In the current practice of pathology, the identification of cell markers and their respective distribution represents an indispensable dialogue for diagnostic, predictive, therapeutic, and research purposes. Early immunohistochemical protocols were limited to direct, fluorescent labeled antibodies, yielding quick results but lacking sensitivity. More recently, the use of indirect techniques –utilization of enzyme labels–and various detection systems have continued to advance the complexity of IHC, increasing both its specificity and sensitivity of detecting one or multiple antigen(s) (Ag) simultaneously. As such, IHC has become an affordable, powerful, and readily available means for the identification of candidate biomarkers (mostly lineage markers) in formalin-fixed, paraffin-embedded (FFPE) tissue samples. Pathologists are now asked to “quantify” expression levels of differential prognostic markers–at microscopic level–using this arguably “non-quantitative” technique. Conventionally, histological grading relies mainly on manual counting of positively immunostained cells, a labour intensive protocol that may be associated with subjectivity, intra- and inter- observer variation and reproducibility issues. The subjectivity and lack of reproducibility has prompted the use of computer-assisted or fully automated image analysis technologies. Digital image acquisition systems are becoming commonplace and as such, the demand for complex assessments of digital images of histological slides must be matched with quantitative platforms. In this study, we aim to introduce a computer-assisted image-computing platform that is both accurate and efficient in quantification of isolated and heterogeneous candidate biomarkers in glioblastoma.