FMT:A Multimodal Pneumonia Detection Model Based on Stacking MOE Framework

📅 2025-03-07
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
To address the poor robustness of clinical pneumonia diagnosis models under incomplete or missing multimodal data, this paper proposes a flexible multimodal detection framework. First, a dynamic masking attention mechanism is designed to explicitly model stochastic modality dropout across imaging and textual modalities. Second, a sequential Mixture-of-Experts (MoE) architecture is introduced to enable hierarchical cross-modal feature fusion and decision refinement. Third, ResNet-50 and BERT are jointly fine-tuned for cross-modal representation learning. Evaluated on a small-sample multimodal pneumonia dataset, the framework achieves 94% accuracy, 95% recall, and 93% F1-score—significantly outperforming unimodal baselines and state-of-the-art methods such as CheXMed. These results demonstrate superior generalization under modality missingness and strong clinical applicability.

Technology Category

Application Category

📝 Abstract
Artificial intelligence has shown the potential to improve diagnostic accuracy through medical image analysis for pneumonia diagnosis. However, traditional multimodal approaches often fail to address real-world challenges such as incomplete data and modality loss. In this study, a Flexible Multimodal Transformer (FMT) was proposed, which uses ResNet-50 and BERT for joint representation learning, followed by a dynamic masked attention strategy that simulates clinical modality loss to improve robustness; finally, a sequential mixture of experts (MOE) architecture was used to achieve multi-level decision refinement. After evaluation on a small multimodal pneumonia dataset, FMT achieved state-of-the-art performance with 94% accuracy, 95% recall, and 93% F1 score, outperforming single-modal baselines (ResNet: 89%; BERT: 79%) and the medical benchmark CheXMed (90%), providing a scalable solution for multimodal diagnosis of pneumonia in resource-constrained medical settings.
Problem

Research questions and friction points this paper is trying to address.

Improves pneumonia diagnosis accuracy using AI and multimodal data.
Addresses challenges like incomplete data and modality loss.
Provides scalable solution for resource-constrained medical settings.
Innovation

Methods, ideas, or system contributions that make the work stand out.

ResNet-50 and BERT for joint learning
Dynamic masked attention for robustness
Sequential MOE for decision refinement
🔎 Similar Papers
No similar papers found.
J
Jingyu Xu
School of Informatics, Computing and Cyber Systems, Northern Arizona University, Arizona, U.S
Y
Yang Wang
Cluster BI Inc, Toronto, Canada