AI assistance in tumor multidisciplinary teams

Area of Science:

• Oncology
• Medical Informatics
• Artificial Intelligence

Background:

• Multidisciplinary teams (MDTs) in cancer care face operational inefficiencies due to manual data synthesis and fragmented decision-making processes.
• Current methods for preparing patient data for MDTs are time-consuming, prone to information overload, and hinder systematic learning.
• Clinical decisions are often stored in static documents, preventing the capture of rationale crucial for research and continuous improvement.

Purpose of the Study:

• To propose artificial intelligence (AI), specifically natural language processing (NLP) and large language models (LLMs), as integrated solutions to enhance MDT operational efficiency.
• To explore how AI can facilitate the seamless integration of multimodal patient data (imaging, histopathology, genomics, clinical data) for improved cancer care.
• To address the challenges associated with AI implementation, including validation, ethical governance, and regulatory oversight in clinical settings.

Main Methods:

• Utilizing a tiered AI approach, from small NLP models for data extraction to foundational generative NLP for evidence synthesis.
• Integrating diverse datasets including imaging, histopathology, genomics, and clinical information.
• Developing validated frameworks and addressing ethical and regulatory considerations for AI adoption in heterogeneous operational environments.

Main Results:

• AI integration promises to enhance diagnostic accuracy and personalize treatment plans by synthesizing complex patient data.
• Streamlined data management and decision-making processes are expected through AI-powered tools.
• Establishment of feedback loops through AI can support continuous learning and research in cancer care.

Conclusions:

• AI, particularly NLP and LLMs, offers a transformative potential to overcome MDT inefficiencies in cancer care.
• Successful AI integration requires addressing validation, ethical governance, and regulatory challenges through international collaborative efforts.
• AI implementation can significantly improve patient outcomes and optimize resource utilization in the complex cancer management journey.