Contrast-enhanced ultrasound imaging relies on nonlinear microbubble signatures for improved visualization of perfused regions. This study aims to enhance the detection of ultrasound contrast agent perfused regions by integrating chirp-coded excitation with nonlinear signal processing. In vitro experiments were conducted in a flow phantom and ex ovo in a chicken embryo using the commercial contrast agent SonoVue. Data were acquired across a range of agent concentrations, excitation pressures, and imaging depths. Radio frequency data were processed using fundamental and subharmonic matched filtering, followed by beamforming, singular value decomposition (SVD)-based clutter suppression, and cubic Volterra filtering. Imaging performance was quantified using contrast-to-tissue ratio (CTR) and axial width measurements. The combined approach of subharmonic matched filtering, SVD, and cubic Volterra filtering resulted in substantial improvements in CTR, with gains of up to 50 dB compared to SVD-enhanced subharmonic imaging across experimental conditions. Improvements were maintained at increased imaging depths, demonstrating robustness to attenuation. Experiments in the chicken embryo model showed an enhanced visualization of perfused structures. The integration of chirp-coded subharmonic excitation with SVD and Volterra filtering improves ultrasound contrast agent imaging performance. Significance: The proposed framework offers potential for advancing contrast-enhanced ultrasound imaging, including for deep targets and low signal-to-noise conditions.

Ultrasound assessment of cervical lymph node metastasis in papillary thyroid carcinoma is crucial but suffers from inter-observer variability. Existing deep learning solutions typically use separate single-task pipelines, resulting in parameter redundancy. We propose MTL-LymphNet, a lightweight, backbone-agnostic multi-task learning framework that simultaneously segments lymph nodes and classifies metastasis within a single forward pass, evaluated across 3 backbones and 4 loss weighting strategies. On 338 images from 2 independent clinical centers, MTL-LymphNet achieved an internal cross-validation Dice coefficient of 0.891 ± 0.022 and AUC of 0.948 ± 0.027. Bidirectional external validation revealed substantial domain shift, with cross-center Dice of 0.865 and 0.779, and AUC of 0.733 and 0.830. The multi-task configuration offered a favorable efficiency-performance trade-off (∆Dice = Δ, -0.014, P < .0001) for 40% fewer parameters than separate models. No backbone or weighting strategy consistently dominated, and calibration degraded substantially under domain shift (ECE: 0.049 internally vs 0.178-0.304 externally), requiring center-specific recalibration before clinical use. To our knowledge, this is the first MTL framework for cervical lymph node analysis with bidirectional multi-center validation and systematic ablation. With a compact size of 22.6 to 29.1 MB and CPU inference of 81 to 131 ms, MTL-LymphNet demonstrates technical feasibility as a CAD tool warranting prospective validation.

Breast cancer (BC) is another leading cause of death for women, after lung cancer. The number of women who pass away from BC can be decreased with early detection. An automated method is necessary for early cancer identification because traditional BC diagnosis takes a long time. To address this issue, this study proposes a Scale-Aware Modulation Meet Deep Dilated Convolutional Network with Weighted-Leader Search Optimization (SMM-DDCNet-WLSO). To begin with, ultrasound (US) pictures are gathered for analysis from the BUSI and UDIAT datasets. After data collection, Trainable Joint Bilateral Filters (TJBF) remove speckle noise from images while preserving important anatomical borders. Next, the DarkNet-53 Convolutional Neural Network (DarkNet-53-CNN) is used for segmentation. This network efficiently captures hierarchical spatial features to differentiate between benign and malignant regions precisely. After that, the Scale-Aware Modulation Meet Deep Dilated Convolutional Neural Network (SMM-DDCNet) method is used for diagnosis. It uses both multi-scale local and global feature representations to improve diagnosis accuracy. Finally, the weighted-leader search optimization (WLSO) algorithm is used to tune the weight parameter efficiently. The populace is guided toward the best answers by the weighted impact of elite leaders in WLSO's simulation of a dynamic leadership-following approach. This harmony between exploration and exploitation greatly enhances convergence precision and stability. This model is unique in that it attains performance measures (99.70% on BUSI, 99.67% on UDIAT), accuracy (99.59% on BUSI, 99.54% on UDIAT), F1-score (99.61% on BUSI, 99.59% on UDIAT), and specificity (99.56% on BUSI, 99.55% on UDIAT), leading to very accurate BC diagnosis.

Ovarian cyst identification is a critical task in gynecological diagnostics, aiming to distinguish between benign and malignant cysts for timely and appropriate treatment. Conventional models often use hand-engineered features, but this approach fails to capture the rich variability and detailed complexity found in ultrasound and MRI images. Additionally, classical approaches are prone to lower accuracy and poor generalizability across diverse patient populations and imaging conditions. Thus, this paper introduces a Quantum-inspired Levenberg-Marquardt Network (Quantum-LMNet) for ovarian cyst detection. Initially, the preprocessing is done using the Contra-Harmonic Mean Filter. The segmentation of cyst regions is carried out with AuxSegNet, employing multiple loss functions including multi-label soft margin loss, Dice loss, and focal loss. Features are subsequently derived from the Discrete Curvelet Transform, Local Quantized Patterns (LQP), and statistical metrics. Finally, ovarian cysts are identified using the proposed Quantum-LMNet, which combines the Levenberg-Marquardt Network (LMNet), with Quantum-Inspired Convolutional Neural Networks (QuCNet). The Quantum-LMNet also showed notable enhancement across several evaluation metrics, including Accuracy, True Positive Rate (TPR), True Negative Rate (TNR), False Omission Rate (FOR), Matthews Correlation Coefficient (MCC), and Cohen's Kappa, achieving values of 96.157%, 95.720%, 97.007%, 0.099, 0.970, and 95.359%, respectively.

Accurate standard-view classification is essential for pediatric echocardiographic image analysis and downstream automated interpretation. This task remains challenging because discriminative view information is often encoded in subtle chamber configurations, outflow-tract morphology, and weak anatomical boundaries, whereas conventional classifiers may underuse shallow and intermediate representations that preserve spatial structure. We propose PVTv2-ASEF, a structure-aware framework for 4-class pediatric echocardiographic standard-view classification. The framework introduces an Adaptive Structural Enhancement Module that performs residual input-side conditioning through channel recalibration, local convolutional mixing, multi-scale structural modeling, and input-dependent branch weighting. It further employs a Dual Auxiliary Fusion Head to transform Stage 2 and Stage 3 representations into class-level evidence and fuse them with the final-stage logits during inference. PVTv2-ASEF was evaluated on a private pediatric ventricular septal defect echocardiography dataset comprising 4 standard views under repeated patient-disjoint evaluation, with macro-F1 used as the primary class-balanced metric. Compared with PVTv2-B2, the proposed framework improved macro-F1 by 0.093 on the private dataset and by 0.023 in cross-task evaluation on FETAL_PLANES_DB. These results support the utility of coordinated input-side structural enhancement and intermediate logit fusion for ultrasound view and plane classification.

To compare pykan and Lightweight KAN with MLP and Linear classifier heads for focal liver lesion classification on B-mode ultrasound, with emphasis on diagnostic sensitivity, interpretability and computational cost. A frozen ResNet18 feature extractor was combined with 4 classifier heads and evaluated on 735 annotated liver ultrasound images (Benign, Malignant, Normal) using 5-fold stratified cross-validation. Weighted cross-entropy was applied to address class imbalance. Performance was assessed using Accuracy, F1-Macro, F1-Weighted, AUROC, class-wise recall, malignant-to-normal misclassification count, inference latency, and pykan-specific spline activeness analysis. Overall metrics were similar across models, and Wilcoxon Signed-Rank testing showed no statistically significant pairwise differences (p > .05). However, clinically relevant class-level differences emerged. pykan achieved the highest Benign recall (65.5%) and the lowest malignant-to-normal error rate (5/435, 1.1%). Lightweight KAN achieved the highest Malignant recall (82.1%) and markedly faster inference (0.69 ms/image) than pykan (4183.67 ms/image). Aggregate metrics alone do not capture clinically meaningful trade-offs among KAN implementations. Lightweight KAN is attractive for rapid screening, whereas pykan shows a directional trend toward fewer malignant-to-normal errors and a more interpretable profile, though this trend did not reach statistical significance and requires confirmation in larger studies.

Breast ultrasound (BUS) images are commonly degraded by speckle noise, intensity inhomogeneity, acoustic shadowing, and low contrast, while breast tumors often exhibit irregular morphology, heterogeneous echogenicity, and blurred boundaries. These factors make accurate lesion segmentation difficult. We propose SFENet, an end-to-end spatial-frequency dual-branch image quality enhancement network for BUS tumor segmentation. Built on a U-shaped segmentation framework, SFENet jointly models lesion morphology, global spectral responses, and local image-quality degradation through a wavelet-frequency deformable spatial (WFDS) block. Specifically, the lesion-adaptive deformable scanning (LADS) module captures irregular lesion morphology and long-range spatial dependencies; the unified spectral filtering attention (USFA) module performs learnable Fourier-domain spectral recalibration; and the wavelet image enhancement module (WIEM) conducts local brightness correction and edge-aware contrast enhancement through wavelet sub-band decomposition. Because the enhancement pathway is optimized under segmentation supervision, it tends to strengthen lesion-relevant structural cues rather than generic visual quality alone. Experiments on the public BUSI data set show that SFENet achieves a Dice coefficient of 81.3% and a Jaccard index of 73.5%, outperforming the representative CNN-, Transformer-, and Mamba-based segmentation models included in this study. Qualitative results and Grad-CAM visualizations further suggest that SFENet produces more stable lesion contours in low-contrast, heterogeneous, and boundary-ambiguous cases. These results indicate that joint spatial-frequency modeling with task-driven wavelet-domain enhancement is an effective strategy for automatic BUS tumor segmentation.

This work presents the first application of the Thru-scan framework to a range of ultrasound images across frequency bands, from 3 MHz human abdominal scans to high frequency preclinical scans. The Thru-scan is a constrained reconstruction of the subresolved scatterers based on a power law distribution and enforced by appropriate selection of power law filters. The resulting reconstructions mitigate the traditional speckle "noise" pattern, with substantially increased resolution at high spatial frequencies. Improvements in contrast and apparent resolution are quantified using the generalized contrast-to-noise ratio (gCNR): Thru-scan increases gCNR by approximately 0.07 to 0.10 units, with p-values of less than 0.001. The Thru-scan processing steps also modify the amplitude distribution of the traditional B-scan, and across all frequency bands and anatomical targets the results are well modeled as a Burr distribution with power law exponents near 2.

Diffuse reverberation noise in ultrasound imaging arises from multiple reflections or secondary scattering of an echo, such as when a strong echo reflects multiple times between a fascial layer and the transducer. Unlike coherent reverberation artifacts, such as multiple reflections within an arterial wall that produce obvious ringdown in the image, these multiple reflections produce speckle-like artifacts that overlay and obscure anatomical targets and are difficult to distinguish from tissue speckle. This noise also degrades image reconstruction and interferes with imaging techniques that rely on displacement or time shifts in the ultrasonic echoes. This work introduces a permuted 2D convolutional neural network (2DCNN) for real-time reverberation suppression in ultrasound channel signals. This technique can be used to improve B-mode image quality or can be used as a pre-processing filter for techniques that require channel or beamsummed signals. The proposed architecture offers significant computational advantages over the previously implementations using 3D convolutional networks, enabling real-time implementation on existing hardware at 15 fps. The 2DCNN was trained on an enhanced dataset that combines Field II and Fullwave simulations, incorporating a broad range of acoustic affects, including reverberation noise, aberration, and attenuation. Validation on liver and kidney scans from 15 volunteers demonstrated improvements in image quality metrics related to the removal of diffuse reverberation noise, including increasing contrast, generalized contrast-to-noise ratio (GCNR), and lag-one coherence (LOC).

Breast cancer remains a significant health concern for women worldwide. Ultrasound imaging is widely adopted for screening due to its non-invasive and radiation-free nature. However, challenges such as low image contrast, blurred tumor boundaries, and diverse tumor morphologies severely hinder accurate segmentation. To address these issues, we propose LKCAU-Net, a novel hybrid multi-scale network that integrates Large Kernel Coordinate Attention (LKCA) and Atrous Spatial Pyramid Pooling (ASPP). The LKCA module enhances the model's ability to capture spatial details and global contextual information, while the ASPP module enhances multi-scale contextual representation. Embedded within a deep U-Net architecture, these components enable effective global-local feature fusion, significantly improving the robustness and accuracy of breast tumor segmentation in complex ultrasound images. We conducted experiments on four widely used public datasets: BUSI, Dataset B, BUSBRA and QAMEBI. For Dataset B, our proposed LKCAU-Net attained a Dice score of 0.8215, a Jaccard index of 0.7167, a precision of 0.8797, a recall of 0.7937, and a specificity of 0.9949. On the BUSI dataset, it reached a Dice score of 0.7973, Jaccard index of 0.6899, precision of 0.8320, recall of 0.7886, and specificity of 0.9825. When evaluated on the BUSBRA dataset, the model achieved a Dice score of 0.9085, Jaccard index of 0.8402, precision of 0.9206, recall of 0.9057, and specificity of 0.9964. On the QAMEBI, it reached a Dice score of 0.8815, Jaccard index of 0.8770, precision of 0.8993, recall of 0.8937, and specificity of 0.9949. Experimental results demonstrate that LKCAU-Net outperforms current state-of-the-art segmentation approaches, providing enhanced accuracy and robustness in breast cancer segmentation from ultrasound images.

Abdominal trauma with bleeding is a leading cause of post-traumatic death, and detecting free fluid in the abdomen or hemoperitoneum can provide critical guidance for clinical management. Rapid and accurate diagnosis of abdominal bleeding using ultrasound is significant for making decisions regarding the need for surgical intervention. This study introduces a multi-task network for the segmentation and classification of ascites in ultrasound images. The network utilizes a U-Net backbone with a ResNext encoder as the basic architecture for the segmentation and classification models. The segmentation network includes a Frequency Channel Attention (FCA) attention module, which effectively broadens the range of captured information and enhances the robustness of channel representation. Furthermore, an Enhanced Channel Attention Multi Feature Fusion (EMFF) was used to extract the interdependencies between feature channels by combining high-order and low-order feature mappings, thereby improving segmentation accuracy. Lastly, a classification branch was created to classify ascites by sharing encoder features. Experiments on the collected ascites ultrasound dataset demonstrated that the proposed method achieved a segmentation Dice of 85.28% and a classification accuracy of 86.18%. It outperformed the leading multi-task SOTA method by 0.7% in Dice and 2.03% in accuracy, establishing a new benchmark for simultaneous ascites assessment. This study showed that the proposed network is valuable for the preliminary diagnosis of ascites in ultrasound and can serve as a potential auxiliary tool for clinical ascites examination in emergency situations.

Ultrasound imaging is widely used in clinical practice due to its non-invasive, real-time, and cost-effective nature. However, speckle noise often degrades image quality, obscuring fine anatomical structures and reducing diagnostic confidence. Existing denoising methods struggle to remove noise effectively while preserving critical details, limiting their clinical utility. Although recent deep learning architectures excel at capturing both local details and global structure, they remain inherently limited in handling speckle noise, as its physical characteristics are not explicitly incorporated. To address this limitation, a Physics-Regularized Self-Supervised Denoising U-Net (PR-SSD-Net) is introduced to reinforce the U-Net's capability for high-quality image restoration. The physics-based constraint guides the network to produce residual noise patterns that align with expected statistical behavior, enhancing image clarity and preserving critical structures. Comprehensive evaluations were conducted on six diverse ultrasound datasets. Significant improvements in Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) were observed, accompanied by reduced variability, as reflected in their standard deviations (SD). An ablation study confirmed the pivotal role of physics-guided regularization, and expert assessments demonstrated high inter-rater agreement (Fleiss), supporting the clinical relevance of the approach. These results highlight the proposed PR-SSD-Net approach as a robust, physically grounded solution for speckle noise reduction, enhancing both the reliability and clinical utility of ultrasound imaging.

This study evaluated the utility of super-resolution ultrasound contrast imaging for differentiating benign from malignant breast tumors. In this retrospective study, 60 patients underwent contrast-enhanced ultrasound acquisition before biopsy or surgery excision at The Third Affiliated Hospital of Nanjing Medical University between March and November 2025. Super-resolution contrast-enhanced ultrasound (SRUS) reconstruction and quantitative analysis were then performed from contrast-enhanced ultrasound cine loops using microbubble localization and tracking software. Using histopathology as the reference standard, patients were classified into malignant and benign groups. The discriminatory value of each parameter for benign versus malignant breast tumors was assessed. Diagnostic performance was quantified using the area under the receiver operating characteristic curve (AUC). The diagnostic performance of each SRUS-derived parameter was evaluated for differentiating benign from malignant breast tumors. On final pathology, SRUS-derived parameters showed good performance for differentiating benign from malignant breast tumors. Compared with benign lesions, malignant tumors exhibited significantly higher FD, number of branching points per area maximum curvature, mean curvature, maximum vessel velocity, mean vessel velocity, maximum diameter, mean diameter, and microvasculature density (all p < .050). Among all parameters, maximum curvature achieved the highest AUC (0.933), with 88.0% sensitivity and 80.0% specificity. Intra- and inter-operator reliability were high, with intraclass correlation coefficients >.85. SRUS-derived quantitative parameters showed diagnostic value for differentiating benign from malignant breast masses and enable clear visualization of hemodynamic alterations, suggesting potential value for preoperative malignant-risk assessment.

The intima media thickness (IMT) of the common carotid artery (CCA) is helpful for the diagnosis of atherosclerosis. The CCA ultrasound image segmentation methods can be used for the IMT measurements. However, the speckle noise and the blurred edges can reduce the accuracy of the IMT measurements. To address this issue, a preprocessing method based on non-subsampled shearlet transform (NSST) and Phase Asymmetry Metric is proposed for the CCA ultrasound images. First, the CCA ultrasound image is decomposed into the NSST domain with a low-frequency sub-band and several high-frequency sub-bands. Next, the high frequency sub-band image is despeckling using an adaptive threshold method. Then, the processed image is reconstructed through NSST. Finally, an edge enhancement method based on the phase asymmetry (PAS) metric is used for enhancing the edges in the reconstructed CCA ultrasound images. In the simulated CCA ultrasound image experiments, the CNR value of proposed method is 1.8191 dB, which is the highest among four state-of-the-art methods. In the clinical CCA ultrasound image experiments, the average CNR values of proposed method is 1.5457 ± 0.6595 (mean ± std in dB), which is the best among four advanced methods. In the real application of IMT measurements, the average absolute error of intima-media thickness measurements of proposed method is 0.0790 ± 0.0588 (mean ± std in mm), which is the lowest among four advanced methods. In conclusion, the proposed approach has improved performance in suppressing the speckle noise and enhancing the intima media complex (IMC) regions, and it is a potential method for preprocessing of the CCA ultrasound images in clinical application.

Poor spatial resolution and contrast remain major challenges in ultrafast ultrasound imaging. Null subtraction imaging (NSI) improves lateral resolution but often degrades speckle quality and contrast. Its extension, dynamic DC-biased NSI (dDC-NSI), mitigates this trade-off by introducing a dynamic DC bias; however, slight speckle suppression may still occur in homogeneous regions, leading to dark-region artifacts. In this work, a generalized null subtraction factor (gNSF) is proposed as a post-processing framework. gNSF applies multiple apodizations, followed by a mirror-flipping and symmetric summation operation, and defines a weighting factor based on the energy ratio between a bias term and a zero-mean sequence. By incorporating the dynamic DC bias, coherent echoes are enhanced while incoherent noise is suppressed. Phantom experiments show that gNSF achieves a contrast performance (gCNR close to 1) comparable to GCF and dDC-NSI, and superior to DAS and NSI. In addition, gNSF improves CR and sSNR by 20% and 21% compared with dDC-NSI, indicating reduced speckle over-suppression and a better balance between contrast and speckle preservation.