SAGEAgent: A Self-Evolving Agent for Cost-Aware Modality Acquisition in Multimodal Survival Prediction

2026-07-10

AI Summary

This study addresses the limitation of existing methods in multimodal cancer survival prediction, which often neglect to actively assess the necessity of acquiring diagnostic modalities, thereby imposing unnecessary clinical burdens. To overcome this, the work formulates modality acquisition as a sequential decision-making problem and proposes a large language model–based self-evolving agent that dynamically devises cost-aware, personalized modality acquisition strategies by integrating episodic and semantic memory. For the first time, this approach incorporates a self-evolution mechanism alongside clinical reasoning tools. Evaluated on a glioma dataset, the method achieves competitive survival prediction accuracy while reducing modality usage by 55% on average.

Abstract

Does every cancer patient truly need a complete diagnostic workup for accurate survival prediction? In multimodal clinical oncology, diagnostic modalities follow a clinically mandated order of escalating burden -- from demographics collected at intake to genomic profiling requiring specialized tissue analysis. Current multimodal survival methods either assume all modalities are available or passively handle missing data, but none actively reason about whether acquiring the next modality is justified for a given patient along this ordered workflow. We formulate this as a sequential decision problem and propose SAGEAgent (Sequential Acquisition Guided by Experience), a self-evolving LLM-based clinical agent that decides which diagnostic modalities to acquire for each patient, balancing predictive accuracy against clinical invasiveness. SAGEAgent reasons about each patient's evolving diagnostic state through clinical tools that translate numerical predictions into text, an episodic memory that retrieves similar past cases, and a semantic memory that accumulates reusable decision patterns from experience. Experiments on a glioma cohort combining TCGA-LGG, TCGA-GBM, and BraTS with four diagnostic modalities demonstrate that SAGEAgent achieves competitive survival prediction accuracy while reducing average acquisition burden by 55%.

Problem

• multimodal survival prediction
• cost-aware modality acquisition
• sequential decision making
• clinical oncology
• diagnostic burden

Innovation

• self-evolving agent
• cost-aware modality acquisition
• multimodal survival prediction
• sequential decision making
• large language model (LLM)