IEEE transactions on ultrasonics, ferroelectrics, and frequency control最新文献_第3页

Adaptive color Doppler for axial velocity imaging of microvessel networks. 微血管网络轴向速度自适应彩色多普勒成像。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-09 DOI: 10.1109/TUFFC.2025.3559238

Yang Liu, Yongchao Wang, Pakpong Chirarattananon, Jianbo Tang

{"title":"Adaptive color Doppler for axial velocity imaging of microvessel networks.","authors":"Yang Liu, Yongchao Wang, Pakpong Chirarattananon, Jianbo Tang","doi":"10.1109/TUFFC.2025.3559238","DOIUrl":"https://doi.org/10.1109/TUFFC.2025.3559238","url":null,"abstract":"Directional filtering has been applied to distinguish between the ascending and descending flows in functional ultrasound imaging, however, it can lead to incorrect measurement of the flow speed and direction when using the directional filtering-based color Doppler ultrasound velocimetry (iCD_US). Specifically, in cases where the frequency spectrum bandwidth of a unidirectional flow extends into both negative and positive frequency domains, directional filtering may erroneously produce bidirectional velocities. Here, we propose an adaptive color Doppler ultrasound technique (aCD_US), which addresses this issue by analyzing the envelope of the Doppler spectrum and then adaptively using the whole spectrum integration or directional filtering-based approach to estimate the flow velocity. The proposed aCD_US was validated through numerical simulations and phantom experiments under various flow conditions, demonstrating superior performance in estimating axial velocities of unidirectional, bidirectional, and horizontal flows. Notably, numerical simulations showed that aCD_US achieved over 90% directional accuracy and less than 15% velocity deviation at signal-to-noise ratios larger than -1 dB. In vivo experiments on mouse cerebral blood flow further highlighted its advantages, with aCD_US surpassing conventional color Doppler velocimetry and iCD_US in reconstructing axial flow velocity maps. The quantitative comparison between aCD_US and vULM shows a strong overall correlation in their axial velocity measurements, with a Pearson correlation coefficient of 0.760 (p = 0.000). These results demonstrate the advantage of aCD_US in precise microvessel networks velocity quantification and its potential to advance microvascular imaging accuracy in both research and clinical applications.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143994327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Respiratory motion effects and mitigation strategies on boiling histotripsy in porcine liver and kidney. 猪肝脏和肾脏沸腾组织的呼吸运动效应及缓解策略。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-09 DOI: 10.1109/TUFFC.2025.3559458

Ekaterina Ponomarchuk, Gilles Thomas, Minho Song, Yak-Nam Wang, Stephanie Totten, George Schade, Vera Khokhlova, Tatiana Khokhlova

{"title":"Respiratory motion effects and mitigation strategies on boiling histotripsy in porcine liver and kidney.","authors":"Ekaterina Ponomarchuk, Gilles Thomas, Minho Song, Yak-Nam Wang, Stephanie Totten, George Schade, Vera Khokhlova, Tatiana Khokhlova","doi":"10.1109/TUFFC.2025.3559458","DOIUrl":"https://doi.org/10.1109/TUFFC.2025.3559458","url":null,"abstract":"Boiling histotripsy (BH) is a pulsed high-intensity focused ultrasound (HIFU)-based method of extracorporeal non-thermal tissue disintegration under real-time ultrasound (US) guidance. Respiratory motion in abdominal targets can affect BH precision and completeness. This study compares two motion mitigation strategies based on pulse/echo US motion tracking: robotic arm-based unidirectional motion compensation by HIFU transducer manipulation and BH pulse gating during expiratory pause. BH ablations were generated in liver and kidney of anesthetized pigs with 2-10ms pulses using 256-element 1.5-MHz HIFU array. A coaxial US imaging probe was used for targeting, tracking skin surface and monitoring real-time bubble activity. The axial (anterior-posterior, AP) displacement of the skin surface was found to be synchronous with liver and kidney motion in both cranio-caudal (CC) and AP directions. BH lesions were produced either with no motion mitigation, or with pulse gating, or with 1D motion compensation. Dimensions of completely fractionated and affected tissue areas were measured histologically. In liver, gating and motion compensation improved fractionation completeness within targeted volumes and reduced off-target tissue damage in AP direction vs no motion mitigation; only gating reduced off-target damage in CC direction. In kidney, gating improved BH completeness in both directions vs no mitigation, but did not affect off-target damage due to lower displacement amplitudes in kidney comparable with gating tolerance limits. In both liver and kidney, gating increased treatment time by 24%. These results suggest that BH pulse gating using US-based AP skin surface tracking is an adequate approach for treating organs with pronounced 3D respiratory motion.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143963274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Contrast-Enhanced Null Subtraction Imaging Method using Dynamic DC Bias in Ultrafast Ultrasound imaging. 一种利用动态直流偏置进行超快超声成像的对比度增强零差成像方法。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-04 DOI: 10.1109/TUFFC.2025.3558017

Xin Yan, Xiaodong Yang, Lingling Jing, Wei Guo, Yingqi Wang, Xinwei Su, Yuanyuan Wang

{"title":"A Contrast-Enhanced Null Subtraction Imaging Method using Dynamic DC Bias in Ultrafast Ultrasound imaging.","authors":"Xin Yan, Xiaodong Yang, Lingling Jing, Wei Guo, Yingqi Wang, Xinwei Su, Yuanyuan Wang","doi":"10.1109/TUFFC.2025.3558017","DOIUrl":"10.1109/TUFFC.2025.3558017","url":null,"abstract":"Enhancing the resolution and contrast of ultrafast ultrasound imaging is imperative for the accuracy of clinical diagnostics. Null subtraction imaging (NSI) is a nonlinear beamforming technique capable of significantly enhancing lateral resolution. However, it suffers from issues of low-quality speckle pattern and poor contrast performance. To address this issue, we propose a novel contrast-enhanced NSI method that utilizes dynamic DC bias. Innovatively, we construct the dynamic DC bias using a generalized coherence factor (GCF) and a sigmoid transformation function that adapts the DC value based on the signal characteristics of different imaging regions. Furthermore, a normalization scheme is proposed to optimize the beamforming output, ensuring uniform pixel intensity throughout the final image. Simulation, phantom, and in vivo data are utilized for ultrasound beamforming to evaluate the performance of the proposed method. Quantitative results show that the proposed method significantly enhances the contrast ratio (CR) by 197%, the contrast-to-noise ratio (CNR) by 341%, the speckle signal-to-noise ratio (sSNR) by 302%, and the generalized contrast-to-noise ratio (gCNR) by 106% compared to the original NSI (in phantom). Point target imaging results indicate that the proposed method achieves a main lobe width slightly wider than the original NSI method, but much narrower than those of DAS and GCF. These findings confirm that the proposed method significantly enhances imaging contrast while preserving high resolution, which is of great significance for the further clinical application of ultrafast ultrasound imaging.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143784458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Computational Ultrasound Carotid Artery Imaging with a Few Transceivers: An Emulation Study. 计算超声颈动脉成像与几个收发器：仿真研究。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-04 DOI: 10.1109/TUFFC.2025.3557374

Yuyang Hu, Didem Dogan, Michael Brown, Geert Leus, Antonius F W van der Steen, Pieter Kruizinga, Johannes G Bosch

{"title":"Computational Ultrasound Carotid Artery Imaging with a Few Transceivers: An Emulation Study.","authors":"Yuyang Hu, Didem Dogan, Michael Brown, Geert Leus, Antonius F W van der Steen, Pieter Kruizinga, Johannes G Bosch","doi":"10.1109/TUFFC.2025.3557374","DOIUrl":"10.1109/TUFFC.2025.3557374","url":null,"abstract":"Ultrasonography could allow operator-independent examination and continuous monitoring of the carotid artery, but normally requires complex and expensive transducers, especially for 3D. By employing computational ultrasound imaging (cUSi), using an aberration mask and model-based reconstruction, a monitoring device could be constructed with a more affordable simple transducer design comprising only a few elements. We aim to apply the cUSi concept to create a carotid artery monitoring system. The system's possible configurations for the 2D imaging case were explored using a linear array setup emulating a cUSi device in silico, followed by in-vitro testing and in-vivo carotid artery imaging. Our study shows enhanced reconstruction performance with the use of an aberrating mask, improved lateral resolution through proper choice of the mask delay variation, and more accurate reconstructions using least-squares with QR decomposition (LSQR) compared to matched filtering. Together, these advancements enable B-mode reconstruction and power Doppler imaging of the carotid artery with sufficient quality for monitoring using a configuration of 12 transceivers coupled with a random aberration mask with a maximum delay variation of 4 wave periods.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143784461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Chirp-Coded Subharmonic Imaging With Volterra Filtering: Histotripsy Bubble Cloud Assessment In Vitro and Ex Vivo 沃尔泰拉滤波的啁啾编码次谐波成像：体外和离体组织泡沫云评估。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-03 DOI: 10.1109/TUFFC.2025.3556030

Vishwas V. Trivedi;Katia Flores Basterrechea;Kenneth B. Bader;Himanshu Shekhar

{"title":"Chirp-Coded Subharmonic Imaging With Volterra Filtering: Histotripsy Bubble Cloud Assessment In Vitro and Ex Vivo","authors":"Vishwas V. Trivedi;Katia Flores Basterrechea;Kenneth B. Bader;Himanshu Shekhar","doi":"10.1109/TUFFC.2025.3556030","DOIUrl":"10.1109/TUFFC.2025.3556030","url":null,"abstract":"Histotripsy is a noninvasive focused ultrasound therapy that liquifies tissue via bubble activity. Conventional ultrasound imaging is used in current clinical practice to monitor histotripsy. Developing surrogate imaging metrics for successful treatment outcomes remains an unmet clinical need. The goal of this work was twofold. First, we investigated whether histotripsy bubble clouds detected with nonlinear imaging (chirp-coded subharmonic imaging with and without Volterra filtering) could be used to assess the ablation zone in vitro. Second, we evaluated the feasibility of improving bubble cloud contrast with this approach in ex vivo porcine kidney. Histotripsy bubble clouds were generated in red blood cell-doped agarose phantoms and imaged with a curvilinear ultrasound probe. The ablation zone was assessed based on images collected with a digital camera. The relationship between the bubble cloud area and the ablation area was assessed using receiver operating characteristic (ROC) analysis, F1 score, accuracy, and Matthews correlation coefficient. Histotripsy bubble clouds were also generated in ex vivo porcine tissue and the ability to improve bubble cloud contrast to tissue was evaluated. Implementing chirp-coded subharmonic imaging with the third-order Volterra filter enhanced contrast-to-tissue ratio (CTR) by up to <inline-formula> <tex-math>$40.06~pm ~0.70$ </tex-math></inline-formula> dB relative to standard imaging in vitro. Furthermore, subharmonic imaging combined with Volterra filtering estimated bubble cloud areas that best matched the ablation zone area based on the analysis metrics. Furthermore, ex vivo studies showed CTR improvement of up to <inline-formula> <tex-math>$26.95~pm ~6.49$ </tex-math></inline-formula> dB. Taken together, these findings advance image guidance and monitoring approaches for histotripsy.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 5","pages":"591-600"},"PeriodicalIF":3.0,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143779804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Directional Coherence Factor for Volumetric Ultrasound Imaging with Matrix Arrays. 矩阵阵列体积超声成像的定向相干系数。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-04-03 DOI: 10.1109/TUFFC.2025.3557519

Xiaochuan Wu, Wei-Ning Lee

{"title":"Directional Coherence Factor for Volumetric Ultrasound Imaging with Matrix Arrays.","authors":"Xiaochuan Wu, Wei-Ning Lee","doi":"10.1109/TUFFC.2025.3557519","DOIUrl":"10.1109/TUFFC.2025.3557519","url":null,"abstract":"Matrix arrays with small apertures limit spatial and contrast resolutions of volumetric ultrasound imaging. Coherence-based beamformers are prevalent for side lobe suppression and resolution improvement. While the spatial coherence of a matrix array is fundamentally a 2D function, conventional coherence factor (CF) methods neglect the directional variation of an M × N matrix array when calculating volumetric coherence. We hereby propose a projectionbased directional coherence factor (DCF) to exploit the 2D nature of volumetric coherence function. Instead of computing the coherent and incoherent summations across the entire 2D aperture, DCF projects aperture data onto azimuthal, elevational, diagonal, and antidiagonal directions and subsequently calculates the coherence factors for each direction separately. The orthogonal coherence pairs, i.e., azimuth and elevation, diagonal and anti-diagonal, are multiplied to obtain DCFRC and DCFDiag, respectively. The Jaccard similarity of the DCFRC and DCFDiag is used to derive the final DCF to weigh the reconstructed images. We evaluated the performance of DCF beamforming in point-target simulations, multi-purpose phantom experiments, and in vivo muscle imaging, and compared it to delay-and-sum (DAS) and CF beamformers. Our DCF achieved side lobe reduction throughout the entire volume compared to conventional CF. Moreover, diagonal weighting significantly improved, on average, the azimuthal resolution by 41.3% vs. DAS and 7.35% vs. CF as well as the elevational resolution by 40.4% vs. DAS and 38.7% vs. CF. Our proposed DCF offers a practical solution for resolution and contrast enhancement of volumetric imaging in 2D matrix array configurations.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143779808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Study of Frequency Trimming Ability and Performance Enhancement of Thin-Film Piezoelectric-on-Silicon MEMS Resonators by Joule Heating via Localized Annealing 局域退火焦耳加热薄膜压电MEMS谐振器的频率微调能力及性能增强研究。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-03-31 DOI: 10.1109/TUFFC.2025.3556305

Ugur Guneroglu;Adnan Zaman;Abdulrahman Alsolami;Ivan F. Rivera;Jing Wang

{"title":"Study of Frequency Trimming Ability and Performance Enhancement of Thin-Film Piezoelectric-on-Silicon MEMS Resonators by Joule Heating via Localized Annealing","authors":"Ugur Guneroglu;Adnan Zaman;Abdulrahman Alsolami;Ivan F. Rivera;Jing Wang","doi":"10.1109/TUFFC.2025.3556305","DOIUrl":"10.1109/TUFFC.2025.3556305","url":null,"abstract":"This article deliberately explores the frequency trimming and performance enhancement of piezoelectric MEMS resonators through localized annealing induced by Joule heating. Targeting the effective postfabrication treatment of thin-film piezoelectric-on-silicon (TPoS) resonators, we employ a novel annealing approach that modifies the silicon resonator body-bottom electrode interface to enable meticulous resonance frequency trimming and enhanced overall performance. By applying a controlled dc current directly through the resonator’s body, precise resonance frequency shifts on the order of 0.1%–0.4% and significant increase in quality factor, from 981 to 2155, from 8214 to 9362, have been realized for rectangular-plate and disk-shaped resonators, respectively. Furthermore, this localized annealing process reduces the motional impedance from 3.43 to 1.65 k<inline-formula> <tex-math>$Omega $ </tex-math></inline-formula> for a rectangular-plate resonator and from 1.79 to 1.58 k<inline-formula> <tex-math>$Omega $ </tex-math></inline-formula> for a disk-shaped resonator, thus demonstrating its viability as a postfabrication treatment technique for a wide variety of MEMS devices. These results highlight the great potential of Joule heating-induced localized annealing in advancing RF systems that demand high precision, reliable filtering, and stable timing functions. This work provides new insights into the thermal annealing effects on MEMS resonators and lays a foundation for future innovations in related microsystem technologies.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 5","pages":"686-696"},"PeriodicalIF":3.0,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143763606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Synthetic Lung Ultrasound Data Generation Using Autoencoder With Generative Adversarial Network 基于生成对抗网络的自编码器合成肺超声数据生成。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-03-27 DOI: 10.1109/TUFFC.2025.3555447

Noreen Fatima;Federico Mento;Sajjad Afrakhteh;Tiziano Perrone;Andrea Smargiassi;Riccardo Inchingolo;Libertario Demi

{"title":"Synthetic Lung Ultrasound Data Generation Using Autoencoder With Generative Adversarial Network","authors":"Noreen Fatima;Federico Mento;Sajjad Afrakhteh;Tiziano Perrone;Andrea Smargiassi;Riccardo Inchingolo;Libertario Demi","doi":"10.1109/TUFFC.2025.3555447","DOIUrl":"10.1109/TUFFC.2025.3555447","url":null,"abstract":"Class imbalance is a significant challenge in medical image analysis, particularly in lung ultrasound (LUS), where severe patterns are often underrepresented. Traditional oversampling techniques, which simply duplicate original data, have limited effectiveness in addressing this issue. To overcome these limitations, this study introduces a novel supervised autoencoder generative adversarial network (SA-GAN) for data augmentation, leveraging advanced generative artificial intelligence (AI) to create high-quality synthetic samples for minority classes. In addition, the traditional data augmentation technique is used for comparison. The SA-GAN incorporates an autoencoder to develop a conditional latent space, effectively addressing weight clipping issues and ensuring higher quality synthetic data. The generated samples are evaluated using similarity metrics and expert analysis to validate their utility. Furthermore, state-of-the-art neural networks are used for multiclass classification, and their performance is compared when trained with GAN-based augmentation versus traditional data augmentation techniques. These contributions enhance the robustness and reliability of AI models in mitigating class imbalance in LUS analysis.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 5","pages":"624-635"},"PeriodicalIF":3.0,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10943232","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143730019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images 无监督测试时间适应肝脂肪变性分级使用超声b型图像。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-03-26 DOI: 10.1109/TUFFC.2025.3555180

Pedro Vianna;Paria Mehrbod;Muawiz Chaudhary;Michael Eickenberg;Guy Wolf;Eugene Belilovsky;An Tang;Guy Cloutier

{"title":"Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images","authors":"Pedro Vianna;Paria Mehrbod;Muawiz Chaudhary;Michael Eickenberg;Guy Wolf;Eugene Belilovsky;An Tang;Guy Cloutier","doi":"10.1109/TUFFC.2025.3555180","DOIUrl":"10.1109/TUFFC.2025.3555180","url":null,"abstract":"Ultrasound (US) is considered a key modality for the clinical assessment of hepatic steatosis (i.e., fatty liver) due to its noninvasiveness and availability. Deep learning methods have attracted considerable interest in this field, as they are capable of learning patterns in a collection of images and achieve clinically comparable levels of accuracy in steatosis grading. However, variations in patient populations, acquisition protocols, equipment, and operator expertise across clinical sites can introduce domain shifts that reduce model performance when applied outside the original training setting. In response, unsupervised domain adaptation techniques are being investigated to address these shifts, allowing models to generalize more effectively across diverse clinical environments. In this work, we propose a test-time batch normalization (TTN) technique designed to handle domain shift, especially for changes in label distribution, by adapting selected features of batch normalization (BatchNorm) layers in a trained convolutional neural network model. This approach operates in an unsupervised manner, allowing robust adaptation to new distributions without access to label data. The method was evaluated on two abdominal US datasets collected at different institutions, assessing its capability in mitigating domain shift for hepatic steatosis classification. The proposed method reduced the mean absolute error in steatosis grading by 37% and improved the area under the receiver operating characteristic curves (AUC) for steatosis detection from 0.78 to 0.97, compared to nonadapted models. These findings demonstrate the potential of the proposed method to address domain shift in US-based hepatic steatosis diagnosis, minimizing risks associated with deploying trained models in various clinical settings.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 5","pages":"601-611"},"PeriodicalIF":3.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143730045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

BentRay-NeRF: Bent-Ray Neural Radiance Fields for Robust Speed-of-Sound Imaging in Ultrasound Computed Tomography BentRay-NeRF：用于超声计算机断层扫描中稳健声速成像的弯曲射线神经辐射场。

IF 3 2区工程技术

IEEE transactions on ultrasonics, ferroelectrics, and frequency control Pub Date : 2025-03-26 DOI: 10.1109/TUFFC.2025.3554223

Shilong Cui;Qing Wu;Yiming Huang;Haizhao Dai;Yuyao Zhang;Jingyi Yu;Xiran Cai

{"title":"BentRay-NeRF: Bent-Ray Neural Radiance Fields for Robust Speed-of-Sound Imaging in Ultrasound Computed Tomography","authors":"Shilong Cui;Qing Wu;Yiming Huang;Haizhao Dai;Yuyao Zhang;Jingyi Yu;Xiran Cai","doi":"10.1109/TUFFC.2025.3554223","DOIUrl":"10.1109/TUFFC.2025.3554223","url":null,"abstract":"Ultrasound computed tomography (USCT) is a promising technique for breast cancer detection because of its quantitative imaging capability of the speed of sound (SOS) of soft tissues and the fact that malignant breast lesions often have a higher SOS compared to healthy tissues in the human breast. Compared to waveform inversion-based USCT, bent-ray tracing USCT is relatively less computationally expensive, which particularly suits breast cancer screening in a large population. However, SOS image reconstruction using bent-ray tracing in USCT is a highly ill-conditioned problem, making it susceptible to measurement errors. This presents significant challenges in achieving stable and high-quality reconstructions. In this study, we show that using implicit neural representation (INR), the ill-conditioned problem can be well mitigated, and stable reconstruction is achievable. This INR approach uses a multilayer perceptron (MLP) with hash encoding to model the slowness map as a continuous function, to better regularize the inverse problem and has been shown more effective than classical approaches of solely adding regularization terms in the loss function. Thereby, we propose the bent-ray neural radiance fields (BentRay-NeRF) method for SOS image reconstruction to address the aforementioned challenges in classical SOS image reconstruction methods, such as the Gauss-Newton method. In silico and in vitro experiments showed that BentRay-NeRF has remarkably improved performance compared to the classical method in many aspects, including the image quality and the robustness of the inversion to different acquisition settings in the presence of measurement errors.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 5","pages":"612-623"},"PeriodicalIF":3.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143730016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0