ABSTRACT
Tissue phenomics is a key component for the successful transition of digital pathology into the next-generation diagnostic pathology. The intelligent combination of integrated workflows with effective user and application interfaces, cognition networks, and hypothesis-free machine learning algorithms based on the conventional expert pathology wisdom will generate new knowledge and diagnostic innovation in the routine diagnostic pathology and tissue biomarker discovery. The major goal of automated image and data analysis in digital pathology is the development of algorithms and methods for clinical applications and advanced diagnostics. The classical histopathological diagnosis is the necessary prerequisite for any reliable treatment of diseases such as cancer and is, by far, the least expensive diagnostic procedure. Computer-aided diagnosis can add or increase accuracy of predictive diagnosis and patient management decisions regarding therapy and outcome evaluation. The future development of decision support systems comprises algorithms and methods for computer-aided diagnosis, quality assurance, image viewing, inspection and context navigation, co-registration, virtual multiplexing, and data analysis.
