Technology Category
Computer Vision: Medical and Biological ImagingIntelligent Robots: Multimodal Perception & Sensor FusionMachine Learning: Multimodal Learning
Multi-Stage Residual-Aware Unsupervised Deep Learning Framework for Consistent Ultrasound Strain Elastography
🤖 AI Summary
Ultrasound strain elastography (USE) faces critical challenges including strong tissue decorrelation noise, absence of ground-truth strain maps, and inconsistent strain estimation across varying deformation conditions.
Method: We propose MUSSE-Net, an unsupervised multi-stage residual-aware network featuring a multi-stream encoder-decoder architecture. It introduces a novel Tri-Cross Attention bottleneck module and a Cross-Attentive Fusion decoder, jointly optimized with a temporal consistency loss and a residual refinement mechanism for end-to-end robust strain map estimation.
Contribution/Results: Evaluated on synthetic and clinical BUET datasets, MUSSE-Net achieves an SNR of 24.54 dB and a CNR of 59.81, significantly enhancing lesion contrast while outperforming state-of-the-art methods in noise suppression and deformation robustness. The framework delivers both high quantitative accuracy and clinically interpretable strain visualization.
Technology Category
Application Category
📝 Abstract
Ultrasound Strain Elastography (USE) is a powerful non-invasive imaging technique for assessing tissue mechanical properties, offering crucial diagnostic value across diverse clinical applications. However, its clinical application remains limited by tissue decorrelation noise, scarcity of ground truth, and inconsistent strain estimation under different deformation conditions. Overcoming these barriers, we propose MUSSE-Net, a residual-aware, multi-stage unsupervised sequential deep learning framework designed for robust and consistent strain estimation. At its backbone lies our proposed USSE-Net, an end-to-end multi-stream encoder-decoder architecture that parallelly processes pre- and post-deformation RF sequences to estimate displacement fields and axial strains. The novel architecture incorporates Context-Aware Complementary Feature Fusion (CACFF)-based encoder with Tri-Cross Attention (TCA) bottleneck with a Cross-Attentive Fusion (CAF)-based sequential decoder. To ensure temporal coherence and strain stability across varying deformation levels, this architecture leverages a tailored consistency loss. Finally, with the MUSSE-Net framework, a secondary residual refinement stage further enhances accuracy and suppresses noise. Extensive validation on simulation, in vivo, and private clinical datasets from Bangladesh University of Engineering and Technology (BUET) medical center, demonstrates MUSSE-Net's outperformed existing unsupervised approaches. On MUSSE-Net achieves state-of-the-art performance with a target SNR of 24.54, background SNR of 132.76, CNR of 59.81, and elastographic SNR of 9.73 on simulation data. In particular, on the BUET dataset, MUSSE-Net produces strain maps with enhanced lesion-to-background contrast and significant noise suppression yielding clinically interpretable strain patterns.
Problem
Research questions and friction points this paper is trying to address.
Address tissue decorrelation noise in ultrasound elastography imaging
Overcome scarcity of ground truth data for strain estimation
Ensure consistent strain measurements under varying deformation conditions
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multi-stage unsupervised deep learning for strain estimation
Residual refinement stage enhances accuracy and suppresses noise
Context-aware fusion with cross-attention for feature processing
🔎 Similar Papers
No similar papers found.
