Graph guided models improve tumor survival predictions from multiple data types
Multi-Modal Tumor Survival Prediction via Graph-Guided Mixture of Experts
Machine LearningMultiagent Systems
Summary
Predicting how long cancer patients might survive is a complex challenge because it involves many types of data, like medical records, cell images, and genetic information. The authors present a new method that combines several specialized models using a graph-based approach, allowing the system to handle different kinds of data together and even manage missing information. Their approach doesn't rely on just one big model but instead organizes many smaller models that work like experts. Testing this on a lung cancer dataset showed it predicted survival better than existing single or simple combined models.
What this means in practice
- •For hospital data teams: Improve patient outcome predictions by integrating clinical, imaging, and genomic data using graph-guided model combinations for lung cancer cases.
- •For medical software developers: Develop decision support tools that handle incomplete multi-modal data inputs for more accurate cancer survival estimates.$Commercial implications: Enables advanced multi-modal survival prediction software targeted at hospitals and oncology clinics enhancing diagnostic tools.
Tested on one dataset.
Authors
H Mathavan, H Liu
Abstract
Large Language Models (LLMs) have displayed impressive capabilities in handling tasks that require few demonstration examples, making them effective few-shot learners. Despite their potential, LLMs face challenges when it comes to addressing complex real-world tasks that involve multiple modalities or reasoning steps. For example, predicting cancer patients' survival period based on clinical data, cell slides, and genomics poses significant logistical complexities. Although several approaches have been proposed to tackle these challenges, they often fall short in achieving promising performance due to their inability to consider all modalities simultaneously or account for missing modalities, variations in modalities, and the integration of multi-modal data, ultimately compromising their effectiveness. This thesis proposes a novel approach for multi-modal tumor survival prediction to address these limitations. Taking inspiration from recent advancements in LLMs, particularly Mixture of Experts (MoE)-based models, a graph-guided MoE framework is introduced. This framework utilizes a graph structure to manage the predictions effectively and combines multiple models to enhance predictive power. Rather than training a single foundation model for end-to-end survival prediction, the approach leverages a MOE-guided ensemble to manage model callings as tools automatically. By leveraging the strengths of existing models and guiding them through a MOE framework, the aim is to achieve better performance and more accurate predictions in complex real-world tasks. Experiments and analysis on the TCGA-LUAD dataset show improved performance over the individual modal and vanilla ensemble models.