Semantic Integrative Digital Pathology: Insights into Microsemiological Semantics and Image Analysis Scalability
Daniel Racoceanu,Frédérique Capron
DOI: https://doi.org/10.1159/000443964
Pathobiology
Abstract:Being able to provide a traceable and dynamic second opinion has become an ethical priority for patients and health care professionals in modern computer-aided medicine. In this perspective, a semantic cognitive virtual microscopy approach has been recently initiated, the MICO project, by focusing on cognitive digital pathology. This approach supports the elaboration of pathology-compliant daily protocols dedicated to breast cancer grading, in particular mitotic counts and nuclear atypia. A proof of concept has thus been elaborated, and an extension of these approaches is now underway in a collaborative digital pathology framework, the FlexMIm project. As important milestones on the way to routine digital pathology, a series of pioneer international benchmarking initiatives have been launched for mitosis detection (MITOS), nuclear atypia grading (MITOS-ATYPIA) and glandular structure detection (GlaS), some of the fundamental grading components in diagnosis and prognosis. These initiatives allow envisaging a consolidated validation referential database for digital pathology in the very near future. This reference database will need coordinated efforts from all major teams working in this area worldwide, and it will certainly represent a critical bottleneck for the acceptance of all future imaging modules in clinical practice. In line with recent advances in molecular imaging and genetics, keeping the microscopic modality at the core of future digital systems in pathology is fundamental to insure the acceptance of these new technologies, as well as for a deeper systemic, structured comprehension of the pathologies. After all, at the scale of routine whole-slide imaging (WSI; ∼0.22 µm/pixel), the microscopic image represents a structured 'genomic cluster', enabling a naturally structured support for integrative digital pathology approaches. In order to accelerate and structure the integration of this heterogeneous information, a major effort is and will continue to be devoted to morphological microsemiology (microscopic morphology semantics). Besides insuring the traceability of the results (second opinion) and supporting the orchestration of high-content image analysis modules, the role of semantics will be crucial for the correlation between digital pathology and noninvasive medical imaging modalities. In addition, semantics has an important role in modelling the links between traditional microscopy and recent label-free technologies. The massive amount of visual data is challenging and represents a characteristic intrinsic to digital pathology. The design of an operational integrative microscopy framework needs to focus on scalable multiscale imaging formalism. In this sense, we prospectively consider some of the most recent scalable methodologies adapted to digital pathology as marked point processes for nuclear atypia and point-set mathematical morphology for architecture grading. To orchestrate this scalable framework, semantics-based WSI management (analysis, exploration, indexing, retrieval and report generation support) represents an important means towards approaches to integrating big data into biomedicine. This insight reflects our vision through an instantiation of essential bricks of this type of architecture. The generic approach introduced here is applicable to a number of challenges related to molecular imaging, high-content image management and, more generally, bioinformatics.