tissues. The use of optical tissue clearing resources for large-scale biological tissues 3D imaging paves the way for future applications in translational and clinical research.

Detection of Rare Objects by Flow Cytometry: Imaging, Cell Sorting, and Deep Learning Approaches

Flow cytometry nowadays is among the main working instruments in modern biology paving the way for clinics to provide early, quick, and reliable diagnostics of many blood-related diseases. The major problem for clinical applications is the detection of rare pathogenic objects in patient blood. These objects can be circulating tumor cells, very rare during the early stages of cancer development, various microorganisms and parasites in the blood during acute blood infections. All of these rare diagnostic objects can be detected and identified very rapidly to save a patient's life. This review outlines the main techniques of visualization of rare objects in the blood flow, methods for extraction of such objects from the blood flow for further investigations and new approaches to identify the objects automatically with the modern deep learning methods.

Flow-regulated endothelial glycocalyx determines metastatic cancer cell activity

Cancer metastasis and secondary tumor initiation largely depend on circulating tumor cell (CTC) and vascular endothelial cell (EC) interactions by incompletely understood mechanisms. Endothelial glycocalyx (GCX) dysfunction may play a significant role in this process. GCX structure depends on vascular flow patterns, which are irregular in tumor environments. This work presents evidence that disturbed flow (DF) induces GCX degradation, leading to CTC homing to the endothelium, a first step in secondary tumor formation. A 2-fold greater attachment of CTCs to human ECs was found to occur under DF conditions, compared to uniform flow (UF) conditions. These results corresponded to an approximately 50% decrease in wheat germ agglutinin (WGA)-labeled components of the GCX under DF conditions, vs UF conditions, with undifferentiated levels of CTC-recruiting E-selectin under DF vs UF conditions. Confirming the role of the GCX, neuraminidase induced the degradation of WGA-labeled GCX under UF cell culture conditions or in Balb/C mice and led to an over 2-fold increase in CTC attachment to ECs or Balb/C mouse lungs, respectively, compared to untreated conditions. These experiments confirm that flow-induced GCX degradation can enable metastatic CTC arrest. This work, therefore, provides new insight into pathways of secondary tumor formation.

Phase contrast time-lapse microscopy datasets with automated and manual cell tracking annotations

Phase contrast time-lapse microscopy is a non-destructive technique that generates large volumes of image-based information to quantify the behaviour of individual cells or cell populations. To guide the development of algorithms for computer-aided cell tracking and analysis, 48 time-lapse image sequences, each spanning approximately 3.5 days, were generated with accompanying ground truths for C2C12 myoblast cells cultured under 4 different media conditions, including with fibroblast growth factor 2 (FGF2), bone morphogenetic protein 2 (BMP2), FGF2 + BMP2, and control (no growth factor). The ground truths generated contain information for tracking at least 3 parent cells and their descendants within these datasets and were validated using a two-tier system of manual curation. This comprehensive, validated dataset will be useful in advancing the development of computer-aided cell tracking algorithms and function as a benchmark, providing an invaluable opportunity to deepen our understanding of individual and population-based cell dynamics for biomedical research.

Background Modeling Method to Identify Interactions Between Circulating Tumor Cells and Dendritic Cells

Interactions between dendritic cells (DCs) and circulating tumor cells (CTCs) have attracted wide attention in tumor immunity research, especially on specifically targeted tumors. However, feature extraction and noninvasive tracking of DCs and CTCs are challenges that have long existed in biomedicine. In this study, we developed an automatic algorithm for identifying, counting, tracking, and segmenting fluorescently-labeled CTCs and DCs from the blood vessels of mouse ears. For fluorescence imaging, we constructed an in vivo image flow cytometry system to capture dual-channel (green and red) fluorescence image sequence simultaneously. To achieve real-time functions for the CTCs and DCs, we developed a motion detection method based on codebook which first performs background modeling and then cone-shaped area search procedures for postprocessing. We validated this novel algorithm through in vivo image sequencing, through which we observed the interaction of CTCs and DCs. Moreover, we used quantitative colocalization to determine the relationship between CTCs and DCs. The quantitative results illustrated the interactions between CTCs and DCs, as did image sequences which are promising for driving research on cancer immunotherapy in the future.

Fluorescence detection, enumeration and characterization of single circulating cells in vivo: technology, applications and future prospects

There are many diseases and biological processes that involve circulating cells in the bloodstream, such as cancer metastasis, immunology, reproductive medicine, and stem cell therapies. This has driven significant interest in new technologies for the study of circulating cells in small animal research models and clinically. Most currently used methods require drawing and enriching blood samples from the body, but these suffer from a number of limitations. In contrast, 'in vivo flow cytometry' (IVFC) refers to set of technologies that allow study of cells directly in the bloodstream of the organism in vivo. In recent years the IVFC field has grown significantly and new techniques have been developed, including fluorescence microscopy, multi-photon, photo-acoustic, and diffuse fluorescence IVFC. In this paper we review recent technical advances in IVFC, with emphasis on instrumentation, contrast mechanisms, and detection sensitivity. We also describe key applications in biomedical research, including cancer research and immunology. Last, we discuss future directions for IVFC, as well as prospects for broader adoption by the biomedical research community and translation to humans clinically.

Endogenous light scattering as an optical signature of circulating tumor cell clusters

Circulating tumor cell clusters (CTCCs) are significantly more likely to form metastases than single tumor cells. We demonstrate the potential of backscatter-based flow cytometry (BSFC) to detect unique light scattering signatures of CTCCs in the blood of mice orthotopically implanted with breast cancer cells and treated with an anti-ADAM8 or a control antibody. Based on scattering detected at 405, 488, and 633 nm from blood samples flowing through microfluidic devices, we identified 14 CTCCs with large scattering peak widths and intensities, whose presence correlated strongly with metastasis. These initial studies demonstrate the potential to detect CTCCs via label-free BSFC.

Toward lower contrast computer vision in vivo flow cytometry

There are many applications in biomedical research where detection and enumeration of circulating cells (CCs) is important. Existing techniques involve drawing and enriching blood samples and analyzing them ex vivo. More recently, small animal "in vivo flow cytometry" (IVFC) techniques have been developed, where fluorescently-labeled cells flowing through small arterioles (ear, retina) are detected and counted. We recently developed a new high-sensitivity IVFC technique termed "Computer Vision(CV)-IVFC". Here, large circulating blood volumes were monitored in the ears of mice with a wide-field video-rate near-infrared (NIR) fluorescent camera. Cells were labeled with a membrane dye and were detected and tracked in noisy image sequences. This technique allowed enumeration of CCs in vivo with overall sensitivity better than 10 cells/mL. However, an ongoing area of interest in our lab is optimization of the system for lower-contrast imaging conditions, e.g. when CCs are weakly labeled, or in the case higher background autofluorescence with visible dyes. To this end, we developed a new optical flow phantom model to control autofluorescence intensity and physical structure to better mimic conditions observed in mice. We acquired image sequences from a series of phantoms with varying levels of contrast and analyzed the distribution of pixel intensities, and showed that we could generate similar conditions to those in vivo. We characterized the performance of our CV-IVFC algorithm in these phantoms with respect to sensitivity and false-alarm rates. Use of this phantom model in optimization of the instrument and algorithm under lower-contrast conditions is the subject of ongoing work in our lab.

In vivo photoswitchable flow cytometry for direct tracking of single circulating tumor cells

Photoswitchable fluorescent proteins (PSFPs) that change their color in response to light have led to breakthroughs in studying static cells. However, using PSFPs to study cells in dynamic conditions is challenging. Here we introduce a method for in vivo ultrafast photoswitching of PSFPs that provides labeling and tracking of single circulating cells. Using in vivo multicolor flow cytometry, this method demonstrated the capability for studying recirculation, migration, and distribution of circulating tumor cells (CTCs) during metastasis progression. In tumor-bearing mice, it enabled monitoring of real-time dynamics of CTCs released from primary tumor, identifying dormant cells, and imaging of CTCs colonizing a primary tumor (self-seeding) or existing metastasis (reseeding). Integration of genetically encoded PSFPs, fast photoswitching, flow cytometry, and imaging makes in vivo single cell analysis in the circulation feasible to provide insights into the behavior of CTCs and potentially immune-related and bacterial cells in circulation.