Oxford Pilots TrustedMDT: Multi-Agent AI Integrated Into Microsoft Teams To Support Cancer Treatment Planning
In Brief
Oxford, in collaboration with Microsoft, has piloted a multi-agent AI assistant within Microsoft Teams to support real-world cancer tumour board decision-making at Oxford University Hospitals.
University of Oxford unveiled TrustedMDT, a multi-agent AI system designed to assist medical specialists during cancer treatment planning meetings.
Developed in collaboration with technology company Microsoft, the AI tool has been integrated into Microsoft Teams and will be piloted at Oxford University Hospitals NHS Foundation Trust, representing one of the first uses of agentic AI in a clinically realistic tumour board environment.
Multidisciplinary Tumour Board meetings in the UK bring together radiologists, pathologists, surgeons, and oncologists to review diagnostic results and develop treatment plans. Rising caseloads, however, are placing increasing pressure on expert capacity.
Research from Cancer Research UK indicates that teams often spend less than two minutes discussing each patient, with critical information gaps contributing to delays in 7% of cases, which can impact treatment timelines, research opportunities, and clinician workload.
TrustedMDT was designed to address these pressures by automating data synthesis and analysis through three coordinated AI agents.
The Clinical Summarisation Agent reviews electronic health records—including radiology, pathology, and biomarker data—to generate concise tumour-specific summaries. The Cancer Staging Agent evaluates disease progression using international staging standards, while the Treatment Planning Agent produces evidence-based treatment recommendations aligned with professional guidelines.
Together, these agents aim to enhance the efficiency and accuracy of tumour board decision-making.
Oxford University Hospitals Pilots TrustedMDT, Assisting Oncology Tumour Board Decisions
Dr. Andrew Soltan, Lead Investigator and Specialty Registrar in Medical Oncology at Oxford University Hospitals, explained that traditional chatbots are insufficient for the complexity of oncology, prompting the development of a hierarchical multi-agent system. In this architecture, each agent comprises sub-agents focused on specific data sets and equipped with relevant tools, requiring the system to reason through clinical guidelines and cross-check recommendations against patient histories to reduce errors.
The Oxford team deployed these custom agents within Microsoft Teams using the healthcare agent orchestrator, integrating the AI directly into existing multidisciplinary tumour board workflows.
Dr. Soltan emphasized that the system is designed to support clinical processes without disruption, functioning as a ‘digital collaborator’ that allows clinicians to provide input in real time and review the rationale behind AI-generated recommendations, with final decisions remaining under human control.
Oxford University Hospitals has received approval to conduct a two-phase pilot study to evaluate TrustedMDT’s accuracy, usability, and technical performance. The first phase benchmarks AI outputs against expert decisions using anonymized cancer cases, while the second phase simulates tumour board meetings to assess how effectively the system summarizes information, supports discussion, and drafts treatment plans within realistic clinical workflows. Clinical support is provided by OUH resident doctors.
Dr. Ben Attwood, Chief Digital Officer at OUH, noted that the hospital is committed to exploring innovations that enhance MDT preparation and operations while adhering to established governance and information security standards.
David Ardman, Corporate Vice President of Microsoft Health and Life Sciences, described the multi-agent system as a novel approach in healthcare AI, enabling clinicians to interact dynamically with specialized agents within Teams to reduce cognitive load and improve decision support.
If validated, TrustedMDT could improve communication among specialists, shorten treatment timelines, and expand access to clinical trials. The pilot study represents an initial step toward demonstrating the system’s potential, generating evidence to inform further technical development and guide future, larger-scale evaluations before clinical deployment.
Disclaimer
In line with the Trust Project guidelines, please note that the information provided on this page is not intended to be and should not be interpreted as legal, tax, investment, financial, or any other form of advice. It is important to only invest what you can afford to lose and to seek independent financial advice if you have any doubts. For further information, we suggest referring to the terms and conditions as well as the help and support pages provided by the issuer or advertiser. MetaversePost is committed to accurate, unbiased reporting, but market conditions are subject to change without notice.
About The Author
Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.
More articles
Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.
