Researchers at the Cambridge Centre for AI in Medicine (CCAIM), together with colleagues from across academia, healthcare, and industry, have published a major new paper titled: Revolutionizing Clinical Trials: A Manifesto for AI-Driven Transformation.
The paper represents a landmark collaboration. Its authors include CCAIM faculty members, students, and researchers, alongside leaders from all the centre’s founding partners -AstraZeneca, GSK, Sanofi, Boehringer Ingelheim, and QuantumBlack. This breadth of expertise reflects CCAIM’s mission: to bring together machine learning, medicine, and industry in pursuit of transformative healthcare.
This manifesto lays out a shared vision for harnessing two of the most promising areas of AI – causal inference and digital twins – to redesign clinical trials. It calls for trials that are faster, more personalised, and more inclusive, while remaining grounded in robust validation and regulatory alignment.
The authors argue that AI can:
- Accelerate discovery and improve probability of success.
- Expand the scope of questions trials can answer.
- Increase inclusivity by better reflecting real-world patient populations.
Crucially, the manifesto emphasises that these advances must be implemented responsibly. It proposes a structured validation framework to ensure AI methods in clinical trials are robust, interpretable, and regulatory-ready. This framework spans nine essential components, from defining context of use, to benchmarking against traditional methods, to ensuring interpretability and ethical compliance.
We have summarised this framework in the figure below:
By embedding validation at the core, the manifesto ensures that innovations such as digital twins can be deployed safely and effectively. By simulating patient trajectories, reducing reliance on large control arms, and personalising treatment paths, digital twins can enable a more dynamic and patient-centred approach to clinical research.
You can find the paper here: https://arxiv.org/pdf/2506.09102
What are digital twins?
To make this vision accessible, CCAIM has introduced digital twins to different audiences through our engagement series:
Revolutionising Healthcare #38 – for clinicians and healthcare professionals, Prof. Mihaela van der Schaar explains how digital twins could transform care through early diagnosis, personalised treatment, and operational efficiency.
Inspiration Exchange #40 – for the machine learning community, Prof. van der Schaar outlines the technical foundations of digital twins, how they differ from prediction and synthetic data, and the challenges ahead.
Together, these sessions show how digital twins are both a technical frontier in machine learning and a practical opportunity for transforming healthcare. You can join these ongoing engagement series to take part in the discussions and explore the future of AI in healthcare with us.
Beyond the manifesto: advancing the frontier
Alongside this high-level vision, CCAIM researchers are developing the technical foundations needed to make digital twins a reality:
- HDTwin (Hybrid Digital Twin): combining mechanistic models with neural networks to create modular, adaptable, and sample-efficient twins.
- CALM-DT (Context-Adaptive Language Model Digital Twin): using large language models to dynamically integrate new variables at inference time without retraining.
- Digital Twin Agents: biologically grounded, high-fidelity twins that can anticipate, reason, and intervene – uniting genomics, clinical practice, and AI reasoning for transformative personalised care.
Support and mission
This paper is also a reflection of CCAIM’s purpose. As outlined in our About section, the CCAIM exists to pioneer next-generation clinical trials and transform personalised healthcare delivery through world-leading research, close partnerships with the NHS and industry, and new education initiatives.
Our partners at AstraZeneca, Boehringer Ingelheim, GSK, McKinsey & Company QuantumBlack, and Sanofi not only provide funding but also strategic input on the challenges that matter in drug development and patient care. This manifesto brings those collaborations to life, uniting CCAIM faculty, students, and industry leaders in a shared vision for how AI can redefine the future of medicine.