dorsal/arxiv
View SchemaW-DUALMINE: Reliability-Weighted Dual-Expert Fusion With Residual Correlation Preservation for Medical Image Fusion
| Authors | Md. Jahidul Islam |
|---|---|
| Categories | |
| ArXiv ID | 2601.08920vv1 |
| URL | https://arxiv.org/abs/2601.08920 |
| License | http://creativecommons.org/licenses/by/4.0/ |
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
{
"annotation_id": "c719543e-3cdc-4ec3-aa51-9fd0182774ff",
"date_created": "2026-02-17T05:53:19.779000Z",
"date_modified": "2026-02-17T05:53:19.779000Z",
"file_hash": "199201b968a6be9700921d282e01b51f3ed19b01af5a7401610c815b4b16b2c8",
"private": false,
"record": {
"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",
"arxiv_id": "2601.08920",
"authors": [
"Md. Jahidul Islam"
],
"categories": [
"eess.IV",
"cs.CV",
"cs.NA",
"math.NA"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "W-DUALMINE: Reliability-Weighted Dual-Expert Fusion With Residual Correlation Preservation for Medical Image Fusion",
"url": "https://arxiv.org/abs/2601.08920",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "92fd8aee-d55b-481e-853b-53e3523ab81b",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}