Pub Date : 2025-09-16DOI: 10.1109/TUFFC.2025.3610361
Keith A. Wear;Christopher R. Fury;Andre V. Alvarenga
Hydrophone spatial resolution and spatial averaging effects are determined by the frequencydependent effective sensitive element diameter deff(f) rather than the geometrical sensitive element diameter dg. The objective of this work was to quantify average deff(f) for needle hydrophones as a function of dg and f. Estimates of effective radii aeff(f) = deff(f)/2 were inferred from directivity measurements from 0.5 to 20 MHz on 16 needle hydrophones with dg = 2ag ranging from 75 to 1000 μm (139 hydrophone/frequency combinations). Effective sensitive element diameter deff(f) exceeded dg by over 100% when λ > 4dg (where λ is the wavelength). For kag > 0.75 (where k = 2π/λ), deff(f) was consistent with the “rigid piston” (RP) theory, reinforcing a previous report from our laboratories. However, for kag < 0.75, deff(f) showed noticeable deviations from RP theory and fell between predictions from RP theory and predictions for an unbaffled (UB) circular piston. Examples: 1) for a needle hydrophone with dg = 75 μm at 1 MHz (kag = 0.16), the data imply that average deff = 505 μm, and 2) for a needle hydrophone with dg = 400 μm at 500 kHz (common parameters for human transcranial neuromodulation; kag = 0.42), the data imply that average deff = 1215 μm.
{"title":"Spatial Resolution Limits for Needle Hydrophones From 0.5 to 20 MHz With Implications for Transcranial Ultrasound","authors":"Keith A. Wear;Christopher R. Fury;Andre V. Alvarenga","doi":"10.1109/TUFFC.2025.3610361","DOIUrl":"10.1109/TUFFC.2025.3610361","url":null,"abstract":"Hydrophone spatial resolution and spatial averaging effects are determined by the frequencydependent effective sensitive element diameter deff(f) rather than the geometrical sensitive element diameter dg. The objective of this work was to quantify average deff(f) for needle hydrophones as a function of dg and f. Estimates of effective radii aeff(f) = deff(f)/2 were inferred from directivity measurements from 0.5 to 20 MHz on 16 needle hydrophones with dg = 2ag ranging from 75 to 1000 μm (139 hydrophone/frequency combinations). Effective sensitive element diameter deff(f) exceeded dg by over 100% when λ > 4dg (where λ is the wavelength). For kag > 0.75 (where k = 2π/λ), deff(f) was consistent with the “rigid piston” (RP) theory, reinforcing a previous report from our laboratories. However, for kag < 0.75, deff(f) showed noticeable deviations from RP theory and fell between predictions from RP theory and predictions for an unbaffled (UB) circular piston. Examples: 1) for a needle hydrophone with dg = 75 μm at 1 MHz (kag = 0.16), the data imply that average deff = 505 μm, and 2) for a needle hydrophone with dg = 400 μm at 500 kHz (common parameters for human transcranial neuromodulation; kag = 0.42), the data imply that average deff = 1215 μm.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 11","pages":"1489-1496"},"PeriodicalIF":3.7,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11165490","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145075093","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}
Pub Date : 2025-09-16DOI: 10.1109/TUFFC.2025.3609832
Darcy M. Dunn-Lawless;Abigail C. Collins;Constantin C. Coussios;Michael D. Gray
Passive acoustic mapping (PAM) is a powerful and widely used method of imaging cavitation activity. However, the presence of a container around a cavitating sample in experiments performed in vitro can introduce significant aberrations into recorded cavitation noise and resulting PAM images. These artifacts may lead to energy being incorrectly estimated or mapped to the wrong place, preventing accurate correlation between cavitation and bioeffects. In this work, we quantify these acoustic effects for six common types of sample containers using an acoustic reciprocity experiment, then use the results to inform the design of a new container with improved acoustic transparency. Existing vessels were found to introduce up to 13-dB broadband insertion loss and change the location and spread of energy in PAM images by up to 1 mm and 25%, respectively. The new container caused up to 1.4-dB insertion loss (the lowest of any container tested) and introduced no significant phase aberration, source location error, or change in energy spread to the PAM images. Testing the new container with real cavitation noise produced very similar insertion loss figures of up to 1.6 dB. These results highlight deficiencies in existing sample containers for the purposes of quantifying cavitation activity with PAM, which is increasingly desired as cavitation matures as a therapy. The guidelines for acoustic transparency developed here may assist researchers in avoiding container aberrations and enable accurate measurement of cavitation energy in future studies.
{"title":"Acoustically Transparent Sample Containers for Quantitative Cavitation Imaging","authors":"Darcy M. Dunn-Lawless;Abigail C. Collins;Constantin C. Coussios;Michael D. Gray","doi":"10.1109/TUFFC.2025.3609832","DOIUrl":"10.1109/TUFFC.2025.3609832","url":null,"abstract":"Passive acoustic mapping (PAM) is a powerful and widely used method of imaging cavitation activity. However, the presence of a container around a cavitating sample in experiments performed in vitro can introduce significant aberrations into recorded cavitation noise and resulting PAM images. These artifacts may lead to energy being incorrectly estimated or mapped to the wrong place, preventing accurate correlation between cavitation and bioeffects. In this work, we quantify these acoustic effects for six common types of sample containers using an acoustic reciprocity experiment, then use the results to inform the design of a new container with improved acoustic transparency. Existing vessels were found to introduce up to 13-dB broadband insertion loss and change the location and spread of energy in PAM images by up to 1 mm and 25%, respectively. The new container caused up to 1.4-dB insertion loss (the lowest of any container tested) and introduced no significant phase aberration, source location error, or change in energy spread to the PAM images. Testing the new container with real cavitation noise produced very similar insertion loss figures of up to 1.6 dB. These results highlight deficiencies in existing sample containers for the purposes of quantifying cavitation activity with PAM, which is increasingly desired as cavitation matures as a therapy. The guidelines for acoustic transparency developed here may assist researchers in avoiding container aberrations and enable accurate measurement of cavitation energy in future studies.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 11","pages":"1476-1488"},"PeriodicalIF":3.7,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11164665","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145075107","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}
Pub Date : 2025-09-02DOI: 10.1109/TUFFC.2025.3605285
Hamza Rasaee;Taha Koleilat;Hassan Rivaz
Accurate and generalizable object segmentation in ultrasound imaging remains a significant challenge due to anatomical variability, diverse imaging protocols, and limited annotated data. In this study, we propose a prompt-driven vision–language model (VLM) that integrates grounding DINO with SAM2 to enable object segmentation across multiple ultrasound organs. A total of 18 public ultrasound datasets, encompassing the breast, thyroid, liver, prostate, kidney, and paraspinal muscle, were utilized. These datasets were divided into 15 for fine-tuning and validation of grounding DINO using low-rank adaptation (LoRA) to the ultrasound domain, and three were held out entirely for testing to evaluate performance in unseen distributions. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art (SOTA) segmentation methods, including UniverSeg, MedSAM, MedCLIP-segment anything model (SAM), BiomedParse, and SAMUS on most seen datasets while maintaining strong performance on unseen datasets without additional fine-tuning. These results underscore the promise of VLMs in scalable and robust ultrasound image analysis, reducing dependence on large, organ-specific annotated datasets. We will publish our code on code.sonography.ai after acceptance.
{"title":"Grounding DINO-US-SAM: Text-Prompted Multiorgan Segmentation in Ultrasound With LoRA-Tuned Vision–Language Models","authors":"Hamza Rasaee;Taha Koleilat;Hassan Rivaz","doi":"10.1109/TUFFC.2025.3605285","DOIUrl":"10.1109/TUFFC.2025.3605285","url":null,"abstract":"Accurate and generalizable object segmentation in ultrasound imaging remains a significant challenge due to anatomical variability, diverse imaging protocols, and limited annotated data. In this study, we propose a prompt-driven vision–language model (VLM) that integrates grounding DINO with SAM2 to enable object segmentation across multiple ultrasound organs. A total of 18 public ultrasound datasets, encompassing the breast, thyroid, liver, prostate, kidney, and paraspinal muscle, were utilized. These datasets were divided into 15 for fine-tuning and validation of grounding DINO using low-rank adaptation (LoRA) to the ultrasound domain, and three were held out entirely for testing to evaluate performance in unseen distributions. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art (SOTA) segmentation methods, including UniverSeg, MedSAM, MedCLIP-segment anything model (SAM), BiomedParse, and SAMUS on most seen datasets while maintaining strong performance on unseen datasets without additional fine-tuning. These results underscore the promise of VLMs in scalable and robust ultrasound image analysis, reducing dependence on large, organ-specific annotated datasets. We will publish our code on code.sonography.ai after acceptance.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1414-1425"},"PeriodicalIF":3.7,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952327","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}
Pub Date : 2025-09-02DOI: 10.1109/TUFFC.2025.3605440
M. Angerer;J. Lu;J. Welsch;M. Lu;C. Luo;W. Plunet;E. Cretu;W. Tetzlaff;R. Rohling
The ability of ultrasound imaging to deliver real-time visualization of tissue structures and surgical instruments can provide essential benefits in guiding medical interventions. In spinal cord injury research, small animal models are commonly used, but their size restricts the applicability of many standard ultrasound systems. Capacitive micromachined ultrasonic transducers (CMUTs) offer advantages over traditional piezoelectric transducers, including a smaller form factor, high design flexibility, and improved acoustic performance. CMUT structures made of polymers (polyCMUTs) can be produced cost-effectively and quickly, while potentially offering flexible, biocompatible transducers for next-generation ultrasound systems. This study introduces the first polyCMUT probe designed for imaging spinal cords in rats. A compact 11 MHz, 64 channel probe with a $12.9times 6.5$ mm small tip was developed through a three-stage fabrication process, combining in-house manufactured polyCMUT arrays with electronics, integrated in a research imaging system. Performance evaluation included electrical impedance measurements, acoustic characterization, and in vitro and ex-vivo imaging. Quality analysis validated the stability of the fabrication process, demonstrating high yield and minimal variability, with a standard deviation in resonance frequency of less than 1%. The probe successfully visualized key anatomical structures like the central canal as well as real-time imaging of needle insertion into tissue. However, distinguishing between gray and white matter remained challenging due to limitations in frequency and bandwidth. This study demonstrates the potential of the polyCMUT technology for developing tailored ultrasound solutions. Future work will focus on optimizing high-frequency performance and advancing toward in vivo applications to provide meaningful tools in spinal cord injury research and therapeutic interventions.
超声成像提供组织结构和手术器械实时可视化的能力可以为指导医疗干预提供必要的好处。在脊髓损伤研究中，常用的是小动物模型，但其尺寸限制了许多标准超声系统的适用性。电容式微机械超声换能器（cmut）比传统的压电换能器具有更小的外形、更高的设计灵活性和更好的声学性能等优点。由聚合物（polycmut）制成的CMUT结构可以经济高效、快速地生产，同时有可能为下一代超声系统提供灵活的、生物相容性的换能器。本研究介绍了首个用于大鼠脊髓成像的polyCMUT探针。紧凑的11mhz， 64通道探针，12.9 x 6.5 mm小尖端，通过三个阶段的制造工艺，结合内部制造的polyCMUT阵列和电子设备，集成在研究成像系统中。性能评估包括电阻抗测量、声学表征、体外和离体成像。质量分析验证了制造工艺的稳定性，证明了高成品率和最小的变异性，共振频率的标准偏差小于1%。该探头成功地显示了中心管等关键解剖结构以及针插入组织的实时成像。然而，由于频率和带宽的限制，区分灰质和白质仍然具有挑战性。这项研究证明了polyCMUT技术在开发量身定制的超声解决方案方面的潜力。未来的工作将集中在优化高频性能和推进体内应用，为脊髓损伤研究和治疗干预提供有意义的工具。
{"title":"A Polymer-Based CMUT Probe for Imaging the Spinal Cord in Rats","authors":"M. Angerer;J. Lu;J. Welsch;M. Lu;C. Luo;W. Plunet;E. Cretu;W. Tetzlaff;R. Rohling","doi":"10.1109/TUFFC.2025.3605440","DOIUrl":"10.1109/TUFFC.2025.3605440","url":null,"abstract":"The ability of ultrasound imaging to deliver real-time visualization of tissue structures and surgical instruments can provide essential benefits in guiding medical interventions. In spinal cord injury research, small animal models are commonly used, but their size restricts the applicability of many standard ultrasound systems. Capacitive micromachined ultrasonic transducers (CMUTs) offer advantages over traditional piezoelectric transducers, including a smaller form factor, high design flexibility, and improved acoustic performance. CMUT structures made of polymers (polyCMUTs) can be produced cost-effectively and quickly, while potentially offering flexible, biocompatible transducers for next-generation ultrasound systems. This study introduces the first polyCMUT probe designed for imaging spinal cords in rats. A compact 11 MHz, 64 channel probe with a <inline-formula> <tex-math>$12.9times 6.5$ </tex-math></inline-formula> mm small tip was developed through a three-stage fabrication process, combining in-house manufactured polyCMUT arrays with electronics, integrated in a research imaging system. Performance evaluation included electrical impedance measurements, acoustic characterization, and in vitro and ex-vivo imaging. Quality analysis validated the stability of the fabrication process, demonstrating high yield and minimal variability, with a standard deviation in resonance frequency of less than 1%. The probe successfully visualized key anatomical structures like the central canal as well as real-time imaging of needle insertion into tissue. However, distinguishing between gray and white matter remained challenging due to limitations in frequency and bandwidth. This study demonstrates the potential of the polyCMUT technology for developing tailored ultrasound solutions. Future work will focus on optimizing high-frequency performance and advancing toward in vivo applications to provide meaningful tools in spinal cord injury research and therapeutic interventions.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 11","pages":"1437-1447"},"PeriodicalIF":3.7,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952365","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}
Pub Date : 2025-08-29DOI: 10.1109/TUFFC.2025.3597498
{"title":"IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control Publication Information","authors":"","doi":"10.1109/TUFFC.2025.3597498","DOIUrl":"https://doi.org/10.1109/TUFFC.2025.3597498","url":null,"abstract":"","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 9","pages":"C2-C2"},"PeriodicalIF":3.7,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11144926","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144918123","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}
Super-resolution ultrasound (SRUS) technology based on contrast agents has shown great potential in in vivo microvascular blood flow imaging and has become a hot topic in the industry in recent years. SRUS represented by ultrasound localization microscopy (ULM) eliminates the point spread function (PSF) caused by diffraction by localizing sparse microbubbles in the image, and then constructs a super-resolution blood flow structure map through long-term image accumulation. It is worth mentioning that almost all current super-resolution strategies, including ULM, adopt postimage processing strategies. In optics, in addition to postimage processing super-resolution techniques, the design of the illumination light field is also a popular super-resolution imaging method. This inspired us to design the acoustic field to achieve a super-resolution blood flow imaging strategy based on structured acoustic field illumination. In this article, we propose a new structured acoustic field illumination design method, which achieves 2-D structured acoustic field design by improving the acoustic field calculation of Gerchberg–Saxton algorithm (GSA) based on Talbot effect. At the same time, we have designed a corresponding structured acoustic field reconstruction strategy that can adaptively reconstruct any structured illumination method, and the reconstructed image frame rate is comparable to that of multiangle plane wave. Combined with image postprocessing methods, we have obtained contrast-enhanced structured illumination ultrasound (CE-SIU), which reduces the full width at half maximum (FWHM) of the microbubble PSF, breaking through the lateral resolution limit of contrast-enhanced imaging. The super-resolution strategy based on acoustic field design proposed in this study provides new insights and perspectives for the development of the overall technical route of SRUS.
{"title":"Contrast-Enhanced Structured Illumination Ultrasound Imaging (CE-SIU)","authors":"Xiaoyu Qian;Jiabin Zhang;Yu Xia;Dongdong Liang;Jue Zhang","doi":"10.1109/TUFFC.2025.3603281","DOIUrl":"10.1109/TUFFC.2025.3603281","url":null,"abstract":"Super-resolution ultrasound (SRUS) technology based on contrast agents has shown great potential in in vivo microvascular blood flow imaging and has become a hot topic in the industry in recent years. SRUS represented by ultrasound localization microscopy (ULM) eliminates the point spread function (PSF) caused by diffraction by localizing sparse microbubbles in the image, and then constructs a super-resolution blood flow structure map through long-term image accumulation. It is worth mentioning that almost all current super-resolution strategies, including ULM, adopt postimage processing strategies. In optics, in addition to postimage processing super-resolution techniques, the design of the illumination light field is also a popular super-resolution imaging method. This inspired us to design the acoustic field to achieve a super-resolution blood flow imaging strategy based on structured acoustic field illumination. In this article, we propose a new structured acoustic field illumination design method, which achieves 2-D structured acoustic field design by improving the acoustic field calculation of Gerchberg–Saxton algorithm (GSA) based on Talbot effect. At the same time, we have designed a corresponding structured acoustic field reconstruction strategy that can adaptively reconstruct any structured illumination method, and the reconstructed image frame rate is comparable to that of multiangle plane wave. Combined with image postprocessing methods, we have obtained contrast-enhanced structured illumination ultrasound (CE-SIU), which reduces the full width at half maximum (FWHM) of the microbubble PSF, breaking through the lateral resolution limit of contrast-enhanced imaging. The super-resolution strategy based on acoustic field design proposed in this study provides new insights and perspectives for the development of the overall technical route of SRUS.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1401-1413"},"PeriodicalIF":3.7,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952304","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}
Pub Date : 2025-08-22DOI: 10.1109/TUFFC.2025.3601660
Junhao Wang;Jiao Xia;Aocheng Bao;Chong Yang;Jinghan Gan;Lei Zhao;Bowen Sheng;Wei Wang;Yipeng Lu
Ultrasound propagation attenuation coefficient detection is a widely used technique for distinguishing different media types. Multifrequency ultrasound detection provides comprehensive information but requires ultrasonic transducers with a broad bandwidth. This study presents a piezoelectric micromachined ultrasonic transducer (PMUT) array that integrates multiple frequency elements with optimized spacing to achieve the fusion of multiple vibration modes, resulting in a −6 dB emission fractional bandwidth of 230% and pulse-echo fractional bandwidth up to 146%. This ultrawide bandwidth facilitates accurate pulse-echo signal reception across a broad frequency range while minimizing signal overlap caused by ringdown effects. Moreover, optimizing the PMUT array size relative to wavelength achieved a highly directional 5° acoustic beam, ensuring effective penetration through high attenuation or multilayered structures. Experimental results indicate that attenuation coefficients in ice and liquid correlate with their material composition and structural properties. These findings highlight the significant potential of the PMUT array for identifying and analyzing surface media, with promising applications in power supply systems, transportation, industrial production, and so on.
{"title":"Surface Condition Sensing With Broadband and Highly Directional PMUT Array","authors":"Junhao Wang;Jiao Xia;Aocheng Bao;Chong Yang;Jinghan Gan;Lei Zhao;Bowen Sheng;Wei Wang;Yipeng Lu","doi":"10.1109/TUFFC.2025.3601660","DOIUrl":"10.1109/TUFFC.2025.3601660","url":null,"abstract":"Ultrasound propagation attenuation coefficient detection is a widely used technique for distinguishing different media types. Multifrequency ultrasound detection provides comprehensive information but requires ultrasonic transducers with a broad bandwidth. This study presents a piezoelectric micromachined ultrasonic transducer (PMUT) array that integrates multiple frequency elements with optimized spacing to achieve the fusion of multiple vibration modes, resulting in a −6 dB emission fractional bandwidth of 230% and pulse-echo fractional bandwidth up to 146%. This ultrawide bandwidth facilitates accurate pulse-echo signal reception across a broad frequency range while minimizing signal overlap caused by ringdown effects. Moreover, optimizing the PMUT array size relative to wavelength achieved a highly directional 5° acoustic beam, ensuring effective penetration through high attenuation or multilayered structures. Experimental results indicate that attenuation coefficients in ice and liquid correlate with their material composition and structural properties. These findings highlight the significant potential of the PMUT array for identifying and analyzing surface media, with promising applications in power supply systems, transportation, industrial production, and so on.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1324-1335"},"PeriodicalIF":3.7,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952336","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}
Pub Date : 2025-08-21DOI: 10.1109/TUFFC.2025.3601216
Jian-Yu Lu
Bessel beams are exact solutions to the isotropic/homogeneous wave equation. In theory, they can propagate to infinite distances without diffraction. In practice, when produced with a finite aperture, they have a very large depth of field, i.e., they can maintain a small beamwidth over a large distance. In addition, they have a self-healing ability after encountering an obstacle. Because of these properties, Bessel beams have applications in optics, electromagnetics, ultrasound, quantum communications, electron beam guidance, and so on. Previously, in ultrasound, Bessel beams were produced with an annular array transducer driven by multiple independent high-voltage radio frequency (RF) power amplifiers that were bulky, heavy, and consumed a lot of power, which limited the Bessel beams in applications such as wearable medical ultrasound imaging and wearable super-resolution imaging. In this article, pulse (broadband) Bessel beams were produced by a single high-voltage RF power amplifier in combination with an RF transformer, reducing the size, weight, and power consumption. Experiments were performed to produce the pulse Bessel beams in water with a custom RF transformer and a custom ten-ring, 50-mm diameter, 2.5-MHz center frequency, and broadband (about 72% -6 dB relative one-way bandwidth) 1–3 lead zirconate titanate (PZT) ceramic/polymer composite annular array transducer driven by a commercial RF power amplifier at about ±90 V. The results show that the pulse Bessel beams produced were very close to those generated with ten independent high-voltage RF power amplifiers, computer simulations, and theory, and the pulse Bessel beams had a -6-dB beamwidth of about 2.53 mm ($4.22lambda $ ) and a depth of field of about 216 mm ($360lambda $ ). The reduced number of high-voltage RF power amplifiers makes it easier to apply Bessel beams in applications such as wearable medical ultrasound imaging and wearable super-resolution imaging, as is illustrated in examples where three-dimensional (3-D) or multi-plane images can be produced using a Bessel beam and a mechanically scanned multi-directional vibrating reflector.
{"title":"Producing Bessel Beams With an RF Transformer","authors":"Jian-Yu Lu","doi":"10.1109/TUFFC.2025.3601216","DOIUrl":"10.1109/TUFFC.2025.3601216","url":null,"abstract":"Bessel beams are exact solutions to the isotropic/homogeneous wave equation. In theory, they can propagate to infinite distances without diffraction. In practice, when produced with a finite aperture, they have a very large depth of field, i.e., they can maintain a small beamwidth over a large distance. In addition, they have a self-healing ability after encountering an obstacle. Because of these properties, Bessel beams have applications in optics, electromagnetics, ultrasound, quantum communications, electron beam guidance, and so on. Previously, in ultrasound, Bessel beams were produced with an annular array transducer driven by multiple independent high-voltage radio frequency (RF) power amplifiers that were bulky, heavy, and consumed a lot of power, which limited the Bessel beams in applications such as wearable medical ultrasound imaging and wearable super-resolution imaging. In this article, pulse (broadband) Bessel beams were produced by a single high-voltage RF power amplifier in combination with an RF transformer, reducing the size, weight, and power consumption. Experiments were performed to produce the pulse Bessel beams in water with a custom RF transformer and a custom ten-ring, 50-mm diameter, 2.5-MHz center frequency, and broadband (about 72% -6 dB relative one-way bandwidth) 1–3 lead zirconate titanate (PZT) ceramic/polymer composite annular array transducer driven by a commercial RF power amplifier at about ±90 V. The results show that the pulse Bessel beams produced were very close to those generated with ten independent high-voltage RF power amplifiers, computer simulations, and theory, and the pulse Bessel beams had a -6-dB beamwidth of about 2.53 mm (<inline-formula> <tex-math>$4.22lambda $ </tex-math></inline-formula>) and a depth of field of about 216 mm (<inline-formula> <tex-math>$360lambda $ </tex-math></inline-formula>). The reduced number of high-voltage RF power amplifiers makes it easier to apply Bessel beams in applications such as wearable medical ultrasound imaging and wearable super-resolution imaging, as is illustrated in examples where three-dimensional (3-D) or multi-plane images can be produced using a Bessel beam and a mechanically scanned multi-directional vibrating reflector.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1426-1435"},"PeriodicalIF":3.7,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952294","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}
Pub Date : 2025-08-20DOI: 10.1109/TUFFC.2025.3600902
Nathan Painchaud;Jérémie Stym-Popper;Pierre-Yves Courand;Nicolas Thome;Pierre-Marc Jodoin;Nicolas Duchateau;Olivier Bernard
Deep learning enables automatic and robust extraction of cardiac function descriptors from echocardiographic sequences, such as ejection fraction (EF) or strain. These descriptors provide fine-grained information that physicians consider, in conjunction with more global variables from the clinical record, to assess patients’ condition. Drawing on novel Transformer models applied to tabular data, we propose a method that considers all descriptors extracted from medical records and echocardiograms to learn the representation of a cardiovascular pathology with a difficult-to-characterize continuum, namely hypertension. Our method first projects each variable into its own representation space using modality-specific approaches. These standardized representations of multimodal data are then fed to a Transformer encoder, which learns to merge them into a comprehensive representation of the patient through the task of predicting a clinical rating. This stratification task is formulated as an ordinal classification to enforce a pathological continuum in the representation space. We observe the major trends along this continuum on a cohort of 239 hypertensive patients, providing unprecedented details in the description of hypertension’s impact on various cardiac function descriptors. Our analysis shows that: 1) the XTab foundation model’s architecture allows to reach high performance (96.8% AUROC) even with limited data (less than 200 training samples); 2) stratification across the population is reproducible between trainings [within 5.7% of mean absolute error (MAE)]; and 3) patterns emerge in descriptors, some of which align with established physiological knowledge about hypertension, while others could pave the way for a more comprehensive understanding of this pathology. The code is available at https://github.com/creatis-myriad/didactic
{"title":"Fusing Echocardiography Images and Medical Records for Continuous Patient Stratification","authors":"Nathan Painchaud;Jérémie Stym-Popper;Pierre-Yves Courand;Nicolas Thome;Pierre-Marc Jodoin;Nicolas Duchateau;Olivier Bernard","doi":"10.1109/TUFFC.2025.3600902","DOIUrl":"10.1109/TUFFC.2025.3600902","url":null,"abstract":"Deep learning enables automatic and robust extraction of cardiac function descriptors from echocardiographic sequences, such as ejection fraction (EF) or strain. These descriptors provide fine-grained information that physicians consider, in conjunction with more global variables from the clinical record, to assess patients’ condition. Drawing on novel Transformer models applied to tabular data, we propose a method that considers all descriptors extracted from medical records and echocardiograms to learn the representation of a cardiovascular pathology with a difficult-to-characterize continuum, namely hypertension. Our method first projects each variable into its own representation space using modality-specific approaches. These standardized representations of multimodal data are then fed to a Transformer encoder, which learns to merge them into a comprehensive representation of the patient through the task of predicting a clinical rating. This stratification task is formulated as an ordinal classification to enforce a pathological continuum in the representation space. We observe the major trends along this continuum on a cohort of 239 hypertensive patients, providing unprecedented details in the description of hypertension’s impact on various cardiac function descriptors. Our analysis shows that: 1) the XTab foundation model’s architecture allows to reach high performance (96.8% AUROC) even with limited data (less than 200 training samples); 2) stratification across the population is reproducible between trainings [within 5.7% of mean absolute error (MAE)]; and 3) patterns emerge in descriptors, some of which align with established physiological knowledge about hypertension, while others could pave the way for a more comprehensive understanding of this pathology. The code is available at <uri>https://github.com/creatis-myriad/didactic</uri>","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1388-1400"},"PeriodicalIF":3.7,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952292","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}
Pub Date : 2025-08-19DOI: 10.1109/TUFFC.2025.3600372
Hyungjoo Park;Tri Vu;Junjie Yao;Yun Jing
This work compares the effectiveness of four image reconstruction techniques for a ring-array photoacoustic computed tomography (PACT) system: the delay-and-sum (DAS), the interpolated model matrix inversion (IMMI), the frequency-domain model-based (FDMB), and the time-reversal (TR) methods. Image quality, computational efficiency, and robustness to noise and spatial/electrical impulse response (EIR) are evaluated to assess the four methods for 2-D imaging. Although it is thought to have limitations in producing high-resolution and artifact-free images, the DAS method is commonly employed in clinical applications because of its simplicity and real-time imaging capability. On the other hand, model-based (MB) methods, such as the IMMI, the FDMB, and the TR methods, use comprehensive physics for a more accurate description of wave propagation through tissue and thus have the potential to provide higher image quality and quantitative information. These approaches, however, require substantially more processing power, which can make them less suitable for real-time applications. Our results show that the DAS method achieves real-time performance while delivering better image quality for an ex vivo elongated target, whereas both the IMMI and the FDMB methods provide enhanced detail and resolution for thin-slice targets, at the cost of considerably slower reconstruction times. The TR method achieves reasonable resolution and detail for all cases but shows the slowest image reconstruction time. This work highlights the strengths and limitations of each method, offering valuable insights into selecting the most appropriate reconstruction technique for ring-array PACT systems.
{"title":"A Comparison of Image Reconstruction Methods for Ring-Array Photoacoustic Computed Tomography","authors":"Hyungjoo Park;Tri Vu;Junjie Yao;Yun Jing","doi":"10.1109/TUFFC.2025.3600372","DOIUrl":"10.1109/TUFFC.2025.3600372","url":null,"abstract":"This work compares the effectiveness of four image reconstruction techniques for a ring-array photoacoustic computed tomography (PACT) system: the delay-and-sum (DAS), the interpolated model matrix inversion (IMMI), the frequency-domain model-based (FDMB), and the time-reversal (TR) methods. Image quality, computational efficiency, and robustness to noise and spatial/electrical impulse response (EIR) are evaluated to assess the four methods for 2-D imaging. Although it is thought to have limitations in producing high-resolution and artifact-free images, the DAS method is commonly employed in clinical applications because of its simplicity and real-time imaging capability. On the other hand, model-based (MB) methods, such as the IMMI, the FDMB, and the TR methods, use comprehensive physics for a more accurate description of wave propagation through tissue and thus have the potential to provide higher image quality and quantitative information. These approaches, however, require substantially more processing power, which can make them less suitable for real-time applications. Our results show that the DAS method achieves real-time performance while delivering better image quality for an ex vivo elongated target, whereas both the IMMI and the FDMB methods provide enhanced detail and resolution for thin-slice targets, at the cost of considerably slower reconstruction times. The TR method achieves reasonable resolution and detail for all cases but shows the slowest image reconstruction time. This work highlights the strengths and limitations of each method, offering valuable insights into selecting the most appropriate reconstruction technique for ring-array PACT systems.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"72 10","pages":"1376-1387"},"PeriodicalIF":3.7,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144882748","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：