Pub Date : 2024-05-23DOI: 10.1109/TUFFC.2024.3404622
Erlend Loland Gundersen, Erik Smistad, Tollef Struksnes Jahren, Svein-Erik Masoy
Deep learning (DL) models have emerged as alternative methods to conventional ultrasound (US) signal processing, offering the potential to mimic signal processing chains, reduce inference time, and enable the portability of processing chains across hardware. This paper proposes a DL model that replicates the fine-tuned BMode signal processing chain of a high-end US system and explores the potential of using it with a different probe and a lower-end system. A deep neural network was trained in a supervised manner to map raw beamformed in-phase and quadrature component data into processed images. The dataset consisted of 30,000 cardiac image frames acquired using the GE HealthCare Vivid E95 system with the 4Vc-D matrix array probe. The signal processing chain includes depth-dependent bandpass filtering, elevation compounding, frequency compounding, and image compression and filtering. The results indicate that a lightweight DL model can accurately replicate the signal processing chain of a commercial scanner for a given application. Evaluation on a 15 patient test dataset of about three thousand image frames gave a structural similarity index measure of 98.56 ± 0.49. Applying the DL model to data from another probe showed equivalent or improved image quality. This indicates that a single DL model may be used for a set of probes on a given system that targets the same application, which could be a cost-effective tuning and implementation strategy for vendors. Further, the DL model enhanced image quality on a Verasonics dataset, suggesting the potential to port features from high-end US systems to lower-end counterparts.
{"title":"Hardware-Independent Deep Signal Processing: A Feasibility Study in Echocardiography.","authors":"Erlend Loland Gundersen, Erik Smistad, Tollef Struksnes Jahren, Svein-Erik Masoy","doi":"10.1109/TUFFC.2024.3404622","DOIUrl":"10.1109/TUFFC.2024.3404622","url":null,"abstract":"<p><p>Deep learning (DL) models have emerged as alternative methods to conventional ultrasound (US) signal processing, offering the potential to mimic signal processing chains, reduce inference time, and enable the portability of processing chains across hardware. This paper proposes a DL model that replicates the fine-tuned BMode signal processing chain of a high-end US system and explores the potential of using it with a different probe and a lower-end system. A deep neural network was trained in a supervised manner to map raw beamformed in-phase and quadrature component data into processed images. The dataset consisted of 30,000 cardiac image frames acquired using the GE HealthCare Vivid E95 system with the 4Vc-D matrix array probe. The signal processing chain includes depth-dependent bandpass filtering, elevation compounding, frequency compounding, and image compression and filtering. The results indicate that a lightweight DL model can accurately replicate the signal processing chain of a commercial scanner for a given application. Evaluation on a 15 patient test dataset of about three thousand image frames gave a structural similarity index measure of 98.56 ± 0.49. Applying the DL model to data from another probe showed equivalent or improved image quality. This indicates that a single DL model may be used for a set of probes on a given system that targets the same application, which could be a cost-effective tuning and implementation strategy for vendors. Further, the DL model enhanced image quality on a Verasonics dataset, suggesting the potential to port features from high-end US systems to lower-end counterparts.</p>","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141086610","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 : 2024-05-03DOI: 10.1109/TUFFC.2024.3396796
Rashid Al Mukaddim, Emily Mackay, Nils Gessert, Ramon Erkamp, Shriram Sethuraman, Jonathan Sutton, Shyam Bharat, Melanie Jutras, Cristiana Baloescu, Christopher L Moore, Balasundar Raju
The reliability of automated image interpretation of point-of-care (POC) echocardiography scans depends on the quality of the acquired ultrasound data. This work reports on the development and validation of spatiotemporal deep learning models to assess the suitability of input ultrasound cine loops collected using a handheld echocardiography device for processing by an automated quantification algorithm (e.g. ejection fraction estimation). POC echocardiograms (n=885 DICOM cine loops from 175 patients) from two sites were collected using a handheld ultrasound device and annotated for image quality at the frame-level. Attributes of high-quality frames for left ventricular (LV) quantification included a temporally-stable LV, reasonable coverage of LV borders, and good contrast between the borders and chamber. Attributes of low-quality frames included temporal instability of the LV and/or imaging artifacts (e.g., lack of contrast, haze, reverberation, acoustic shadowing). Three different neural network architectures were investigated - (a) frame-level convolutional neural network (CNN) which operates on individual echo frames (VectorCNN), (b) single-stream sequence-level CNN which operates on a sequence of echo frames (VectorCNN+LSTM) and (c) two-stream sequence-level CNNs which operate on a sequence of echo and optical flow frames (VectorCNN+LSTM+Average, VectorCNN+LSTM+MinMax, and VectorCNN+LSTM+ConvPool). Evaluation on a sequestered test dataset containing 76 DICOM cine loops with 16,914 frames showed that VectorCNN+LSTM can effectively utilize both spatial and temporal information to regress the quality of an input frame (accuracy: 0.925, sensitivity = 0.860, specificity = 0.952), compared to the frame-level VectorCNN that only utilizes spatial information in that frame (accuracy: 0.903, sensitivity = 0.791, specificity = 0.949). Furthermore, an independent sample t-test indicated that the cine loops classified to be of adequate quality by the VectorCNN+LSTM model had a statistically significant lower bias in the automatically estimated EF (mean bias = - 3.73 ± 7.46 %, versus a clinically obtained reference EF) compared to the loops classified as inadequate (mean bias = -15.92 ± 12.17 %) (p = 0.007). Thus, cine loop stratification using the proposed spatiotemporal CNN model improves the reliability of automated point-of-care echocardiography image interpretation.
{"title":"Spatiotemporal Deep Learning-Based Cine Loop Quality Filter for Handheld Point-of-Care Echocardiography.","authors":"Rashid Al Mukaddim, Emily Mackay, Nils Gessert, Ramon Erkamp, Shriram Sethuraman, Jonathan Sutton, Shyam Bharat, Melanie Jutras, Cristiana Baloescu, Christopher L Moore, Balasundar Raju","doi":"10.1109/TUFFC.2024.3396796","DOIUrl":"https://doi.org/10.1109/TUFFC.2024.3396796","url":null,"abstract":"<p><p>The reliability of automated image interpretation of point-of-care (POC) echocardiography scans depends on the quality of the acquired ultrasound data. This work reports on the development and validation of spatiotemporal deep learning models to assess the suitability of input ultrasound cine loops collected using a handheld echocardiography device for processing by an automated quantification algorithm (e.g. ejection fraction estimation). POC echocardiograms (n=885 DICOM cine loops from 175 patients) from two sites were collected using a handheld ultrasound device and annotated for image quality at the frame-level. Attributes of high-quality frames for left ventricular (LV) quantification included a temporally-stable LV, reasonable coverage of LV borders, and good contrast between the borders and chamber. Attributes of low-quality frames included temporal instability of the LV and/or imaging artifacts (e.g., lack of contrast, haze, reverberation, acoustic shadowing). Three different neural network architectures were investigated - (a) frame-level convolutional neural network (CNN) which operates on individual echo frames (VectorCNN), (b) single-stream sequence-level CNN which operates on a sequence of echo frames (VectorCNN+LSTM) and (c) two-stream sequence-level CNNs which operate on a sequence of echo and optical flow frames (VectorCNN+LSTM+Average, VectorCNN+LSTM+MinMax, and VectorCNN+LSTM+ConvPool). Evaluation on a sequestered test dataset containing 76 DICOM cine loops with 16,914 frames showed that VectorCNN+LSTM can effectively utilize both spatial and temporal information to regress the quality of an input frame (accuracy: 0.925, sensitivity = 0.860, specificity = 0.952), compared to the frame-level VectorCNN that only utilizes spatial information in that frame (accuracy: 0.903, sensitivity = 0.791, specificity = 0.949). Furthermore, an independent sample t-test indicated that the cine loops classified to be of adequate quality by the VectorCNN+LSTM model had a statistically significant lower bias in the automatically estimated EF (mean bias = - 3.73 ± 7.46 %, versus a clinically obtained reference EF) compared to the loops classified as inadequate (mean bias = -15.92 ± 12.17 %) (p = 0.007). Thus, cine loop stratification using the proposed spatiotemporal CNN model improves the reliability of automated point-of-care echocardiography image interpretation.</p>","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"PP ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2024-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140858828","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 : 2024-04-30DOI: 10.1109/TUFFC.2024.3395490
Di Xiao;Pat De la Torre;Alfred C. H. Yu
Speed-of-sound (SoS) is an intrinsic acoustic property of human tissues and has been regarded as a potential biomarker of tissue health. To foster the clinical use of this emerging biomarker in medical diagnostics, it is important for SoS estimates to be derived and displayed in real time. Here, we demonstrate that concurrent global SoS estimation and B-mode imaging can be achieved live on a portable ultrasound scanner. Our innovation is hinged upon the design of a novel pulse-echo SoS estimation framework that is based on steered plane wave imaging. It has accounted for the effects of refraction and imaging depth when the medium SoS differs from the nominal value of 1540 m/s that is conventionally used in medical imaging. The accuracy of our SoS estimation framework was comparatively analyzed with through-transmit time-of-flight measurements in vitro on 15 custom agar phantoms with different SoS values (1508–1682 m/s) and in vivo on human calf muscles ( ${N} =9$ ; SoS range: 1560–1586 m/s). Our SoS estimation framework has a mean signed difference (MSD) of $- 0.6 , pm , 2.3$ m/s in vitro and $- 2.2 , pm , 11.2$ m/s in vivo relative to the reference measurements. In addition, our real-time system prototype has yielded simultaneous SoS estimates and B-mode imaging at an average frame rate of 18.1 fps. Overall, by realizing real-time tissue SoS estimation with B-mode imaging, our innovation can foster the use of tissue SoS as a biomarker in medical ultrasound diagnostics.
声速（SoS）是人体组织固有的声学特性，被视为组织健康的潜在生物标志物。为了促进这一新兴生物标志物在医疗诊断中的临床应用，必须实时得出并显示声速估计值。在这里，我们展示了在便携式超声扫描仪上可同时实现全局 SoS 估计和 B 型成像。我们的创新在于设计了一种基于转向平面波成像的新型脉冲回波 SoS 估计框架。当介质 SoS 不同于医学成像中通常使用的 1540 米/秒的标称值时，它考虑了折射和成像深度的影响。我们的 SoS 估计框架的准确性通过飞行时间测量进行了比较分析，这些测量在体外对 15 个定制的琼脂模型进行，其 SoS 值各不相同（1508 至 1682 m/s），在体内对人体小腿肌肉进行（N = 9；SoS 范围：1560 至 1586 m/s）。与参考测量值相比，我们的 SoS 估算框架在体外的平均符号差分别为 -0.6±2.3 m/s，在体内的平均符号差为 -2.2±11.2 m/s。此外，我们的实时系统原型还能以 18.1 帧/秒的平均帧频同时进行 SoS 估计和 B 模式成像。总之，通过实现实时组织SoS估计和B模式成像，我们的创新可以促进组织SoS作为生物标志物在医学超声诊断中的应用。
{"title":"Real-Time Speed-of-Sound Estimation In Vivo via Steered Plane Wave Ultrasound","authors":"Di Xiao;Pat De la Torre;Alfred C. H. Yu","doi":"10.1109/TUFFC.2024.3395490","DOIUrl":"10.1109/TUFFC.2024.3395490","url":null,"abstract":"Speed-of-sound (SoS) is an intrinsic acoustic property of human tissues and has been regarded as a potential biomarker of tissue health. To foster the clinical use of this emerging biomarker in medical diagnostics, it is important for SoS estimates to be derived and displayed in real time. Here, we demonstrate that concurrent global SoS estimation and B-mode imaging can be achieved live on a portable ultrasound scanner. Our innovation is hinged upon the design of a novel pulse-echo SoS estimation framework that is based on steered plane wave imaging. It has accounted for the effects of refraction and imaging depth when the medium SoS differs from the nominal value of 1540 m/s that is conventionally used in medical imaging. The accuracy of our SoS estimation framework was comparatively analyzed with through-transmit time-of-flight measurements in vitro on 15 custom agar phantoms with different SoS values (1508–1682 m/s) and in vivo on human calf muscles (\u0000<inline-formula> <tex-math>${N} =9$ </tex-math></inline-formula>\u0000; SoS range: 1560–1586 m/s). Our SoS estimation framework has a mean signed difference (MSD) of \u0000<inline-formula> <tex-math>$- 0.6 , pm , 2.3$ </tex-math></inline-formula>\u0000 m/s in vitro and \u0000<inline-formula> <tex-math>$- 2.2 , pm , 11.2$ </tex-math></inline-formula>\u0000 m/s in vivo relative to the reference measurements. In addition, our real-time system prototype has yielded simultaneous SoS estimates and B-mode imaging at an average frame rate of 18.1 fps. Overall, by realizing real-time tissue SoS estimation with B-mode imaging, our innovation can foster the use of tissue SoS as a biomarker in medical ultrasound diagnostics.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 6","pages":"673-686"},"PeriodicalIF":3.6,"publicationDate":"2024-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10511251","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140862015","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 : 2024-04-30DOI: 10.1109/TUFFC.2024.3395525
Zhiyi Wen;Dawei Wu;Kentaro Nakamura
Scaling down linear actuators is crucial for various industrial applications, yet the efficiency of electromagnetic linear actuators decreases significantly as they are miniaturized. Millimeter-scale miniature ultrasonic motors, on the other hand, maintain high efficiency. This article describes a new approach to facilitate the miniaturization of traveling wave linear ultrasonic motors by attaching bimorph transducers to the ends of the stator beams. To control the resonance frequency and facilitate the generation of a traveling flexural wave in the beam, grooves are incorporated into the bimorph structure. Mechanical output is improved by amplifying the transverse displacement through the addition of teeth to the beam. Utilizing the finite element method (FEM), a prototype measuring $10times 10times 160$ mm was designed, fabricated, and tested. It achieved an output speed of 53.7 mm/s and a thrust of 0.83 N at a peak-to-peak voltage of 300 V and a frequency of 32.7 kHz. The results show that the proposed ultrasonic linear motors with small size, simple structure, and overall compactness have promising applications in robotics, precision machining, medical equipment, and other fields requiring miniature compact linear motions.
对于各种工业应用而言，缩小线性致动器的尺寸至关重要，但电磁线性致动器的效率会随着微型化而显著降低。而毫米级微型超声波电机却能保持高效率。本文介绍了一种促进行波线性超声波电动机微型化的新方法，即在定子横梁的两端安装双态传感器。为了控制共振频率并促进在横梁中产生行波挠性波，双晶结构中加入了凹槽。通过在横梁上增加齿来放大横向位移，从而提高机械输出。利用有限元法（FEM），设计、制造并测试了一个尺寸为 10 mm × 10 mm × 160 mm 的原型。在峰-峰电压为 300 V、频率为 32.7 kHz 时，输出速度达到 53.7 mm/s，推力为 0.83 N。结果表明，所提出的超声波直线电机体积小、结构简单、整体紧凑，在机器人、精密加工、医疗设备和其他需要微型紧凑直线运动的领域具有广阔的应用前景。
{"title":"A Miniature Traveling Wave Ultrasonic Linear Motor Utilizing Bimorph Transducers","authors":"Zhiyi Wen;Dawei Wu;Kentaro Nakamura","doi":"10.1109/TUFFC.2024.3395525","DOIUrl":"10.1109/TUFFC.2024.3395525","url":null,"abstract":"Scaling down linear actuators is crucial for various industrial applications, yet the efficiency of electromagnetic linear actuators decreases significantly as they are miniaturized. Millimeter-scale miniature ultrasonic motors, on the other hand, maintain high efficiency. This article describes a new approach to facilitate the miniaturization of traveling wave linear ultrasonic motors by attaching bimorph transducers to the ends of the stator beams. To control the resonance frequency and facilitate the generation of a traveling flexural wave in the beam, grooves are incorporated into the bimorph structure. Mechanical output is improved by amplifying the transverse displacement through the addition of teeth to the beam. Utilizing the finite element method (FEM), a prototype measuring \u0000<inline-formula> <tex-math>$10times 10times 160$ </tex-math></inline-formula>\u0000 mm was designed, fabricated, and tested. It achieved an output speed of 53.7 mm/s and a thrust of 0.83 N at a peak-to-peak voltage of 300 V and a frequency of 32.7 kHz. The results show that the proposed ultrasonic linear motors with small size, simple structure, and overall compactness have promising applications in robotics, precision machining, medical equipment, and other fields requiring miniature compact linear motions.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 6","pages":"698-706"},"PeriodicalIF":3.6,"publicationDate":"2024-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140861107","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}
The development of high-temperature nondestructive testing (NDT) requires ultrasonic transducers with good temperature resistance and high sensitivity for improved detection efficiency. Piezoelectric composite can improve the performance of transducers because of its high electromechanical coupling coefficient and adjustable acoustic impedance. In this study, 1-3-2 composites and 1-3-2 high-temperature composite ultrasonic transducers (HTCUTs) based on 0.36BiScO3-0.64PbTiO3 (BSPT), which is preferred piezoelectric materials at $200~^{circ }$ C– $300~^{circ }$ C, and high-temperature epoxy with a center frequency of 6 MHz were designed and fabricated. From $25~^{circ }$ C to $250~^{circ }$ C, 1-3-2 composites show a higher electromechanical coupling coefficient ${k}_{t}$ especially at high temperatures (~0.53 at $25~^{circ }$ C and ~0.64 at $250~^{circ }$ C) than monolithic BSPT (~0.5). The signal of the pulse-echo response of 1-3-2 HTCUTs is distinguishable up to $250~^{circ }$ C and remains stable ( ${V}_{text {pp}}~sim 500$ mV) below $150~^{circ }$ C, exhibiting higher sensitivity (improved by 7 dB) than that of monolithic BSPT high-temperature ultrasonic transducers (HTUTs). Bandwidth has been greatly enhanced especially at high temperatures (~103% at $250~^{circ }$ C) compared with that of monolithic BSPT HTUTs (~30% at $250~^{circ }$ C). To verify the excellent performance, B-mode scanning imaging measurement of a stepped steel block and defect location detection of a steel block was performed, showing the potential for high-temperature NDT applications.
{"title":"0.36BiScO3-0.64PbTiO3/Epoxy 1-3-2 High- Temperature Composite Ultrasonic Transducer for Nondestructive Testing Applications","authors":"Juan Zhang;Chenxue Hou;Tian-Long Zhao;Kefei Shi;Yi Li;Yecheng Wang;Mengqing Zhou;Xinhao Sun;Yi Quan;Zhaoxi Li;Yintang Yang;Chunlong Fei","doi":"10.1109/TUFFC.2024.3388552","DOIUrl":"10.1109/TUFFC.2024.3388552","url":null,"abstract":"The development of high-temperature nondestructive testing (NDT) requires ultrasonic transducers with good temperature resistance and high sensitivity for improved detection efficiency. Piezoelectric composite can improve the performance of transducers because of its high electromechanical coupling coefficient and adjustable acoustic impedance. In this study, 1-3-2 composites and 1-3-2 high-temperature composite ultrasonic transducers (HTCUTs) based on 0.36BiScO3-0.64PbTiO3 (BSPT), which is preferred piezoelectric materials at \u0000<inline-formula> <tex-math>$200~^{circ }$ </tex-math></inline-formula>\u0000C–\u0000<inline-formula> <tex-math>$300~^{circ }$ </tex-math></inline-formula>\u0000C, and high-temperature epoxy with a center frequency of 6 MHz were designed and fabricated. From \u0000<inline-formula> <tex-math>$25~^{circ }$ </tex-math></inline-formula>\u0000C to \u0000<inline-formula> <tex-math>$250~^{circ }$ </tex-math></inline-formula>\u0000C, 1-3-2 composites show a higher electromechanical coupling coefficient \u0000<inline-formula> <tex-math>${k}_{t}$ </tex-math></inline-formula>\u0000 especially at high temperatures (~0.53 at \u0000<inline-formula> <tex-math>$25~^{circ }$ </tex-math></inline-formula>\u0000C and ~0.64 at \u0000<inline-formula> <tex-math>$250~^{circ }$ </tex-math></inline-formula>\u0000C) than monolithic BSPT (~0.5). The signal of the pulse-echo response of 1-3-2 HTCUTs is distinguishable up to \u0000<inline-formula> <tex-math>$250~^{circ }$ </tex-math></inline-formula>\u0000C and remains stable (\u0000<inline-formula> <tex-math>${V}_{text {pp}}~sim 500$ </tex-math></inline-formula>\u0000 mV) below \u0000<inline-formula> <tex-math>$150~^{circ }$ </tex-math></inline-formula>\u0000C, exhibiting higher sensitivity (improved by 7 dB) than that of monolithic BSPT high-temperature ultrasonic transducers (HTUTs). Bandwidth has been greatly enhanced especially at high temperatures (~103% at \u0000<inline-formula> <tex-math>$250~^{circ }$ </tex-math></inline-formula>\u0000C) compared with that of monolithic BSPT HTUTs (~30% at \u0000<inline-formula> <tex-math>$250~^{circ }$ </tex-math></inline-formula>\u0000C). To verify the excellent performance, B-mode scanning imaging measurement of a stepped steel block and defect location detection of a steel block was performed, showing the potential for high-temperature NDT applications.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 6","pages":"638-647"},"PeriodicalIF":3.6,"publicationDate":"2024-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140677001","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 : 2024-04-18DOI: 10.1109/TUFFC.2024.3390807
Sean J. Z. Wong;Kaustav Roy;Chengkuo Lee;Yao Zhu
Thin-film piezoelectric micromachined ultrasound transducers (PMUTs) are an increasingly relevant and well-researched field, and their biomedical importance has been growing as the technology continues to mature. This review article briefly discusses their history in biomedical use, provides a simple explanation of their principles for newer readers, and sheds light on the materials selection for these devices. Primarily, it discusses the significant applications of PMUTs in the biomedical industry and showcases recent progress that has been made in each application. The biomedical applications covered include common historical uses of ultrasound such as ultrasound imaging, ultrasound therapy, and fluid sensing, but additionally new and upcoming applications such as drug delivery, photoacoustic imaging, thermoacoustic imaging, biometrics, and intrabody communication. By including a device comparison chart for different applications, this review aims to assist microelectromechanical systems (MEMS) designers that work with PMUTs by providing a benchmark for recent research works. Furthermore, it puts forth a discussion on the current challenges being faced by PMUTs in the biomedical field, current and likely future research trends, and opportunities for PMUT development areas, as well as sharing the opinions and predictions of the authors on the state of this technology as a whole. The review aims to be a comprehensive introduction to these topics without diving excessively deep into existing literature.
{"title":"Thin-Film Piezoelectric Micromachined Ultrasound Transducers in Biomedical Applications: A Review","authors":"Sean J. Z. Wong;Kaustav Roy;Chengkuo Lee;Yao Zhu","doi":"10.1109/TUFFC.2024.3390807","DOIUrl":"10.1109/TUFFC.2024.3390807","url":null,"abstract":"Thin-film piezoelectric micromachined ultrasound transducers (PMUTs) are an increasingly relevant and well-researched field, and their biomedical importance has been growing as the technology continues to mature. This review article briefly discusses their history in biomedical use, provides a simple explanation of their principles for newer readers, and sheds light on the materials selection for these devices. Primarily, it discusses the significant applications of PMUTs in the biomedical industry and showcases recent progress that has been made in each application. The biomedical applications covered include common historical uses of ultrasound such as ultrasound imaging, ultrasound therapy, and fluid sensing, but additionally new and upcoming applications such as drug delivery, photoacoustic imaging, thermoacoustic imaging, biometrics, and intrabody communication. By including a device comparison chart for different applications, this review aims to assist microelectromechanical systems (MEMS) designers that work with PMUTs by providing a benchmark for recent research works. Furthermore, it puts forth a discussion on the current challenges being faced by PMUTs in the biomedical field, current and likely future research trends, and opportunities for PMUT development areas, as well as sharing the opinions and predictions of the authors on the state of this technology as a whole. The review aims to be a comprehensive introduction to these topics without diving excessively deep into existing literature.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 6","pages":"622-637"},"PeriodicalIF":3.6,"publicationDate":"2024-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140688989","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 : 2024-04-10DOI: 10.1109/TUFFC.2024.3387351
Minho Song;Oleg A. Sapozhnikov;Vera A. Khokhlova;Tatiana D. Khokhlova
Pulsed high-intensity focused ultrasound (pHIFU) can induce sparse de novo inertial cavitation without the introduction of exogenous contrast agents, promoting mild mechanical disruption in targeted tissue. Because the bubbles are small and rapidly dissolve after each HIFU pulse, mapping transient bubbles and obtaining real-time quantitative metrics correlated with tissue damage are challenging. Prior work introduced Bubble Doppler, an ultrafast power Doppler imaging method as a sensitive means to map cavitation bubbles. The main limitation of that method was its reliance on conventional wall filters used in Doppler imaging and its optimization for imaging blood flow rather than transient scatterers. This study explores Bubble Doppler enhancement using dynamic mode decomposition (DMD) of a matrix created from a Doppler ensemble for mapping and extracting the characteristics of transient cavitation bubbles. DMD was first tested in silico with a numerical dataset mimicking the spatiotemporal characteristics of backscattered signal from tissue and bubbles. The performance of DMD filter was compared to other widely used Doppler wall filter-singular value decomposition (SVD) and infinite impulse response (IIR) high-pass filter. DMD was then applied to an ex vivo tissue dataset where each HIFU pulse was immediately followed by a plane wave Doppler ensemble. In silico DMD outperformed SVD and IIR high-pass filter and ex vivo provided physically interpretable images of the modes associated with bubbles and their corresponding temporal decay rates. These DMD modes can be trackable over the duration of pHIFU treatment using k-means clustering method, resulting in quantitative indicators of treatment progression.
{"title":"Dynamic Mode Decomposition for Transient Cavitation Bubbles Imaging in Pulsed High-Intensity Focused Ultrasound Therapy","authors":"Minho Song;Oleg A. Sapozhnikov;Vera A. Khokhlova;Tatiana D. Khokhlova","doi":"10.1109/TUFFC.2024.3387351","DOIUrl":"10.1109/TUFFC.2024.3387351","url":null,"abstract":"Pulsed high-intensity focused ultrasound (pHIFU) can induce sparse de novo inertial cavitation without the introduction of exogenous contrast agents, promoting mild mechanical disruption in targeted tissue. Because the bubbles are small and rapidly dissolve after each HIFU pulse, mapping transient bubbles and obtaining real-time quantitative metrics correlated with tissue damage are challenging. Prior work introduced Bubble Doppler, an ultrafast power Doppler imaging method as a sensitive means to map cavitation bubbles. The main limitation of that method was its reliance on conventional wall filters used in Doppler imaging and its optimization for imaging blood flow rather than transient scatterers. This study explores Bubble Doppler enhancement using dynamic mode decomposition (DMD) of a matrix created from a Doppler ensemble for mapping and extracting the characteristics of transient cavitation bubbles. DMD was first tested in silico with a numerical dataset mimicking the spatiotemporal characteristics of backscattered signal from tissue and bubbles. The performance of DMD filter was compared to other widely used Doppler wall filter-singular value decomposition (SVD) and infinite impulse response (IIR) high-pass filter. DMD was then applied to an ex vivo tissue dataset where each HIFU pulse was immediately followed by a plane wave Doppler ensemble. In silico DMD outperformed SVD and IIR high-pass filter and ex vivo provided physically interpretable images of the modes associated with bubbles and their corresponding temporal decay rates. These DMD modes can be trackable over the duration of pHIFU treatment using k-means clustering method, resulting in quantitative indicators of treatment progression.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 5","pages":"596-606"},"PeriodicalIF":3.6,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140718504","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 : 2024-04-04DOI: 10.1109/TUFFC.2024.3384815
Ufuk Soylu;Michael L. Oelze
A transfer function approach was recently demonstrated to mitigate data mismatches at the acquisition level for a single ultrasound scanner in deep learning (DL)-based quantitative ultrasound (QUS). As a natural progression, we further investigate the transfer function approach and introduce a machine-to-machine (M2M) transfer function, which possesses the ability to mitigate data mismatches at a machine level. This ability opens the door to unprecedented opportunities for reducing DL model development costs, enabling the combination of data from multiple sources or scanners, or facilitating the transfer of DL models between machines. We tested the proposed method utilizing a SonixOne machine and a Verasonics machine with an L9-4 array and an L11-5 array. We conducted two types of acquisitions to obtain calibration data: stable and free-hand, using two different calibration phantoms. Without the proposed method, the mean classification accuracy when applying a model on data acquired from one system to data acquired from another system was 50%, and the mean average area under the receiver operator characteristic (ROC) curve (AUC) was 0.405. With the proposed method, mean accuracy increased to 99%, and the AUC rose to the 0.999. Additional observations include the choice of the calibration phantom led to statistically significant changes in the performance of the proposed method. Moreover, robust implementation inspired by Wiener filtering provided an effective method for transferring the domain from one machine to another machine, and it can succeed using just a single calibration view. Lastly, the proposed method proved effective when a different transducer was used in the test machine.
{"title":"Machine-to-Machine Transfer Function in Deep Learning-Based Quantitative Ultrasound","authors":"Ufuk Soylu;Michael L. Oelze","doi":"10.1109/TUFFC.2024.3384815","DOIUrl":"10.1109/TUFFC.2024.3384815","url":null,"abstract":"A transfer function approach was recently demonstrated to mitigate data mismatches at the acquisition level for a single ultrasound scanner in deep learning (DL)-based quantitative ultrasound (QUS). As a natural progression, we further investigate the transfer function approach and introduce a machine-to-machine (M2M) transfer function, which possesses the ability to mitigate data mismatches at a machine level. This ability opens the door to unprecedented opportunities for reducing DL model development costs, enabling the combination of data from multiple sources or scanners, or facilitating the transfer of DL models between machines. We tested the proposed method utilizing a SonixOne machine and a Verasonics machine with an L9-4 array and an L11-5 array. We conducted two types of acquisitions to obtain calibration data: stable and free-hand, using two different calibration phantoms. Without the proposed method, the mean classification accuracy when applying a model on data acquired from one system to data acquired from another system was 50%, and the mean average area under the receiver operator characteristic (ROC) curve (AUC) was 0.405. With the proposed method, mean accuracy increased to 99%, and the AUC rose to the 0.999. Additional observations include the choice of the calibration phantom led to statistically significant changes in the performance of the proposed method. Moreover, robust implementation inspired by Wiener filtering provided an effective method for transferring the domain from one machine to another machine, and it can succeed using just a single calibration view. Lastly, the proposed method proved effective when a different transducer was used in the test machine.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 6","pages":"687-697"},"PeriodicalIF":3.6,"publicationDate":"2024-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10491402","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141300547","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 : 2024-04-03DOI: 10.1109/TUFFC.2024.3374508
{"title":"Retraction Notice: New Mega-Farad Ultracapacitors","authors":"","doi":"10.1109/TUFFC.2024.3374508","DOIUrl":"https://doi.org/10.1109/TUFFC.2024.3374508","url":null,"abstract":"","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 4","pages":"516-516"},"PeriodicalIF":3.6,"publicationDate":"2024-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10485188","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140345486","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 : 2024-04-02DOI: 10.1109/TUFFC.2024.3383660
William M. K. Pitman;Di Xiao;Billy Y. S. Yiu;Adrian J. Y. Chee;Alfred C. H. Yu
For high-frame-rate ultrasound imaging, it remains challenging to implement on compact systems as a sparse imaging configuration with limited array channels. One key issue is that the resulting image quality is known to be mediocre not only because unfocused plane-wave excitations are used but also because grating lobes would emerge in sparse-array configurations. In this article, we present the design and use of a new channel recovery framework to infer full-array plane-wave channel datasets for periodically sparse arrays that operate with as few as one-quarter of the full-array aperture. This framework is based on a branched encoder-decoder convolutional neural network (CNN) architecture, which was trained using full-array plane-wave channel data collected from human carotid arteries (59 864 training acquisitions; 5-MHz imaging frequency; 20-MHz sampling rate; plane-wave steering angles between −15° and 15° in 1° increments). Three branched encoder-decoder CNNs were separately trained to recover missing channels after differing degrees of channelwise downsampling (2, 3, and 4 times). The framework’s performance was tested on full-array and downsampled plane-wave channel data acquired from an in vitro point target, human carotid arteries, and human brachioradialis muscle. Results show that when inferred full-array plane-wave channel data were used for beamforming, spatial aliasing artifacts in the B-mode images were suppressed for all degrees of channel downsampling. In addition, the image contrast was enhanced compared with B-mode images obtained from beamforming with downsampled channel data. When the recovery framework was implemented on an RTX-2080 GPU, the three investigated degrees of downsampling all achieved the same inference time of 4 ms. Overall, the proposed framework shows promise in enhancing the quality of high-frame-rate ultrasound images generated using a sparse-array imaging setup.
{"title":"Branched Convolutional Neural Networks for Receiver Channel Recovery in High-Frame-Rate Sparse-Array Ultrasound Imaging","authors":"William M. K. Pitman;Di Xiao;Billy Y. S. Yiu;Adrian J. Y. Chee;Alfred C. H. Yu","doi":"10.1109/TUFFC.2024.3383660","DOIUrl":"10.1109/TUFFC.2024.3383660","url":null,"abstract":"For high-frame-rate ultrasound imaging, it remains challenging to implement on compact systems as a sparse imaging configuration with limited array channels. One key issue is that the resulting image quality is known to be mediocre not only because unfocused plane-wave excitations are used but also because grating lobes would emerge in sparse-array configurations. In this article, we present the design and use of a new channel recovery framework to infer full-array plane-wave channel datasets for periodically sparse arrays that operate with as few as one-quarter of the full-array aperture. This framework is based on a branched encoder-decoder convolutional neural network (CNN) architecture, which was trained using full-array plane-wave channel data collected from human carotid arteries (59 864 training acquisitions; 5-MHz imaging frequency; 20-MHz sampling rate; plane-wave steering angles between −15° and 15° in 1° increments). Three branched encoder-decoder CNNs were separately trained to recover missing channels after differing degrees of channelwise downsampling (2, 3, and 4 times). The framework’s performance was tested on full-array and downsampled plane-wave channel data acquired from an in vitro point target, human carotid arteries, and human brachioradialis muscle. Results show that when inferred full-array plane-wave channel data were used for beamforming, spatial aliasing artifacts in the B-mode images were suppressed for all degrees of channel downsampling. In addition, the image contrast was enhanced compared with B-mode images obtained from beamforming with downsampled channel data. When the recovery framework was implemented on an RTX-2080 GPU, the three investigated degrees of downsampling all achieved the same inference time of 4 ms. Overall, the proposed framework shows promise in enhancing the quality of high-frame-rate ultrasound images generated using a sparse-array imaging setup.","PeriodicalId":13322,"journal":{"name":"IEEE transactions on ultrasonics, ferroelectrics, and frequency control","volume":"71 5","pages":"558-571"},"PeriodicalIF":3.6,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10486969","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140754941","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号