AI-Powered Cancer Patient Application for Colorectal Cancer Care

Coach Name

Juan Juan

EU Organization

Arcella Holding B.V. (Netherlands)

Members

  • Anna O’Leary
  • Pavels Kartasevs

US Organization

Hekademeia Research Solutions (USA)

Members

  • Blake Anderson
  • Grant Anderson

Project Overview

ARC advances Arcella’s next-generation AI-powered Cancer Patient Application, a digital medical device designed to support patients with colorectal cancer (CRC) throughout diagnosis, treatment, recovery, and long-term follow-up. The experiment validated how AI technologies, federated learning, decentralised infrastructures, and GDPR/HIPAA-aligned architectures can empower patients while enabling secure, cross-border data collaboration.

The project delivered and tested 25+ interactive features including symptom and mood tracking, AI-enabled journaling, personalised treatment guidance, medication and appointment support, clinical-trial matching, structured medical folders, well-being tools, and a moderated peer community. These modules were co-designed with CRC patients, oncologists, nurses, and specialists to ensure clinical relevance, inclusivity, and real-world usability.

ARC demonstrates that a patient-centred, privacy-first oncology companion can meaningfully improve daily disease management while generating de-identified, ethically sourced real-world evidence for research. The resulting platform is technically ready for clinical validation in 2026.

Methods and approaches

AI-driven, decentralised digital health architecture

ARC implemented a GDPR- and HIPAA-compliant cloud infrastructure, using AES-256 encryption, role-based access controls, containerisation, and federated learning to enable secure EU–US data processing. The system integrates an OAuth-based API layer, OpenAPI 3.1 documentation, and a structured database schema for high-quality, interoperable medical data.

Human-centred design combining medical, social & behavioural insights

The team conducted interviews and informal consultations with CRC patients and clinicians to shape the app’s health tracking, journaling, educational content, UI/UX, and communication style. These insights ensured empathy, clarity, emotional safety, and accessibility (WCAG 2.1 AA).

Key Achievements

Fully developed MVP infrastructure with secure backend, medical folder, dashboard, and AI content engine.

AI-driven modules, including symptom and mood tracking, NLP journaling, personalised health blog, and AI companion.

Clinical-trial matching engine, achieving ~80% accuracy in preliminary tests.

Cross-border interoperability validated between EU and US infrastructures, confirming decentralised compliance.

Community chat forum with <5s latency and a moderation protocol ensuring <24h response.

Complete documentation set: system architecture, cybersecurity protocols, API schemas, UI/UX design, maintenance and recovery plans, and clinical-trial benchmarking.

Large-scale synthetic datasets for testing (patient records and trial profiles) and a verified oncologist database (EU + US).

All KPIs across the four stages were successfully completed.

Impact & Results

Scientific Impact

ARC advances decentralised, ethical, and interoperable digital health infrastructures capable of supporting AI-driven clinical decision support, real-world evidence generation, and cross-border oncology research.

Societal Impact

By addressing gaps in cancer communication, accessibility, and personalised support, ARC empowers patients across diverse demographic and geographic backgrounds—especially those in remote areas or underserved regions. It reduces anxiety, improves symptom management, and fosters equitable access to cancer information and clinical trials.

Economic Impact

ARC has the potential to save over €100 million annually in avoidable EU healthcare costs by improving adherence, reducing emergency visits, preventing complications, and enhancing care coordination. It also enables future revenue models (B2B licensing, reimbursement, research partnerships).

Industrial Impact

The project provides a scalable model for integrating AI, decentralised data governance, and secure clinical-grade tools into digital oncology workflows—enhancing innovation in the broader European health-tech ecosystem.

EU/US Collaboration

The partnership with Hekademeia strengthened cross-border technical capabilities and compliance workflows, which will continue into the clinical validation and expansion phases.

Publications and Open-Source Contributions

  • Open-source conceptual and technical materials published under CC BY 4.0:
  • User testing videos, prototype demonstrations, UI mockups, API documentation, and cybersecurity protocols (public links included in the report).
  • MVP accessible in test mode: https://arcella.research-app.org

Future directions

  • Launch clinical validation at Radboud University Medical Center in 2026.
  • Enhance AI assistant features, multilingual support, and clinical-trial matching.
  • Expand interoperability testing with research organisations and hospitals.
  • Continue EU–US collaboration for international scaling.
  • Prepare additional scientific publications on federated learning, accessibility, and oncology digital twins.

Subscribe to our newsletter
SUBSCRIBE

Horizon Europe – Grant Agreement number 101092887

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Union’s Horizon Europe research and innovation programme. Neither the European Union nor the granting authority can be held responsible for them.