Differential Phase Contrast Image Fusion via Wavelet Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical X-ray imaging techniques, such as mammography, face challenges in effectively fusing absorption, differential phase contrast, and scattering data into a single, informative image that radiologists can easily interpret, as existing fusion methods do not align with conventional gray-level images and can introduce noise or reduce resolution.
Innovation Solution
A method and system for image fusion using a multiple-resolution framework that decomposes images into sub-bands, applies intra- and inter-band processing to enhance signal-to-noise ratio and selectively weight contributions from absorption, differential phase contrast, and scattering data, ensuring the fused image maintains resolution and noise levels similar to conventional mammograms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If RGB or IHS color-coded image fusion method is used to merge AC, DPC, and DFC signals, then the fusion is efficient and natural for human vision, but it does not comply with conventional gray-level images in mammography and high-resolution clinical monitors
Solution Approach 1:
The patent transforms the color-coded fusion output into grayscale by applying a luminance weighting function Y = 0.299R + 0.587G + 0.114B to the RGB fused image, converting it to a format compatible with conventional gray-level clinical displays while preserving the enhanced diagnostic information from DPC and DFC signals
2Measurement precision
If image fusion is performed to enhance detail features from DPC and DFC signals, then diagnostic information is improved, but noise level may increase and resolution may be reduced
Solution Approach 1:
The patent decomposes the input images into multiple frequency sub-bands using wavelet transform, separating low-frequency components (from AC image) and high-frequency components (from DPC and DFC images) to selectively fuse only the beneficial detail information while avoiding noise amplification
Solution Approach 2:
The patent applies spatially adaptive weighting in the wavelet domain, where different weights are assigned to different frequency sub-bands and spatial locations based on local image characteristics, allowing enhancement of true diagnostic features while suppressing noise in specific regions
3Productivity
If multiple signals are fused into one image to reduce diagnosis time and complexity, then radiologist workflow is improved, but the fusion algorithm complexity increases
Solution Approach 1:
The patent introduces the wavelet transform domain as an intermediary representation space where the fusion of AC, DPC, and DFC signals is performed more efficiently than in the spatial domain, enabling automatic multi-parameter fusion without requiring complex manual intervention while maintaining computational feasibility
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The proposed method successfully merges absorption, differential phase contrast, and scattering data into a single gray-level image with enhanced detail features, improving diagnostic accuracy by providing intuitive clinical information while maintaining noise levels and resolution, thus facilitating more accurate diagnoses.
Implementation Method 1
The decomposition therefore emphasizes the contribution of the DPC and the DFC signals to the fused image since the most interesting contributions of the DPC and DFC signals materializes rather in the high frequency domain than the contribution of AC signal which contributes rather in the low frequency domain
Data Source
AI summary
The latest progresses in breast imaging using differential phase contrast techniques pose the question of how to fuse multiple information sources, yielded by absorption, differential phase, and scattering signals, into a single, informative image for clinical diagnosis. It is proposed to use an image fusion scheme based on a multiple-resolution framework. The three signals are first transformed into multiple bands presenting information at different frequencies and then a two-step processing follows: (1) intra-band processing enhances the local signal-to-noise ratio using a novel noise estimation method and context modeling; and (2) inter-band processing weights each band by considering their characteristics and contributions, and suppressing the global noise level. The fused image, looking similar to a conventional mammogram but with significantly enhanced detail features, is reconstructed by inverse transform. The fused image is compatible with clinical settings and enables the radiologists to use their years of diagnosis experiences in mammography.