Skip to content

Energy- and Contrast-Aware NSST-Based Medical Image Fusion with Anisotropic Diffusion Enhancement

Jul 2026 · Recent Patents on Engineering · Vol 20 · 0 citations

TL;DR

ENCAF-NSST-AD is an effective MIF approach that can enhance clinical decision support in medical imaging, and achieves superior fusion quality compared with existing state-of-the-art methods, exhibiting enhanced visual clarity and higher values across standard quantitative metrics.

Abstract

Medical image fusion (MIF) integrates complementary information from multiple modalities to enhance diagnostic accuracy. However, existing approaches often struggle to suppress noise while preserving critical anatomical details. We propose ENCAF-NSST-AD, a fusion technique that combines the Non-Subsampled Shearlet Transform (NSST) with anisotropic diffusion (AD) refinement. Lowfrequency base layers (LFBL) are fused using an energy-based weighting strategy, while high-frequency detail layers (HFDL) are integrated through a contrast-aware rule. The reconstructed image is further enhanced using AD filtering with optimized parameters to improve clarity and preserve structural information. Experiments on standard medical image datasets demonstrate that ENCAF-NSST-AD consistently outperforms existing methods in both visual quality and quantitative metrics, including FMI, FF, entropy, and standard deviation. Experiments on benchmark datasets demonstrate that ENCAF-NSST-AD achieves superior fusion quality compared with existing state-of-the-art methods, exhibiting enhanced visual clarity and higher values across standard quantitative metrics. The proposed method effectively balances noise suppression with edge preservation, addressing limitations of traditional MIF schemes. By combining adaptive fusion rules with edgepreserving diffusion, ENCAF-NSST-AD preserves global contrast and fine anatomical boundaries, resulting in fused images more suitable for clinical interpretation. Its energy- and contrast- aware fusion with AD refinement represents a clinically relevant and patent-oriented advancement. The integration of NSST with adaptive fusion rules and AD refinement produces robust and diagnostically meaningful fusion outcomes. ENCAF-NSST-AD is an effective MIF approach that can enhance clinical decision support in medical imaging.

View source

Similar papers

Conference Sep 2026

Task-Aware Enhancement of Low-SNR Prostate DWI on the FastMRI Prostate Dataset

Prostate diffusion-weighted imaging (DWI) is highly vulnerable to low signal-to-noise ratio (SNR), which can degrade image quality and obscure diagnostically important tissue characteristics. This paper proposes a task-aware enhancement framework for low-SNR prostate DWI using the public fastMRI Prostate dataset. The m...

Omar Ahmed Mahmood, A. Yousif, Z. Omar et al. · 0 citations
Aug 2026

MSGF-Net: a multi-scale spectral-gradient fusion network for perceptual image quality assessment of low-dose CT

The proposed MSGF-Net embeds a Hybrid Gradient-Frequency Attention Module (HG-FAM) within the residual bottlenecks to achieve dual-domain synergistic perception of global spectral energy and spatial gradient details and significantly outperforms existing main-stream methodologies.

Changke Yao, Li-Bin Lu, Yujie Luo et al. · 0 citations
Sep 2026

A Reliability-Guided Fusion and Dynamic Dual-Domain Mask-Modulated Shearlet Network for Medical Image Segmentation.

A reliability-guided fusion and dynamic dual-domain mask-modulated Shearlet network for medical image segmentation (RF-DMSNet for short) that achieves superior segmentation performance compared to state-of-the-art approaches for medical image segmentation.

Qing-Ting Jiang, Hai-Liang Ye, Rui Zhang et al. · 0 citations
2026

[Quantitative susceptibility reconstruction guided by a multi-scale T1-weighted image attention mechanism].

T1w-ADQSM outperformed QSMnet across all the 4 quantitative metrics with statistically significant differences (P<0.05), and validation based on the COSMOS reference reconstructed from multi-orientation data showed that T1w-ADQSM exhibited better structural consistency than QSMnet, and the reconstruction results were c...

Piao-Ran Li, Q. Tao, Fusheng Oyang et al. · 0 citations
Aug 2026

Machine Learning-based Enhanced Hybrid Preprocessing and Segmentation Framework for SPIR-based Liver Images

Biomedical imaging modalities such as Spectral Pre-saturation with Inversion Recovery (SPIR) MRI frequently suffer from non-Gaussian Rician noise, spatial field inhomogeneities, and weak parenchymal boundaries. This paper presents a complete, automated computer-aided diagnostic (CAD) pipeline designed for real-time edg...

T. C. M. Daniel Nixon · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.