AI Digital Twin Platform Supports Personalized Cancer Treatment Decisions

Oncology teams face persistent challenges in tailoring treatment plans as tumors evolve and responses vary across patients. Static snapshots from imaging or genomics offer limited guidance for complex decisions, especially in immunotherapy where outcomes are difficult to predict. The need for adaptive, data-rich models is pressing to improve precision and resource use. To help address this challenge, researchers have now developed a patient‑centric artificial intelligence platform that builds a dynamic digital twin to guide personalized cancer care.

The Artificial Intelligence (AI) Virtual Patient Simulation System from The Hong Kong Polytechnic University (PolyU) creates a continuously updated digital twin by integrating multimodal data, including genomic profiles, medical imaging, and clinical records. The platform tracks changes in a patient’s condition in real time and predicts the potential effectiveness of different treatment options. It is designed to support more precise, individualized planning and to streamline multidisciplinary decision-making.

The system pairs a dedicated platform for health care professionals with a patient‑facing mobile application. Clinicians can review aggregated genomic data, imaging, pathology reports, laboratory results, and clinical notes to gain a comprehensive view of disease status and treatment response. Patients can upload records, log daily symptoms, and track status, while an encrypted Deep Feature QR code enables secure transfer of medical data across clinics, hospitals, and devices.

To advance cancer decision support, the team introduced a clinical, data‑driven, multiscale artificial intelligence framework for predicting immunotherapy response in non‑small cell lung cancer. The Visual‑Global Relation Fusion Network (ViGNet) integrates histopathological image features with clinical data, including gene expression profiles and cancer‑type text. ViGNet uses a multiscale visual encoder and a gene‑driven encoder to relate image and genomic features to treatment response, and in qualitative and quantitative evaluations it outperformed baseline approaches, achieving 82.55% discrimination performance in predicting immunotherapy response. The work is published in Medical Image Analysis.

The platform is positioned for use in cancer and critical care settings where disease trajectories are complex and choices are numerous. By turning static records into dynamic resources, it aims to enhance diagnostic support, condition monitoring, and treatment assessment. The approach emphasizes close collaboration between professionals and patients, encouraging active participation and more efficient information sharing.

“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 health care teams when a patient’s biomarkers or symptoms show abnormalities. This technology helps shorten diagnosis and assessment times, supporting health care professionals in developing more precise, effective and personalized treatment plans for patients with cancer or other critical illnesses,” said Lawrence Chan, associate professor of the PolyU Department of Health Technology and Informatics.

Related Links
The Hong Kong Polytechnic University