[論文レビュー] W-DUALMINE: Reliability-Weighted Dual-Expert Fusion With Residual Correlation Preservation for Medical Image Fusion

Md. Jahidul Islam|arXiv (Cornell University)|Jan 13, 2026

Advanced Image Fusion Techniques被引用数 0

ひとこと要約

tldr: W-DUALMINE uses reliability-weighted dual experts and a residual-to-average fusion to preserve global statistics (MI/CC) while enhancing local details across CT–MRI, PET–MRI, and SPECT–MRI datasets, outperforming AdaFuse and ASFE-Fusion.

ABSTRACT

Medical image fusion integrates complementary information from multiple imaging modalities to improve clinical interpretation. However, existing deep learningbased methods, including recent spatial-frequency frameworks such as AdaFuse and ASFE-Fusion, often suffer from a fundamental trade-off between global statistical similaritymeasured by correlation coefficient (CC) and mutual information (MI)and local structural fidelity. This paper proposes W-DUALMINE, a reliability-weighted dual-expert fusion framework designed to explicitly resolve this trade-off through architectural constraints and a theoretically grounded loss design. The proposed method introduces dense reliability maps for adaptive modality weighting, a dual-expert fusion strategy combining a global-context spatial expert and a wavelet-domain frequency expert, and a soft gradient-based arbitration mechanism. Furthermore, we employ a residual-to-average fusion paradigm that guarantees the preservation of global correlation while enhancing local details. Extensive experiments on CT-MRI, PET-MRI, and SPECT-MRI datasets demonstrate that W-DUALMINE consistently outperforms AdaFuse and ASFE-Fusion in CC and MI metrics while

研究の動機と目的

• Objective 1: Address the trade-off between global statistical similarity and local structural fidelity in medical image fusion.
• Objective 2: Introduce dense reliability maps to suppress artifacts from unreliable regions before fusion.
• Objective 3: Develop a dual-expert fusion architecture (spatial and wavelet frequency) with a soft gradient arbitration mechanism.
• Objective 4: Adopt a residual-to-average fusion paradigm to theoretically preserve high CC and MI with source inputs.

提案手法

• Method 1: Siamese multi-scale encoders extract hierarchical features from each modality.
• Method 2: Dense reliability maps predict pixel-wise reliability scores to weight feature fusion adaptively.
• Method 3: Dual-expert fusion at each scale: a Global Context Spatial Expert and a Wavelet Frequency Expert.
• Method 4: Soft Gradient Mixer dynamically arbiters between spatial and wavelet outputs based on edge strength.
• Method 5: Residual-to-Average Decoder reconstructs a fused image by adding a residual to the mean of inputs, ensuring global statistical preservation.
• Method 6: Compound loss with five terms (L_avg, L_grad, L_cc, L_mi, L_rec) balances content fidelity, edge preservation, correlation, information, and reconstruction.

Figure 1: Architecture of the Reliability-Weighted Dual-Expert Fusion Network. The framework processes multi-modal inputs (e.g., CT/PET and MRI) through a Siamese encoder composed of ResBlocks. Feature maps are weighted by a Reliability Estimation module before entering two parallel expert branches:

実験結果

リサーチクエスチョン

• RQ1 研究課題1: Reliability-weighted feature modeling は信頼できない領域からのアーチファクトを抑制し、融合品質を向上させるか？
• RQ2 研究課題2: ソフトグラデーション仲裁を備えた双专家（空間・ウェーブレット）融合経路は globally statistics を保持しつつ局所的ディテールを強化するか？
• RQ3 研究課題3: Residual-to-Average 融合スキームはソースモーダリティと高い相互情報量と相関係数を保証するか？

主な発見

Method	EN	MI	CC	PSNR	FMI
AdaFuse (CT–MRI)	5.0592±0.2346	3.3570±0.1978	0.8306±0.0238	64.0004±0.7757	0.4343±0.0170
ASFE-Fusion (CT–MRI)	5.4855±0.2734	3.1463±0.1605	0.8302±0.0238	63.9884±0.7845	0.4066±0.0180
W-DUALMINE (CT–MRI)	4.3394±0.2502	3.6059±0.2419	0.8308±0.0238	64.0891±0.7917	0.4746±0.0210

• 発見1: CT–MRI で W-DUALMINE は MI = 3.6059 および CC = 0.8308 を達成し、グローバル統計的類似度で競合を上回る。
• 発見2: CT–MRI で W-DUALMINE は PSNR = 64.0891 および FMI = 0.4746 を示し、エッジ保存と特徴忠実度が高い。
• 発見3: PET–MRI で W-DUALMINE は MI = 4.3068 および FMI = 0.5064、CC = 0.8686 を達成し、機能情報伝達とテクスチャ保存が改善。
• 発見4: SPECT–MRI で W-DUALMINE は MI = 4.0016、CC = 0.9116、PSNR = 64.9084 を記録し、解像度差とノイズ下での堅牢な性能を強調。

Figure 2: The overall framework of W-DUALMINE. The architecture consists of Siamese encoders extracting multi-scale features, which are fused and projected for contrastive learning. The network is optimized via a composite loss function comprising: (1) Average Content Loss ( $\mathcal{L}_{avg}$ ) fo