Mixture of Multicenter Experts in Multimodal Generative AI for Advanced Radiotherapy Target Delineation

2024-09-27

arXiv.org

AI Summary

Poor generalizability and regional bias of medical AI in radiotherapy target delineation stem from inter-institutional clinical practice heterogeneity. Method: We propose the first few-shot adaptive generative framework that requires no cross-institutional data sharing. It introduces a novel Multi-center Mixture of Experts (MoME) architecture, integrating multimodal generative AI, Mixture of Experts (MoE), and cross-site federated few-shot adaptation—enabling collaborative modeling of heterogeneous clinical knowledge using only small amounts of local imaging data and structured textual annotations per site. Results: Our method significantly outperforms baselines in prostate cancer target delineation; achieves the largest gains under severe distribution shift or extreme data scarcity; and enables rapid deployment of robust, personalized AI-assisted systems in resource-constrained healthcare settings.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Clinical experts employ diverse philosophies and strategies in patient care, influenced by regional patient populations. However, existing medical artificial intelligence (AI) models are often trained on data distributions that disproportionately reflect highly prevalent patterns, reinforcing biases and overlooking the diverse expertise of clinicians. To overcome this limitation, we introduce the Mixture of Multicenter Experts (MoME) approach. This method strategically integrates specialized expertise from diverse clinical strategies, enhancing the AI model's ability to generalize and adapt across multiple medical centers. The MoME-based multimodal target volume delineation model, trained with few-shot samples including images and clinical notes from each medical center, outperformed baseline methods in prostate cancer radiotherapy target delineation. The advantages of MoME were most pronounced when data characteristics varied across centers or when data availability was limited, demonstrating its potential for broader clinical applications. Therefore, the MoME framework enables the deployment of AI-based target volume delineation models in resource-constrained medical facilities by adapting to specific preferences of each medical center only using a few sample data, without the need for data sharing between institutions. Expanding the number of multicenter experts within the MoME framework will significantly enhance the generalizability, while also improving the usability and adaptability of clinical AI applications in the field of precision radiation oncology.

Problem

• Addressing AI bias in medical models without data sharing
• Integrating diverse clinical expertise to enhance model generalizability
• Enabling localized customization for radiotherapy target delineation

Innovation

• Mixture of Multicenter Experts framework
• Few-shot multimodal training approach
• Cross-center generalization without data sharing