Beyond Ontologies: What Does It Take to Build Dependable Cancer Research Infrastructure? A Systematic Review

This project conducts a systematic literature review (SLR) to analyze how knowledge models—such as ontologies, semantic schemas, conceptual models, and knowledge graphs—are used to support data representation, integration, and interoperability in cancer research infrastructures. Modern oncology research depends on the management of heterogeneous data sources, including clinical records, genomic data, imaging, and research datasets distributed across multiple repositories and institutions. The lack of semantic alignment among these data sources creates significant challenges for data reuse, integration, and collaborative research. Knowledge models have emerged as key mechanisms for structuring biomedical data and enabling semantic interoperability. However, existing research on these models in oncology is fragmented across different modeling approaches, data types, and application contexts. There is currently no comprehensive synthesis examining what types of knowledge models are employed, where they are applied, and what limitations and opportunities remain. To address this gap, this study follows the PRISMA 2020 guidelines for systematic reviews and is methodologically grounded in established SLR practices from software engineering research. The review covers peer-reviewed literature published between 2015 and 2026 and systematically examines: the types of knowledge models used in cancer research data systems, the biomedical data domains in which they are applied, the research and clinical contexts where they are deployed, and the challenges and opportunities associated with their use for data integration and interoperability. The expected outcomes include: (1) a structured classification of knowledge modeling approaches in oncology; (2) an overview of their application domains and functional roles; (3) identification of recurring technical and semantic challenges; and (4) research directions for improving cancer research infrastructures through better knowledge representation practices. By synthesizing evidence across disciplines, this review aims to support researchers, system developers, and data infrastructure designers working toward interoperable and semantically robust cancer research ecosystems.