Study: Human Tissue in Systems Medicine

Human tissue is inherently heterogeneous which, along with ischaemic and laboratory preparation artefacts, creates a headache for the systems pathologist. Classical histopathology is very adept at dealing with tissue artefacts and heterogeneity as the pathologist merely ignores, assesses or allows for them. Upon biochemical quantification of the artefactual and heterogeneous tissue the problem becomes apparent; imperfect tissue into the predictive model means imperfect results out. Automated image analysis of whole tissue sections can aid in quantifying heterogeneity as well as act as a QC checkpoint. Sophisticated image analysis packages, such as Definiens Developer™, allow the full quantification of heterogeneously expressed biomarkers as well as morphological and histopathological features across whole tissue sections, while negating certain tissue and staining artefacts. Automated image analysis can also aid in quantifying the per cent of tumour area within a tissue section and in identifying heterogeneous subpopulations which can be purified for use in genomic testing. Systems medicine relies on incorporating quantitative morphological, proteomic and genomic data to create a prognostic and predictive model which directly impacts within the clinic. It therefore needs to acknowledge and account for the imperfection of the tissue sample which is often sub-optimal and heterogeneous. In the short review linked with this blog we argue that a systems medicine approach to pathology will not seek to replace but rather augment classical histopathology and that automated image analysis is an extremely valuable tool for the systems pathologist.

http://www.ncbi.nlm.nih.gov/pubmed/24118991