Advances in Imaging Modalities, Artificial Intelligence, and Single Cell Biomarker Analysis, and Their Applications in Cytopathology

Abstract

Several advances in recent decades in digital imaging, artificial intelligence, and multiplex modalities have improved our ability to automatically analyze and interpret imaging data. Imaging technologies such as optical coherence tomography, optical projection tomography, and quantitative phase microscopy allow analysis of tissues and cells in 3-dimensions and with subcellular granularity. Improvements in computer vision and machine learning have made algorithms more successful in automatically identifying important features to diagnose disease. Many new automated multiplex modalities such as antibody barcoding with cleavable DNA (ABCD), single cell analysis for tumor phenotyping (SCANT), fast analytical screening technique fine needle aspiration (FAST-FNA), and portable fluorescence-based image cytometry analyzer (CytoPAN) are under investigation. These have shown great promise in their ability to automatically analyze several biomarkers concurrently with high sensitivity, even in paucicellular samples, lending themselves well as tools in FNA. Not yet widely adopted for clinical use, many have successfully been applied to human samples. Once clinically validated, some of these technologies are poised to change the routine practice of cytopathology.

Keywords: computational cytopathology; computational pathology; molecular cytopathology; multiplex immunofluorescence; single cell biomarker analysis.

Figure 1

ABCD (antibody barcoding with cleavable DNA). FNA biopsy is performed to isolate cells of interest, which may undergo a purification step. Cells are incubated with DNA-barcoded antibodies to proteins of interest. DNA barcodes are cleaved from antibodies. (A) Released DNA-barcodes can undergo PCR amplification and gel electrophoresis for semi-quantitative measurement of protein expression; (B) Released DNA-barcodes can undergo multiplex quantitative PCR; (C) Released DNA-barcodes can be hybridized with fluorescent labeled capture probes and automatically imaged and analyzed with nanostring technology.

Figure 2

SCANT (single cell analysis for tumor phenotyping). Cells are obtained via fine needle aspiration biopsy and incubated with DNA-antibody conjugates. DNA strands are hybridized to complementary strands with two flurochromes; antibodies with different flurochromes will fluoresce at different channels. Cells are imaged and subjected to automated image analysis. Fluorescent strands can be washed off and capped between cycles to reduce cycle-to-cycle background.

Figure 3

FAST-FNA (fast analytical screening technique fine needle aspiration). FNA biopsy samples are fixed and stained with fluorescent-labeled antibodies. Antibodies are conjugated to trans-cyclooctene (TCO), conjugated to tetrazine (Tz), conjugated to a fluorescent label. Images are processed in an automated image cytometer; deep learning algorithms quantify marker expression. Tz-activated black hole quenchers quench the fluorescent signal in mere seconds, allowing much shorter time intervals between cycles.

Figure 4

CYTO-PAN (portable fluorescence-based image cytometry analyser). This specific system was used to diagnose and subtype breast cancer in FNA specimens. FNA biopsy samples are processed with prefabricated kits with preselected antibodies. Slides with stained cells are analyzed by the CytoPAN device, a fluorescence based image cytometer with five optical channels. Custom-developed algorithms identify cancer cells and biomarker information, producing quantitative reports. The entire process could be performed within 1 h.