Digital Image Analysis in Pathology (DIAPath)

(Code: ULB723)

DIAPath is a multidisciplinary and inter-faculty unit (Faculty of Medicine and Brussels School of Engineering/ École polytechnique de Bruxelles) included
in the Center for Microscopy and Molecular Imaging (CMMI, Biopark of Gosselies). DIAPath develops an integrated approach for the characterization,
validation and monitoring of tissue-based biomarkers in animal and human tissue samples (in close collaboration avec the Pathology Department of the Erasme
hospital). The methodology involves histological techniques, immunohistochemistry (IHC) and chromogenic in situ hybridization (CISH) as well as image analysis
(involving Artificial Intelligence), biostatistics and data mining. IHC and CISH have the advantage to preserve tissue morphology and thus allows target
location at histological and cell levels. By simultaneously processing thousands of samples, the tissue microarray (TMA) technology allows standardized
screening of protein expression using IHC and thus provides a very efficient way for biomarker validation in preclinical and clinical studies. The ''cell-block''
technique allows the study of cell lines with the same methodology. The unit has also a slide scanner for complete slide imaging and virtual slide making.
This enables objective and quantitative staining characterization using image analysis. Specific techniques are developed to efficiently process and analyze
the big slide images (> 4 Giga pixels), in collaboration with the LISA (Laboratory of Image Synthesis and Analysis, EPB). These techniques aim to describe
staining heterogeneity by means of original features as well as to characterize colocalization of staining carried out on serial slides. Databases are
then set up on large sample series by merging biological and clinical and/or pharmaceutical data, depending of the study context. These data are then
submitted to modern methods of multivariate data analysis. The general aim is to extract information useful in understanding pathological processes and therapy
responses, as well as to identify and to validate new biomarkers beneficial for diagnostic, prognostic and therapeutic procedures.

×

Frascati domains:

 • 1.2 Computer and information sciences

 • 2.6 Medical engineering

 • 3.2 Clinical medicine

 • 3.4 Health biotechnology

Disciplines CRef :

 • Anatomopathology

 • Artificial intelligence

 • Biomedical engineering

 • Biomedical sciences

 • Cancerology

 • Engineering sciences

 • Histology

 • Histopathology

 • Imaging and image processing

 • Medical imaging, radiology, tomography