Colorectal cancer (CRC) remains a major public health challenge, demanding effective, accessible, and patient-friendly screening strategies. ONCOSCREEN, a Mission Cancer EU-funded initiative, aims to transform CRC screening by leveraging novel technologies, artificial intelligence, and a thoughtfully co-designed system architecture that unites clinicians, researchers, patients, and policymakers (Figure 1). From early design to advanced development phases, end users have been actively involved in shaping the system’s requirements and functionalities—ensuring their insights are embedded into the architecture. This iterative, participatory process guarantees that the final product is not only technically robust but also aligned with real-world needs.
From its inception, ONCOSCREEN has been envisioned as an ecosystem of modules designed to work together to address every critical stage of the CRC screening process—from early detection and personalized risk assessment to clinical decision-making and policy support. To fulfill its mission, ONCOSCREEN’s architecture is built to be flexible, secure, compliant, and scalable—allowing seamless integration across diverse healthcare environments.
The front end of the platform, ONCO-CAWA (Figure 2), is a personalized companion mobile fitness application designed for self-assessment, monitoring, and citizen awareness. This application serves as the primary interface for individuals and patients’ main access point to the ONCOSCREEN platform. The ecosystem’s diagnostic toolkit includes ONCO-VOC, a breath analyzer for VOC-based CRC detection; ONCO-CRISPR, a cutting-edge liquid biopsy tool using CRISPR-Cas technology; ONCO-CTC, a novel solution for isolating circulating tumor cells; and ONCO-NMR, an innovative protocol for uncovering a rich spectrum of metabolic, lipoprotein, and glycoprotein parameters for CRC screening.
The outputs of these four diagnostic tools feed into a personalized risk stratification engine, ONCO-RISTE, which, combined with other parameters, stratifies individuals into different CRC risk groups. This comprehensive risk profile, along with personalized data from ONCO-CAWA, feeds into the Clinical Decision Support Platform (ONCO-CLIDE), which uses AI-driven analytics to identify CRC-specific patterns with high sensitivity and specificity. ONCO-CLIDE serves as the second access point of the project for decision support and integrated diagnosis, specially designed to address the needs of clinicians and healthcare professionals (Figure 3).
Other key modules include ONCO-EVIDA, a policymaker-oriented visualization platform; the training platforms ONCO-AICO and ONCO-AITI; and the data fusion and cataloguing infrastructure ONCO-BIOBA. ONCO-EVIDA uses the output from ONCO-RISTE to offer policymakers a clear view of the geospatial distribution of risk groups and provides actionable recommendations to support evidence-based decisions to advance the goal of precision public health. Meanwhile, ONCO-AICO and ONCO-AITI aim to enhance the training of healthcare professionals by offering user-friendly tools that improve diagnostic accuracy while reducing training time, thereby supporting widespread adoption and building trust.
The data fusion engine, ONCO-BIOBA, is designed to ingest, store, process, and analyze heterogeneous data streams across the project—delivering insights and identifying patterns. The engine functions as a centralized data catalogue that presents descriptions of datasets in a harmonized, FAIR-compliant format, making it easy for research institutions, organizations, and industry to explore available resources and establish collaborations.
The platform employs a hybrid data management model, combining federated databases with a centralized repository. This setup enables hospitals and research centers to maintain control over their data while allowing the system to extract, harmonize, and analyze key information efficiently and securely. Security and compliance are embedded throughout the design. Following a “security-by-design” approach, the system incorporates encrypted data pipelines, anonymization protocols, and federated management—all within a Virtual Private Network (VPN), orchestrated through technologies like Kafka SSL.
All medical data adhere to leading standards such as FHIR and SNOMED CT, ensuring interoperability, privacy, and data protection across the ecosystem while still maintaining human oversight at all stages. Furthermore, architecture is modular and adaptable, allowing the integration of new tools or data sources as technology evolves.
With this comprehensive and interconnected architecture, ONCOSCREEN offers an end-to-end CRC screening solution that is scalable across Europe. It holds the potential to make early detection more accurate, accessible, and equitable, ultimately reducing the burden of disease through innovation and collaboration.
By combining cutting-edge diagnostics with a co-designed digital infrastructure, ONCOSCREEN exemplifies how digital health can be built inclusively and effectively to have a long-lasting impact.
