FAST-PATH · Fast-Tracking Pathology via Automated Image Analysis and High-Performance Computing: Application to Prostate Cancer Diagnostics

Histopathology has traditionally been a low-throughput, labour-intensive technique. Conventional manual annotation of tissue slides requires a pathologist to examine the tissue and cellular components to grade the level of disease progression. However, digital pathology and automated digital image analysis solutions can provide a more rapid solution. In recent years, there has been a rapid uptake of digital pathology in both the academic and industrial sectors, each creating large image libraries with related manual and automated annotations, all stored on local systems. There is now a requirement to streamline this process and to provide a comprehensive set of tools for data integration and mining of this valuable, but often not fully exploited, information. There is also a need for inter-lab standardisation and the development of high-performance computing approaches to enable integration of multiple libraries and knowledge discovery. The FAST-PATH project will address these key issues to maximise the capabilities of the digital pathology workflow. Specifically, FAST-PATH will develop a set of tools, available through a web-based interface, to address the standardisation, integration and knowledge discovery of high-throughput digital pathology libraries and related manual annotations. We will focus on integration of oncology-based data generated from the academic and industrial partners involved, with a particular emphasis on prostate cancer. This inter-sectoral study will also involve validation of novel prognostic biomarkers via automated analysis of immunohistochemical images. Importantly, FAST-PATH will bring together 4 major European academic institutes and 2 SME partners with the central aim of streamlining digital pathology via the development of data standardisation, integration and discovery tools.