# AdaFuse: Adaptive Multimodal Fusion for Lung Cancer Risk Prediction via Reinforcement Learning

arXiv: https://arxiv.org/abs/2602.00347 · HTML: https://mlsanity.com/paper/2602.00347/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for clinical treatment and surgery, one of 15 directions in Deep reinforcement learning methods.
- Builds on 15 of its 40 references that are in the map.
- First posted on arXiv 30 Jan 2026.

## Placed in

- Deep reinforcement learning methods / [RL for clinical treatment and surgery](https://mlsanity.com/deeprl/rl-for-clinical-treatment-and-surgery/)

## Builds on (most cited first, 15 of 15)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [BioBERT: a pre-trained biomedical language representation model for biomedical text mining](https://mlsanity.com/paper/1901.08746/) (2019)
- [Tensor Fusion Network for Multimodal Sentiment Analysis](https://mlsanity.com/paper/1707.07250/) (2017)
- [Efficient Low-rank Multimodal Fusion with Modality-Specific Factors](https://mlsanity.com/paper/1806.00064/) (2018)
- [Pathomic Fusion: An Integrated Framework for Fusing Histopathology and Genomic Features for Cancer Diagnosis and Prognosis](https://mlsanity.com/paper/1912.08937/) (2019)
- [FuseMoE: Mixture-of-Experts Transformers for Fleximodal Fusion](https://mlsanity.com/paper/2402.03226/) (2024)
- [Dynamic Multimodal Fusion](https://mlsanity.com/paper/2204.00102/) (2022)
- [Multi-modal Gated Mixture of Local-to-Global Experts for Dynamic Image Fusion](https://mlsanity.com/paper/2302.01392/) (2023)
- [Survival Prediction of Brain Cancer with Incomplete Radiology, Pathology, Genomics, and Demographic Data](https://mlsanity.com/paper/2203.04419/) (2022)
- [Reinforcement Learning in Medical Image Analysis: Concepts, Applications, Challenges, and Future Directions](https://mlsanity.com/paper/2206.14302/) (2022)
- [Low Rank Fusion based Transformers for Multimodal Sequences](https://mlsanity.com/paper/2007.02038/) (2020)
- [G-MIND: An End-to-End Multimodal Imaging-Genetics Framework for Biomarker Identification and Disease Classification](https://mlsanity.com/paper/2101.11656/) (2021)
- [CoRe: An Efficient Coarse-refined Training Framework for BERT](https://mlsanity.com/paper/2011.13633/) (2020)
- [M2Fusion: Bayesian-based Multimodal Multi-level Fusion on Colorectal Cancer Microsatellite Instability Prediction](https://mlsanity.com/paper/2401.07854/) (2024)

## Abstract (arXiv)

Multimodal fusion has emerged as a promising paradigm for disease diagnosis and prognosis, integrating complementary information from heterogeneous data sources such as medical images, clinical records, and radiology reports. However, existing fusion methods process all available modalities through the network, either treating them equally or learning to assign different contribution weights, leaving a fundamental question unaddressed: for a given patient, should certain modalities be used at all? We present AdaFuse, an adaptive multimodal fusion framework that leverages reinforcement learning (RL) to learn patient-specific modality selection and fusion strategies for lung cancer risk prediction. AdaFuse formulates multimodal fusion as a sequential decision process, where the policy network iteratively decides whether to incorporate an additional modality or proceed to prediction based on the information already acquired. This sequential formulation enables the model to condition each selection on previously observed modalities and terminate early when sufficient information is available, rather than committing to a fixed subset upfront. We evaluate AdaFuse on the National Lung Screening Trial (NLST) dataset. Experimental results demonstrate that AdaFuse achieves the highest AUC (0.762) compared to the best single-modality baseline (0.732), the best fixed fusion strategy (0.759), and adaptive baselines including DynMM (0.754) and MoE (0.742), while using fewer FLOPs than all triple-modality methods. Our work demonstrates the potential of reinforcement learning for personalized multimodal fusion in medical imaging, representing a shift from uniform fusion strategies toward adaptive diagnostic pipelines that learn when to consult additional modalities and when existing information suffices for accurate prediction.