Digital Image Analysis in Clinical and Experimental Pathology: An Ode to Microscopy

Introduction

Clinical pathologist is a physician who specializes in the diagnosis and management of human diseases using appropriate laboratory methods. Conventional analysis of surgical or core biopsy specimens (histopathology) and body fluids, needle aspiration and exfoliative materials (cytopathology) using a light microscope is rapidly shifting towards digital integration. Digital imaging plays an increasing role in clinical diagnosis, biomedical research, seminar and continuing medical education. Currently, pathology platforms are composed of clinical, molecular, and translational pathologists and engineers with the sole intention of investigating cellular and molecular basis of human health through applied research in disease aetiology, pathogenesis, diagnosis and treatment. Molecular diagnosis using technical advances and the application of specific biomarkers in clinical practice are the two main pillars of modern personalized medicine especially in oncology. There are two levels of biomarker analysis in clinical trials; specific biomarker analysis to determine the patient’s stratification and general biomarker analysis to predict patient response to treatment (Salto-Tellez et al., 2014).

Clinical and molecular pathology platforms and translational research laboratories use the same tissue and fluid samples, phenotypic analysis, digital imaging and advanced technologies, under strict standards, in both clinical diagnosis and biomedical research. They play a central role in identifying mechanisms of carcinogenesis, gene expression, signalling aberrations, tumour microenvironment, cancer stem cells, tumour invasion and metastasis, metabolism, extracellular matrix regulation, angiogenesis, autophagy, ribonucleic acid (RNA) and protein expression, deoxyribonucleic acid (DNA) damage, non-coding RNA (ncRNA) deregulation and apoptosis (Meseure et al., 2014; Drak Alsibai & Meseure 2018).

The precision medicine and new scientific research approach of disease generate enormous amounts of information “big data”, that require database construction, analysis, storing and managing their storage (i.e. biobanking and bioinformatics). As technique complexity and data increase, so too do the amount and the complexity of data analysis. The complexity of data highlights the importance of integrating an alternative multi-parametric data assays into the clinical practice and scientific research (Gondhalekar et al., 2018).

It has become evident that accredited clinical and molecular pathology laboratories can make the most of specific biomarker analyses as well as incorporating other key aspects of translational research and data analysis. Currently, these laboratories are well equipped by tissue-based techniques {i.e. tissue microarray (TMA), immunohistochemistry (IHC), fluorescence in situ hybridization (FISH), multiplexed hybridisations}, digital image techniques, nucleic acid extractions and polymerase chain reaction (PCR) technologies {i.e. Sanger sequencing, pyrosequencing, quantitative PCR}, and high-throughput technologies {i.e. next-generation sequencing (NGS), high-throughput gene expression arrays, high-throughput methylation, gene copy number variation, high-throughput proteomics and metabolomics, imaging mass spectrometry (IMS) and matrix-assisted laser desorption/ionization-IMS (MALDI-IMS)} (Salto-Tellez et al., 2015).

This chapter will briefly describe the tissue processing and tissue microarrays protocols, and the main steps for acquiring digital images from tissue samples, and the imaging techniques relevant for this purpose such as whole-slide imaging. Then, the chapter will highlight the major high-throughput vectorial and non-vectorial technologies in molecular pathology, and will empathize on their mechanistic details and main applications in diagnostic pathology, oncology and biomedical research (Table 1).