A new AI system transforms how doctors treat cancer by creating a continuously updated digital model of each patient, integrating genomic data, medical imaging, and clinical records to predict which treatments will work best. Rather than analyzing a single CT scan or static medical report, the system tracks real-time changes in a patient's condition and simulates the effectiveness of different treatment options, helping healthcare teams develop more precise and personalized medical solutions.

What Makes This Digital Twin Approach Different?

Traditional medical AI tools often rely on isolated pieces of information, making it difficult to understand how a patient's condition changes over time. The "AI Virtual Patient Simulation System," developed by Prof. Lawrence Chan and his team at Hong Kong Polytechnic University's Department of Health Technology and Informatics, takes a fundamentally different approach by combining multiple data streams into one dynamic model.

The system works through two integrated platforms: one for healthcare professionals and one for patients. The professional platform consolidates genomic data, medical imaging, pathology reports, laboratory test results, and clinical records into a comprehensive overview. Meanwhile, the patient-facing mobile application allows individuals to upload medical records, log daily symptoms, and track their health status in real time.

"The AI Virtual Patient Simulation System is an innovative and comprehensive platform that integrates diagnosis, monitoring and treatment assessment. In addition to identifying subtle yet crucial pathological connections across multimodal data, the system can also act as a 'monitoring sentinel', alerting healthcare teams when a patient's biomarkers or symptoms show abnormalities," said Prof. Lawrence Chan.

Prof. Lawrence Chan, Associate Professor, Department of Health Technology and Informatics, Hong Kong Polytechnic University

The system is particularly well-suited for cancer and critical care, where disease progression can be complex, treatment options are diverse, and medical costs are high. By enabling patients to shift from passively receiving treatment to actively participating in their health management, the platform strengthens doctor-patient collaboration and streamlines multidisciplinary consultations and referral processes.

How Does the AI Predict Treatment Response?

To advance cancer care applications, the research team introduced a specialized AI framework called the Visual-Global Relation Fusion Network, or ViGNet. This framework is designed specifically to predict immunotherapy response in patients with non-small cell lung cancer by integrating histopathological image features with clinical data, including gene expression profiles and cancer-type information.

In testing, ViGNet achieved 82.55% discrimination performance in predicting immunotherapy response, outperforming baseline approaches in both qualitative and quantitative evaluations. This level of accuracy represents a significant advancement in supporting practical deployment of AI methods in clinical settings and provides important insights for personalized treatment and clinical decision-making.

Steps to Implement AI-Driven Personalized Cancer Care

• Data Integration: Consolidate genomic data, medical imaging, pathology reports, laboratory results, and clinical records into a single dynamic patient model that updates in real time as new information becomes available.
• Multidisciplinary Collaboration: Use the healthcare professional platform to streamline consultations across different specialists, enabling faster diagnosis and more coordinated treatment planning for complex cases.
• Patient Engagement: Implement patient-facing mobile applications that allow individuals to upload records, log symptoms, and track health status, transforming them from passive recipients to active participants in their care.
• Predictive Analytics: Deploy specialized AI frameworks like ViGNet to predict treatment response and simulate the effectiveness of different therapeutic options before implementation.
• Real-Time Monitoring: Establish alert systems that notify healthcare teams when patient biomarkers or symptoms show abnormalities, enabling early intervention and proactive care adjustments.

What Are the Real-World Applications and Next Steps?

The system has already gained significant recognition in the healthcare technology sector. It was recently showcased at Mobile World Congress 2026 in Barcelona, Spain, where it was shortlisted as a finalist for the 2026 Global Mobile Awards in the category of Best Mobile Innovation for Connected Health and Wellbeing.

The project has received funding from the Hong Kong Polytechnic University Micro Fund and Seed Fund, as well as the Guangdong-Hong Kong-Macau Greater Bay Area Innovation and Entrepreneurship Incubation Programme. It has also been conditionally accepted into the Hong Kong Science and Technology Park's Incubation Programme and is advancing into commercialization and industrialization.

As the system is deployed in clinical settings, the real-world data it collects will inform drug development, clinical trials, and treatment plan optimization, further enhancing the medical innovation ecosystem and benefitting more patients. The ability to securely transfer medical data across different clinics, hospitals, and devices through an encrypted Deep Feature QR code also addresses a critical challenge in healthcare: maintaining privacy while improving data-sharing efficiency.

This advancement reflects a broader shift in healthcare technology toward integrated, patient-centric AI systems that move beyond single-task analysis to comprehensive care orchestration. By combining real-time monitoring, predictive analytics, and patient engagement, the AI Virtual Patient Simulation System demonstrates how technology can support more precise, effective, and personalized treatment plans for cancer and other critical illnesses.