[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

Search Results (330)

Search Parameters:
Keywords = ultrasound array

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
16 pages, 5459 KiB  
Article
Impact of Cell Layout on Bandwidth of Multi-Frequency Piezoelectric Micromachined Ultrasonic Transducer Array
by Wanli Yang, Huimin Li, Yuewu Gong, Zhuochen Wang, Xingli Xu, Xiaofan Hu, Pengfei Niu and Wei Pang
Micromachines 2025, 16(1), 49; https://doi.org/10.3390/mi16010049 - 31 Dec 2024
Viewed by 361
Abstract
Piezoelectric micromachined ultrasonic transducers (PMUTs) show considerable promise for application in ultrasound imaging, but the limited bandwidth of the traditional PMUTs largely affects the imaging quality. This paper focuses on how to arrange cells with different frequencies to maximize the bandwidth and proposes [...] Read more.
Piezoelectric micromachined ultrasonic transducers (PMUTs) show considerable promise for application in ultrasound imaging, but the limited bandwidth of the traditional PMUTs largely affects the imaging quality. This paper focuses on how to arrange cells with different frequencies to maximize the bandwidth and proposes a multi-frequency PMUT (MF-PMUT) linear array. Seven cells with gradually changing frequencies are arranged in a monotonic trend to form a unit, and 32 units are distributed across four lines, forming one element. To investigate how the arrangement of cells affects the bandwidth, three different arrays were designed according to the extent of unit aggregation from the same frequency. Underwater experiments were conducted to assess the acoustic performance, especially the bandwidth. We found that the densest arrangement of the same cells produced the largest bandwidth, achieving a 92% transmission bandwidth and a 50% burst-echo bandwidth at 6 MHz. The mechanism was investigated from the coupling point of view by finite element analysis and laser Doppler vibrometry, focusing on the cell displacements. The results demonstrated strong ultrasound coupling in the devices, resulting in larger bandwidths. To exploit the advanced bandwidth but reduce the crosstalk, grooves for isolation were fabricated between elements. This work proposes an effective strategy for developing advanced PMUT arrays that would benefit ultrasound imaging applications. Full article
(This article belongs to the Section A:Physics)
Show Figures

Figure 1

Figure 1
<p>Schematic diagram of the structure of a single cell and the geometric parameters of its simulation model.</p>
Full article ">Figure 2
<p>Arrangement (<b>a</b>–<b>c</b>) of cells with different resonant frequencies.</p>
Full article ">Figure 3
<p>(<b>a</b>) The FEA model for the PMUT array, displaying boundary conditions. The model is not depicted in proportion to show the configuration details. Simulation model mesh examples of the B-type combination: (<b>b</b>) free triangular mesh of the combination and (<b>c</b>) global mesh sweep.</p>
Full article ">Figure 4
<p>Displacement of the (<b>a</b>) A-type, (<b>b</b>) B-type, and (<b>c</b>) C-type combinations under a 1 Vpp excitation at a frequency of 4 MHz applied to the top electrode.</p>
Full article ">Figure 5
<p>Resonant frequencies of cells corresponding to various cavity sizes in the simulation of a single cell and the A-type, B-type, and C-type combinations.</p>
Full article ">Figure 6
<p>Displacement of each cell: (<b>a</b>) when the cell with a resonant frequency of about 4 MHz is in resonance; (<b>b</b>) when the cell with a resonant frequency of about 6 MHz in resonance; (<b>c</b>) when the cell with a resonant frequency of about 8 MHz in resonance.</p>
Full article ">Figure 7
<p>Optical images of (<b>a</b>) A-type, (<b>b</b>) B-type, and (<b>c</b>) C-type multi-frequency PMUT arrays. (<b>d</b>) A cross-section of a cell and (<b>e</b>) its enlarged view at the edge of the cavity.</p>
Full article ">Figure 8
<p>Impedance and phase characteristics of multi-frequency PMUT arrays under the (<b>a</b>) A-type, (<b>b</b>) B-type, and (<b>c</b>) C-type designs; (<b>d</b>) comparison schematic of actual measured resonant frequency and simulation results.</p>
Full article ">Figure 9
<p>Waveform of the signal received by the hydrophone from the PMUT array transmission.</p>
Full article ">Figure 10
<p>The amplitude–frequency variation curve of the received signal during the frequency sweep transmission of the (<b>a</b>) A-type, (<b>b</b>) B-type, and (<b>c</b>) C-type PMUT array.</p>
Full article ">Figure 11
<p>Time-domain response signals and their spectra of the 1-cycle burst echo of the (<b>a</b>) A-type, (<b>b</b>) B-type, and (<b>c</b>) C-type PMUT arrays under single-pulse sine wave excitation.</p>
Full article ">Figure 12
<p>Diagram of the use of an LDV to measure crosstalk.</p>
Full article ">Figure 13
<p>Comparison of normalized vibration displacement between excitation elements and their neighbor elements: comparison of arrays with different arrangements.</p>
Full article ">Figure 14
<p>Comparison of normalized vibration displacement between excitation elements and their neighbor elements: comparison of arrays with isolation grooves of different depths.</p>
Full article ">Figure 15
<p>Time-domain response signals and their FFT spectra of PMUT arrays with isolation grooves of different depths under single-pulse sinusoidal signal excitation: (<b>a</b>) grooves etched to a depth of 2 μm; (<b>b</b>) grooves etched to a depth of 5 μm; (<b>c</b>) grooves etched to a depth of 20 μm.</p>
Full article ">
15 pages, 4130 KiB  
Article
Delivering Volumetric Hyperthermia to Head and Neck Cancer Patient-Specific Models Using an Ultrasound Spherical Random Phased Array Transducer
by Muhammad Zubair, Imad Uddin, Robert Dickinson and Chris J. Diederich
Bioengineering 2025, 12(1), 14; https://doi.org/10.3390/bioengineering12010014 - 28 Dec 2024
Viewed by 361
Abstract
In exploring adjuvant therapies for head and neck cancer, hyperthermia (40–45 °C) has shown efficacy in enhancing chemotherapy and radiation, as well as the delivery of liposomal drugs. Current hyperthermia treatments, however, struggle to reach large deep tumors uniformly and non-invasively. This study [...] Read more.
In exploring adjuvant therapies for head and neck cancer, hyperthermia (40–45 °C) has shown efficacy in enhancing chemotherapy and radiation, as well as the delivery of liposomal drugs. Current hyperthermia treatments, however, struggle to reach large deep tumors uniformly and non-invasively. This study investigates the feasibility of delivering targeted uniform hyperthermia deep into the tissue using a non-invasive ultrasound spherical random phased array transducer. Simulations in 3D patient-specific models for thyroid and oropharyngeal cancers assessed the transducer’s proficiency. The transducer consisting of 256 elements randomly positioned on a spherical shell, operated at a frequency of 1 MHz with various phasing schemes and power modulations to analyze 40, 41, and 43 °C isothermal volumes and the penetration depth of the heating volume, along with temperature uniformity within the target area using T10, T50, and T90 temperatures, across different tumor models. Intensity distributions and volumetric temperature contours were calculated to define moderate hyperthermia boundaries. The results indicated the array’s ability to produce controlled heating volumes from 1 to 48 cm3 at 40 °C, 0.35 to 27 cm3 at 41 °C, and 0.1 to 8 cm3 at 43 °C. The heating depths ranged from 7 to 39 mm minimum and 52 to 59 mm maximum, measured from the skin’s inner surface. The transducer, with optimal phasing and water-cooled bolus, confined the heating to the targeted regions effectively. Multifocal sonications also improved the heating homogeneity, reducing the length-to-diameter ratio by 38% when using eight foci versus a single one. This approach shows potential for treating a range of tumors, notably deep-seated and challenging oropharyngeal cancers. Full article
(This article belongs to the Section Biomedical Engineering and Biomaterials)
Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Front face of the phased array transducer showing element distribution on the surface of the array, with four quadrants and a central hole for holding an imaging transducer, (<b>b</b>) 3D patient-specific model and bio-acoustic thermal simulation for targeted volumetric hyperthermia delivery with a random phased array transducer (1 MHz, 256 elements, 170 mm outer diameter (OD), 130 mm focal depth), positioned to target energy at a tumor in the neck. A 3D model showing the tumor and surrounding anatomy including skull bone, trachea, soft tissue, fat, and skin.</p>
Full article ">Figure 2
<p>A cross-sectional plane from patient-specific models used in this study (<b>a</b>) Model 1, (<b>b</b>) Model 2, and (<b>c</b>) Model 3 showing the tumor and surrounding anatomy including bone (B), trachea (Tr), skin (S), and tumor (Tu). Models 1 and 3 are superficial tumors with varying volumes, whereas Model 2 is a deep oropharyngeal tumor.</p>
Full article ">Figure 3
<p>Pressure profile [MPa] projected from the phased array transducer to the patient-specific anatomical models focused at the geometric center of the array inside the tumor for a single focus in model 1 (<b>a</b>,<b>b</b>), four simultaneous foci at (x,y) = ± 2.5 mm (<b>c</b>,<b>d</b>), and eight simultaneous foci at (x,y) = ±5 mm and ±10 mm, in the axial (<b>a</b>,<b>c</b>,<b>e</b>) and transverse (<b>b</b>,<b>d</b>,<b>f</b>) planes.</p>
Full article ">Figure 4
<p>Acoustic and biothermal simulations in a 3D patient specific model (Model 1): (<b>a</b>) normalized pressure distribution projected from the phased array transducer to the model with 0.17 MPa surface pressure equivalent to 1 W/cm<sup>2</sup> input surface intensity to generate four simultaneous foci (x = ±5 mm, y = ±5 mm) at the geometric center of the array (z = 130 mm); (<b>b</b>) distribution of temperature along a central axial plane; (<b>c</b>) temperature distribution across a transverse plane at the geometric center of the transducer (130 mm depth from center of transducer); (<b>d</b>) iso-temperature volume of 41 °C within the target region. Tu: tumor, B: bone, Tr: trachea.</p>
Full article ">Figure 5
<p>Acoustic and biothermal simulations in a 3D patient specific model (Model 2): (<b>a</b>) normalized pressure distribution projected from the phased array transducer to the model with 0.17 MPa surface pressure equivalent to 1 W/cm<sup>2</sup> input surface intensity for generating a single focus at the geometric center; (<b>b</b>) distribution of temperature along a central axial plane; (<b>c</b>) temperature distribution across a transverse plane at the geometric center of the transducer (130 mm depth from center of transducer) overlaid on the voxelized image with the tumor and surrounding anatomy shown; (<b>d</b>) iso-temperature volume of 41 °C within the target region. S: skin, Tu: tumor, B: bone, Tr: trachea.</p>
Full article ">Figure 6
<p>Acoustic and biothermal simulations in a 3D patient specific model (Model 3): (<b>a</b>) normalized pressure distribution projected from the phased array transducer to the model with 0.17 MPa surface pressure equivalent to 1 W/cm<sup>2</sup> input surface intensity to generate eight simultaneous foci at the geometric center of the array (z = 130 mm); (<b>b</b>) temperature distribution across a central axial plane with the target delineated; (<b>c</b>) temperature distribution across a transverse plane at the geometric center of the transducer (130 mm depth from center of transducer); (<b>d</b>) iso-temperature volume of 41 °C within the target region. Tu: tumor, B: bone, Tr: trachea.</p>
Full article ">
14 pages, 1190 KiB  
Systematic Review
The Efficacy of Transversus Abdominis Plane (TAP) Blocks When Completed by Anesthesiologists Versus by Surgeons: A Systematic Review and Meta-Analysis
by Dylan Irvine, Christopher Rennie, Emily Coughlin, Imani Thornton, Rahul Mhaskar and Jeffrey Huang
Healthcare 2024, 12(24), 2586; https://doi.org/10.3390/healthcare12242586 - 22 Dec 2024
Viewed by 554
Abstract
Background/Objectives: Current literature has demonstrated the benefits of transversus abdominis plane (TAP) blocks for reducing postoperative pain and opioid consumption for an array of surgical procedures. Some randomized controlled trials and retrospective studies have compared ultrasound guidance TAP blocks completed by anesthesiologists [...] Read more.
Background/Objectives: Current literature has demonstrated the benefits of transversus abdominis plane (TAP) blocks for reducing postoperative pain and opioid consumption for an array of surgical procedures. Some randomized controlled trials and retrospective studies have compared ultrasound guidance TAP blocks completed by anesthesiologists (US-TAP) to laparoscopic guidance TAP blocks completed by surgeons (LAP-TAP). However, the findings of these studies have not been consolidated to improve recommendations and patient outcomes. Our objective is to consolidate and summarize current literature regarding the efficacy of TAP blocks for postoperative pain control and opioid consumption when performed with ultrasound guidance (US-TAP, compared to laparoscopic guidance (LAP-TAP). Methods: We performed a systematic review and meta-analysis of RCTs and retrospective studies to evaluate US-TAP versus LAP-TAP blocks for postoperative pain control and opioid consumption. We searched PubMed/MEDLINE, CINAHL, Cochrane, and Web of Science databases for all articles meeting the search criteria until the time of article extraction in February 2024. The primary outcome variables were postoperative pain scores and opioid consumption. The secondary outcome variables were complications, time taken to perform the block, length of stay (LOS) in the hospital, and cost of performing the block. Results: Of the 1673 articles initially identified, 18 studies met the inclusion criteria for evaluation. Of the included studies, 88.9% and 77.8% found no significant difference in postoperative pain scores or opioid consumption, respectively, between US-TAP and LAP-TAP groups. Six studies (33.3%) found that LAP-TAP was faster to perform than US-TAP. Meta-analysis demonstrated no statistically significant differences in postoperative pain scores or opioid consumption between groups but showed that block times were significantly longer in the US-TAP group. Conclusions: US-TAP and LAP-TAP blocks may be equivocal in terms of reducing postoperative pain and opioid consumption. LAP-TAPs may be less time-consuming and more cost-effective and viable alternatives to US-TAP blocks in the perioperative setting. Full article
(This article belongs to the Section Pain Management)
Show Figures

Figure 1

Figure 1
<p>PRISMA flow chart—identification included studies from a search of databases.</p>
Full article ">Figure 2
<p>Postoperative pain scores between US-TAP and LAP-TAP groups [<a href="#B17-healthcare-12-02586" class="html-bibr">17</a>,<a href="#B20-healthcare-12-02586" class="html-bibr">20</a>,<a href="#B22-healthcare-12-02586" class="html-bibr">22</a>,<a href="#B23-healthcare-12-02586" class="html-bibr">23</a>,<a href="#B25-healthcare-12-02586" class="html-bibr">25</a>,<a href="#B26-healthcare-12-02586" class="html-bibr">26</a>,<a href="#B28-healthcare-12-02586" class="html-bibr">28</a>,<a href="#B30-healthcare-12-02586" class="html-bibr">30</a>].</p>
Full article ">Figure 3
<p>Postoperative opioid consumption between US-TAP and LAP-TAP groups [<a href="#B20-healthcare-12-02586" class="html-bibr">20</a>,<a href="#B22-healthcare-12-02586" class="html-bibr">22</a>,<a href="#B23-healthcare-12-02586" class="html-bibr">23</a>,<a href="#B25-healthcare-12-02586" class="html-bibr">25</a>,<a href="#B26-healthcare-12-02586" class="html-bibr">26</a>,<a href="#B28-healthcare-12-02586" class="html-bibr">28</a>,<a href="#B31-healthcare-12-02586" class="html-bibr">31</a>,<a href="#B32-healthcare-12-02586" class="html-bibr">32</a>].</p>
Full article ">Figure 4
<p>Block completion time between US-TAP and LAP-TAP groups [<a href="#B17-healthcare-12-02586" class="html-bibr">17</a>,<a href="#B23-healthcare-12-02586" class="html-bibr">23</a>,<a href="#B26-healthcare-12-02586" class="html-bibr">26</a>,<a href="#B28-healthcare-12-02586" class="html-bibr">28</a>,<a href="#B30-healthcare-12-02586" class="html-bibr">30</a>].</p>
Full article ">
30 pages, 1605 KiB  
Review
A Practical Clinical Approach to Navigate Pulmonary Embolism Management: A Primer and Narrative Review of the Evolving Landscape
by Kevin Benavente, Bradley Fujiuchi, Hafeez Ul Hassan Virk, Pavan K. Kavali, Walter Ageno, Geoffrey D. Barnes, Marc Righini, Mahboob Alam, Rachel P. Rosovsky and Chayakrit Krittanawong
J. Clin. Med. 2024, 13(24), 7637; https://doi.org/10.3390/jcm13247637 - 15 Dec 2024
Viewed by 803
Abstract
Advances in imaging, pharmacological, and procedural technologies have rapidly expanded the care of pulmonary embolism. Earlier, more accurate identification and quantification has enhanced risk stratification across the spectrum of the disease process, with a number of clinical tools available to prognosticate outcomes and [...] Read more.
Advances in imaging, pharmacological, and procedural technologies have rapidly expanded the care of pulmonary embolism. Earlier, more accurate identification and quantification has enhanced risk stratification across the spectrum of the disease process, with a number of clinical tools available to prognosticate outcomes and guide treatment. Direct oral anticoagulants have enabled a consistent and more convenient long-term therapeutic option, with a greater shift toward outpatient treatment for a select group of low-risk patients. The array of catheter-directed therapies now available has contributed to a more versatile and nuanced armamentarium of treatment options, including ultrasound-facilitated thrombolysis and mechanical thrombectomy. Research into supportive care for pulmonary embolism have explored the optimal use of vasopressors and volume resuscitation, as well as utilization of various devices, including right ventricular mechanical support and extracorporeal membrane oxygenation. Even in the realm of surgery, outcomes have steadily improved in experienced centers. This rapid expansion in diagnostic and therapeutic data has necessitated implementation of pulmonary embolism response teams to better interpret the available evidence, manage the utilization of advanced therapies, and coordinate multidisciplinary care. We provide a narrative review of the risk stratification and management of pulmonary embolism, with a focus on structuralizing the multidisciplinary approach and organizing the literature on new and emerging therapies. Full article
Show Figures

Figure 1

Figure 1
<p>Representation of the cyclic hemodynamic effects of acute pulmonary embolism.</p>
Full article ">Figure 2
<p>A summary of the evidence for the management of intermediate high-risk PE.</p>
Full article ">Figure 3
<p>A summary of the evidence for the management of high-risk PE.</p>
Full article ">Figure 4
<p>Trials examining the effects of CDT and anticoagulation against anticoagulation alone that are actively enrolling.</p>
Full article ">
16 pages, 3435 KiB  
Article
Ultrasound Corrosion Mapping on Hot Stainless Steel Surfaces
by Jan Lean Tai, Mohamed Thariq Hameed Sultan, Farah Syazwani Shahar, Andrzej Łukaszewicz, Zbigniew Oksiuta and Rafał Grzejda
Metals 2024, 14(12), 1425; https://doi.org/10.3390/met14121425 - 12 Dec 2024
Viewed by 498
Abstract
This study investigates the application of Phased Array Corrosion Mapping (PACM) as a non-destructive testing (NDT) method for detecting and monitoring corrosion growth on hot stainless steel (SS) surfaces, specifically focusing on SS 304 and SS 316. Conducted across a temperature range of [...] Read more.
This study investigates the application of Phased Array Corrosion Mapping (PACM) as a non-destructive testing (NDT) method for detecting and monitoring corrosion growth on hot stainless steel (SS) surfaces, specifically focusing on SS 304 and SS 316. Conducted across a temperature range of 30 °C to 250 °C, the research evaluates the effectiveness of PACM in high-temperature environments typical of the petrochemical industry. Experiments were conducted using specimens with machined slots and flat-bottom holes (FBHs) to simulate corrosion defects. The results demonstrate that PACM effectively detects and maps corrosion indicators, with color-coded C-scan data facilitating easy interpretation. Temperature variations significantly influenced ultrasound signal characteristics, leading to observable changes in FBH indications, particularly at elevated temperatures. Increased ultrasound attenuation necessitated adjustments in decibel settings to maintain accuracy. SS 304 and SS 316 exhibited distinct responses to temperature changes, with SS 316 showing higher dB values and unique signal behaviors, including increased scattering and noise echoes at elevated temperatures. Detected depths for slots and FBHs correlated closely with designed depths, with deviations generally less than 0.5 mm; however, some instances showed deviations exceeding 2 mm, underscoring the need for careful interpretation. At temperatures above 230 °C, the disbanding of probe elements led to weak or absent signals, complicating data interpretation and requiring adjustments in testing protocols. This study highlights the feasibility and effectiveness of PACM for corrosion detection on hot SS surfaces, providing critical insights into material behavior under thermal conditions. Future research should include physical examination of samples using Scanning Electron Microscopy (SEM) to validate and enhance the reliability of the findings. The integration of non-contact NDT methods and optimization of calibration techniques are essential for improving PACM performance at elevated temperatures. Full article
(This article belongs to the Section Corrosion and Protection)
Show Figures

Figure 1

Figure 1
<p>Schematic of the test specimen design with identification number.</p>
Full article ">Figure 2
<p>(<b>a</b>) Corrosion mapping data for SS 304 at 30 °C; (<b>b</b>) 100 °C; (<b>c</b>) 110 °C; (<b>d</b>) 200 °C; (<b>e</b>) 230 °C; (<b>f</b>) 250 °C.</p>
Full article ">Figure 3
<p>(<b>a</b>) Corrosion mapping data for SS 316 at 30 °C; (<b>b</b>) 100 °C; (<b>c</b>) 190 °C; (<b>d</b>) 210 °C; (<b>e</b>) 230 °C; (<b>f</b>) 250 °C.</p>
Full article ">Figure 3 Cont.
<p>(<b>a</b>) Corrosion mapping data for SS 316 at 30 °C; (<b>b</b>) 100 °C; (<b>c</b>) 190 °C; (<b>d</b>) 210 °C; (<b>e</b>) 230 °C; (<b>f</b>) 250 °C.</p>
Full article ">Figure 4
<p>Indication dimension measurement.</p>
Full article ">
51 pages, 8216 KiB  
Article
Optimization of Combined Ultrasound and Microwave-Assisted Extraction for Enhanced Bioactive Compounds Recovery from Four Medicinal Plants: Oregano, Rosemary, Hypericum, and Chamomile
by Konstantina Theodora Laina, Christina Drosou, Chrysanthos Stergiopoulos, Panagiota Maria Eleni and Magdalini Krokida
Molecules 2024, 29(23), 5773; https://doi.org/10.3390/molecules29235773 - 6 Dec 2024
Viewed by 623
Abstract
This study presents the synergistic application of ultrasound- and microwave-assisted extraction (UAE–MAE) as a novel and efficient method for recovering bioactive compounds from the medicinal plants oregano, rosemary, Hypericum perforatum, and chamomile. Extraction parameters, including microwave (MW) power, ultrasound (US) power, and [...] Read more.
This study presents the synergistic application of ultrasound- and microwave-assisted extraction (UAE–MAE) as a novel and efficient method for recovering bioactive compounds from the medicinal plants oregano, rosemary, Hypericum perforatum, and chamomile. Extraction parameters, including microwave (MW) power, ultrasound (US) power, and extraction time, were optimized using the response surface methodology (RSM), with ethanol as the solvent. Extracts were evaluated for total phenolic content (TPC) via the Folin–Ciocalteu method and antioxidant activity (IC50) using the DPPH assay. High-performance liquid chromatography with diode array detection (HPLC–DAD) identified the main bioactive compounds contributing to their antioxidant and therapeutic potential. The optimized UAE–MAE conditions enhanced phenolic recovery and antioxidant potential across all plants. Notably, Hypericum perforatum exhibited the highest TPC (53.7 mg GAE/g) and strongest antioxidant activity (IC50 29.8 mg extract/g) under 200 W MW, 450 W US, and 12 min, yielding 14.5%. Rosemary achieved the highest yield (23.36%) with a TPC of 26.35 mg GAE/g and an IC50 of 40.75 mg extract/g at 200 W MW, 700 W US, and 8 min. Oregano’s optimal conditions (500 W MW, 700 W US, 12 min) produced a TPC of 34.99 mg GAE/g and an IC50 of 50.31 mg extract/g. Chamomile extracts demonstrated lower phenolic content and antioxidant activity but achieved significant yields under 500 W MW, 700 W US, and 5 min. This study highlights UAE–MAE’s superior efficiency, showcasing its potential to maximize phenolic recovery sustainably, making it a promising technique for industrial and therapeutic applications. Full article
(This article belongs to the Special Issue Current Emerging Trends of Extraction and Encapsulation in Food)
Show Figures

Figure 1

Figure 1
<p>Plots of predicted versus actual values for the <span class="html-italic">Y</span> (%) of oregano extracts.</p>
Full article ">Figure 2
<p>Plots of predicted versus actual values for the IC50 (mg extract/g raw material) of UAE of oregano extracts.</p>
Full article ">Figure 3
<p>Plots of predicted versus actual values for the TPC (mg GAE/g raw material) of oregano extracts.</p>
Full article ">Figure 4
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on the extraction yield (%) of oregano extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W); (<b>c</b>) US power vs. extraction time (MW power: 200 W).</p>
Full article ">Figure 5
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on IC<sub>50</sub> (mg extract/g raw material) values of oregano extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W).</p>
Full article ">Figure 6
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on TPC (mg GAE/g raw material) values of oregano extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W).</p>
Full article ">Figure 7
<p>Plots of predicted versus actual values for the Y (%) of rosemary extracts.</p>
Full article ">Figure 8
<p>Plots of predicted versus actual values for the IC<sub>50</sub> (mg extract/g raw material) of UAE of rosemary extracts.</p>
Full article ">Figure 9
<p>Plots of predicted versus actual values for the TPC (mg GAE/g raw material) of rosemary extracts.</p>
Full article ">Figure 10
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on IC<sub>50</sub> (mg extract/g raw material) values of rosemary extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W); (<b>c</b>) US power vs. extraction time (MW power: 200 W).</p>
Full article ">Figure 11
<p>Response surface and contour plots showing the effects of MW power and US power on TPC (mg GAE/g raw material) values of rosemary extracts.</p>
Full article ">Figure 12
<p>Plots of predicted versus actual values for the <span class="html-italic">Y</span> (%) of hypericum extracts.</p>
Full article ">Figure 13
<p>Plots of predicted versus actual values for the IC<sub>50</sub> (mg extract/g raw material) of UAE of hypericum extracts.</p>
Full article ">Figure 14
<p>Plots of predicted versus actual values for the TPC (mg GAE/g raw material) of hypericum extracts.</p>
Full article ">Figure 15
<p>Response surface and contour plots showing the effects of MW power and US power on the extraction yield (%) of hypericum extracts.</p>
Full article ">Figure 16
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on IC<sub>50</sub> (mg extract/g raw material) values of hypericum extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W); (<b>c</b>) US power vs. extraction time (MW power: 200 W).</p>
Full article ">Figure 17
<p>Response surface and contour plots showing the effects of MW power and US power on TPC (mg GAE/g raw material) values of hypericum extracts.</p>
Full article ">Figure 18
<p>Plots of predicted versus actual values for the <span class="html-italic">Y</span> (%) of chamomile extracts.</p>
Full article ">Figure 19
<p>Plots of predicted versus actual values for the IC<sub>50</sub> (mg extract/g raw material) of UAE of chamomile extracts.</p>
Full article ">Figure 20
<p>Plots of predicted versus actual values for the TPC (mg GAE/g raw material) of chamomile extracts.</p>
Full article ">Figure 21
<p>Response surface and contour plots showing the effects of MW power and US power on the extraction yield (%) of chamomile extracts.</p>
Full article ">Figure 22
<p>Response surface and contour plots showing the effects of MW power, US power, extraction time on IC<sub>50</sub> (mg extract/g raw material) values of chamomile extracts. (<b>a</b>) MW power vs. US power (extraction time: 8 min); (<b>b</b>) MW power vs. extraction time (US power: 450 W).</p>
Full article ">Figure 23
<p>Response surface and contour plots showing the effects of MW power and US power on TPC (mg GAE/g raw material) values of chamomile extracts.</p>
Full article ">Figure 24
<p>HPLC chromatograms of oregano extract recorded at 280 nm (for detection of rosmarinic acid and carvacrol).</p>
Full article ">Figure 25
<p>HPLC chromatograms of rosemary extract recorded at 280 nm (for detection of rosmarinic acid, carnosol, and carnosic acid).</p>
Full article ">Figure 26
<p>HPLC chromatograms of hypericum extract recorded at: (<b>a</b>) 272 nm (for detection of hyperforin), and (<b>b</b>) 520 nm (for detection of hypericin).</p>
Full article ">Figure 27
<p>HPLC chromatogram of chamomile extract recorded at 360 nm (for detection of rutin and quercetin).</p>
Full article ">
18 pages, 7815 KiB  
Article
Feasibility of Backscattering Coefficient Evaluation of Soft Tissue Using High-Frequency Annular Array Probe
by Jungtaek Choi, Jeffrey A. Ketterling, Jonathan Mamou, Cameron Hoerig, Shinnosuke Hirata, Kenji Yoshida and Tadashi Yamaguchi
Sensors 2024, 24(22), 7118; https://doi.org/10.3390/s24227118 - 5 Nov 2024
Viewed by 641
Abstract
The objective of this work is to address the need for versatile and effective tissue characterization in abdominal ultrasound diagnosis using a simpler system. We evaluated the backscattering coefficient (BSC) of several tissue-mimicking phantoms utilizing three different ultrasonic probes: a single-element transducer, a [...] Read more.
The objective of this work is to address the need for versatile and effective tissue characterization in abdominal ultrasound diagnosis using a simpler system. We evaluated the backscattering coefficient (BSC) of several tissue-mimicking phantoms utilizing three different ultrasonic probes: a single-element transducer, a linear array probe for clinical use, and a laboratory-made annular array probe. The single-element transducer, commonly used in developing fundamental BSC evaluation methods, served as a benchmark. The linear array probe provided a clinical comparison, while the annular array probe was tested for its potential in high-frequency and high-resolution ultrasonic observations. Our findings demonstrate that the annular array probe meets clinical demands by providing accurate BSC measurements, showcasing its capability for high-frequency and high-resolution imaging with a simpler, more versatile system. Full article
(This article belongs to the Special Issue Biomedical Imaging, Sensing and Signal Processing)
Show Figures

Figure 1

Figure 1
<p>Appearance and configuration of the annular array probe.</p>
Full article ">Figure 2
<p>Schematic of synthetic focusing.</p>
Full article ">Figure 3
<p>Difference in delay time due to interpolation.</p>
Full article ">Figure 4
<p>Averaged amplitude envelopes of phantoms of basic study.</p>
Full article ">Figure 5
<p>B–mode images of phantoms.</p>
Full article ">Figure 6
<p>Frequency characteristics of phantoms of basic study.</p>
Full article ">Figure 7
<p>Evaluated BSCs of phantoms.</p>
Full article ">Figure 8
<p>Averaged amplitude envelopes of phantoms of comparative study.</p>
Full article ">Figure 9
<p>B–mode images of phantoms acquired using different ultrasound probes.</p>
Full article ">Figure 10
<p>Frequency characteristics of phantoms observed using different ultrasound probes.</p>
Full article ">Figure 11
<p>Evaluated BSCs of phantoms ((<b>a</b>) single element transducer; (<b>b</b>) annular array probe).</p>
Full article ">Figure 12
<p>Deviation of evaluated BSCs from theoretical values in comparative study within single element transducer and annular array probe.</p>
Full article ">Figure 13
<p>Frequency characteristics of phantoms of comparative study.</p>
Full article ">Figure 14
<p>Evaluated BSCs of phantoms ((<b>a</b>) annular array probe; (<b>b</b>) linear array probe).</p>
Full article ">Figure 15
<p>Deviation of evaluated BSCs from theoretical values in comparative study within annular array probe and linear array probe.</p>
Full article ">
12 pages, 3235 KiB  
Article
Dynamic Acoustic Holography: One-Shot High-Precision and High-Information Methodology
by Zhaoxi Li, Yiheng Yang, Qi Lu, Xiongwei Wei, Chenxue Hou, Yi Quan, Xiaozhou Lü, Weimin Bao, Yintang Yang and Chunlong Fei
Micromachines 2024, 15(11), 1316; https://doi.org/10.3390/mi15111316 - 29 Oct 2024
Viewed by 988
Abstract
Acoustic holography technology is widely used in the field of ultrasound due to its capability to achieve complex acoustic fields. The traditional acoustic holography method based on single-phase holograms is limited due to its inability to complete acoustic field control with high dynamics [...] Read more.
Acoustic holography technology is widely used in the field of ultrasound due to its capability to achieve complex acoustic fields. The traditional acoustic holography method based on single-phase holograms is limited due to its inability to complete acoustic field control with high dynamics and accuracy. Here, we propose a method for constructing an acoustic holographic model, introducing an ultrasonic array to provide dynamic amplitude control degrees of freedom, and combining the dynamically controllable ultrasonic array and high-precision acoustic hologram to achieve the highest acoustic field accuracy and dynamic range. This simulation method has been proven to be applicable to both simple linear patterns and complex surface patterns. Moreover, it is possible to reconstruct the degree of freedom of the target plane amplitude effectively and achieve a breakthrough in high information content. This high-efficiency acoustic field control capability has potential applications in ultrasound imaging, acoustic tweezers, and neuromodulation. Full article
Show Figures

Figure 1

Figure 1
<p>Diagram of the control system of the electric machine and the equipment to be tested.</p>
Full article ">Figure 2
<p>(<b>a</b>) Ideal acoustic field in the target plane, with the “XDU” portion having an amplitude of “1” and the rest being “0”. (<b>b</b>,<b>c</b>), respectively, show the target surface acoustic field distribution at a depth of 7 mm obtained by IASA after keeping the array element amplitudes the same and optimizing them.</p>
Full article ">Figure 3
<p>Acoustic holograms formed and target depth acoustic fields generated after random selection of array elements. (<b>a</b>) Two ideal target plane acoustic fields, with the “XDU” and “SME” portion having an amplitude of “1” and the rest being “0”. (<b>b</b>,<b>c</b>), respectively, show the acoustic hologram phase and acoustic field generated when the target acoustic fields are “XDU” and “SME” with equal excitation amplitudes for each array element. (<b>d</b>,<b>e</b>), respectively, show the final acoustic hologram phase and target acoustic field formed after amplitude optimization when the target acoustic fields are “XDU” and “SME”.</p>
Full article ">Figure 4
<p>(<b>a</b>) Ideal acoustic field on the target plane, where the amplitudes represented by the four “petals” increase uniformly clockwise from the 6 o’clock direction. (<b>b</b>,<b>c</b>), respectively, show the target plane acoustic field distribution at a depth of 7mm obtained by inverse acoustic source amplitude (IASA) after maintaining uniform amplitudes for array elements and after amplitude optimization for array elements. (<b>d</b>) Comparison of reconstruction efficiency and reconstruction similarity after optimization.</p>
Full article ">Figure 5
<p>(<b>a</b>) Ideal acoustic field on the target plane. (<b>b</b>) Acoustic hologram phase and acoustic field formed when array elements are activated at regular intervals with uniform amplitudes. (<b>c</b>) Acoustic hologram phase and acoustic field formed when half of the array elements are randomly selected and activated with uniform amplitudes. (<b>d</b>) Acoustic hologram phase and acoustic field formed after array element amplitude optimization followed by random selection of 512 array elements. (<b>e</b>) Comparison of reconstruction efficiency and reconstruction similarity between two schemes.</p>
Full article ">Figure 6
<p>(<b>a</b>) Ideal acoustic field for each target plane. (<b>b</b>) Distribution of element amplitudes optimized for each specific array element. (<b>c</b>) The corresponding phase of acoustic holography after optimization. (<b>d</b>) Dynamic acoustic field animation achieved by creating an acoustic hologram by superimposing the three phases and using a phased array.</p>
Full article ">
16 pages, 5483 KiB  
Article
Periodically Sinusoidal Magnetic Stray Field and Improved Film Quality of CoMnP Micro-Magnet Arrays for Magnetic Encoders by Electrodeposition with the Assistance of Ultrasound
by Geng-Hua Xu, Jung-Yen Chang, Hsiang-Chun Hsueh and Chiao-Chi Lin
Coatings 2024, 14(10), 1340; https://doi.org/10.3390/coatings14101340 - 21 Oct 2024
Viewed by 2833
Abstract
Magnetic encoders are composed of a magnetic sensor, a hard magnetic recording medium and a signal processing circuit. Electrodeposited micro-magnet arrays produced by micro-fabrication are promising recording media for enhancing encoder performance. However, two major engineering issues have yet to be resolved. One [...] Read more.
Magnetic encoders are composed of a magnetic sensor, a hard magnetic recording medium and a signal processing circuit. Electrodeposited micro-magnet arrays produced by micro-fabrication are promising recording media for enhancing encoder performance. However, two major engineering issues have yet to be resolved. One issue is an unknown relationship between the feature sizes of micro-magnet arrays and their stray field shapes, and another issue is the formation of micro-cracks due to the built-up residual stresses of thick films. In this study, we investigated the effect of feature sizes on the emanating stray field shape at various observation heights. Feature sizes include two height (i.e., film thickness) values of 78 μm and 176 μm, and both width and spacing with three values of 360 μm, 520 μm and 680 μm. Ultrasound-assisted agitation was adopted for investigating the effects of electrodepositing current densities on the film crystalline microstructures and magnetic properties. Narrowing the width of micro-magnets helps the stray field to become a sinusoidal profile. Thinner film, i.e., thickness 78 μm in this study, supports the stray field taking on a sinusoidal profile. Moreover, the spacing between the micro-magnets plays a key factor in determining the shape of the stray field. Under 37 kHz/156 W ultrasound agitation, the optimal hard magnetic properties of electrodeposited CoMnP films are residual magnetization 2329 G and coercivity 968 Oe by a current density of 10.0 mA/cm2. Ultrasound-assisted electrodeposition, along with duly designed feature size, facilitates the micro-magnet arrays having a sinusoidal stray field shape using high quality films. Furthermore, for the first time, a systematic understanding of feature-size-dependent stray field evolution and improved polarities quality has been realized for the recording media of sinusoidal magnetic encoders. Full article
(This article belongs to the Special Issue Functional Coatings and Surface Science for Precision Engineering)
Show Figures

Figure 1

Figure 1
<p>Schematic diagrams of three main parts of the present work: (<b>a</b>) The experimental process for the studies of micro-magnet feature sizes by micro-fabrication and electrodepositing processes using a low-carbon steel substrate with a thickness of 4.0 mm. (<b>b</b>) Studies of electrodepositing current density under ultrasound-assisted electrodeposition using a copper substrate with a thickness of 0.5 mm for investigating film quality and materials properties. (<b>c</b>) Integration and optimization of the techniques from (<b>a</b>,<b>b</b>) using a low-carbon steel substrate to realize fine pole pitch micro-magnet arrays with mitigated surface micro-cracks having a sinusoidal stray field.</p>
Full article ">Figure 2
<p>Setups for magnetization and magnetic field measurement: (<b>a</b>) Capacitive discharge pulse magnetizer connecting with a C-shaped magnetizing head for magnetization. The scheme on the right indicates the applied magnetic field (Happ) generated from the two yokes, pointing to the OP direction of the sample. Direction of Happ is represented by the green arrows. (<b>b</b>) A photograph showing the setup for measuring magnetic flux density as a function of spatial distribution using a Hall sensor. (<b>c</b>) A schematic illustration of the magnetic field profile measurement: the magnetic field strength in both the OP and IP directions is measured at 20 µm intervals, in a stepwise manner.</p>
Full article ">Figure 3
<p>Top view SEM images of the micro-magnet arrays in the format of gratings with different micro-magnet width (W) and spacing (D): (<b>a</b>) W360-D360; (<b>b</b>) W360-D520; (<b>c</b>) W360-D680; (<b>d</b>) W520-D360; (<b>e</b>) W680-D360. (<b>f</b>) Optical image showing a magnetic field viewing card placed on a W680-D360 sample after magnetization. On the bottom of the image a portion of the sample appeared.</p>
Full article ">Figure 4
<p>Measurement results of the stray field profiles of (<b>a</b>) W360-D360; (<b>b</b>) W360-D520; (<b>c</b>) W360-D680; (<b>d</b>) W520-D360; (<b>e</b>) W680-D360 samples with 176 μm height (T-series) micro-magnets. The insets show the enlarged portions of the measurement results.</p>
Full article ">Figure 5
<p>Measurement results of the stray field profiles of (<b>a</b>) W360-D360; (<b>b</b>) W360-D520; (<b>c</b>) W360-D680; (<b>d</b>) W520-D360; (<b>e</b>) W680-D360 samples with 78 μm height (F-series) micro-magnets. The insets show the enlarged portions of the measurement results.</p>
Full article ">Figure 6
<p>SEM images showing the morphologies of CoMnP hard magnetic films by ultrasound-assisted electrodeposition (denoted as U) at various depositing current density: (<b>a</b>) 5.0 mA/cm<sup>2</sup>; (<b>b</b>) 7.5 mA/cm<sup>2</sup>; (<b>c</b>) 10.0 mA/cm<sup>2</sup>; (<b>d</b>) 12.5 mA/cm<sup>2</sup>; and by conventional electrodeposition (denoted as P) at various depositing current density: (<b>e</b>) 5.0 mA/cm<sup>2</sup>; (<b>f</b>) 7.5 mA/cm<sup>2</sup>; (<b>g</b>) 10.0 mA/cm<sup>2</sup>; (<b>h</b>) 12.5 mA/cm<sup>2</sup>.</p>
Full article ">Figure 7
<p>(<b>a</b>) X-ray diffraction patterns of CoMnP hard magnetic films by ultrasound-assisted electrodeposition (denoted as U) at various depositing current densities and by conventional electrodeposition (denoted as P) at 5.0 mA/cm<sup>2</sup> current density; (<b>b</b>) JCPDS (Joint Committee on Powder Diffraction Standards) card information of crystalline cobalt.</p>
Full article ">Figure 8
<p>Hysteresis curves (normalized) measured in the OP and IP directions for CoMnP hard magnetic films by ultrasound-assisted electrodeposition under different electrodepositing current density: (<b>a</b>) 5.0 mA/cm<sup>2</sup>; (<b>b</b>) 7.5 mA/cm<sup>2</sup>; (<b>c</b>) 10.0 mA/cm<sup>2</sup>; (<b>d</b>) 12.5 mA/cm<sup>2</sup>.</p>
Full article ">Figure 9
<p>Magnetic measurement results of the second quadrant B-H loops acquired in the (<b>a</b>) IP direction; and (<b>b</b>) OP direction for samples by ultrasound-assisted electrodeposition (denoted as U) at various depositing current density and by conventional electrodeposition (denoted as P) at 5.0 mA/cm<sup>2</sup> current density.</p>
Full article ">Figure 10
<p>Top view SEM images of the micro-magnet arrays by the combined air-bubbling and ultrasound-assisted agitation electrodeposition, in different magnification: (<b>a</b>) 20×; (<b>b</b>) 100×; (<b>c</b>) 800×; (<b>d</b>) 1500×. The zoomed-in areas are denoted by white blocks.</p>
Full article ">Figure 11
<p>(<b>a</b>) Measurement results of the stray field profiles of W200-D360 sample. The inset shows an enlarged portion of the measurement results; (<b>b</b>) Amplitude of OP fields as a function of observation height for different feature sizes of micro-magnet arrays.</p>
Full article ">
17 pages, 6429 KiB  
Article
Element Array Optimization for Skin-Attachable Ultrasound Probes to Improve the Robustness against Positional and Angular Errors
by Takumi Noda, Takashi Azuma, Ichiro Sakuma and Naoki Tomii
Appl. Sci. 2024, 14(20), 9320; https://doi.org/10.3390/app14209320 - 12 Oct 2024
Viewed by 920
Abstract
Skin-attachable ultrasound probes face challenges in imaging the intended cross-section due to the difficulty in precisely adjusting the position and angle of attachment. While matrix element arrays are capable of imaging any cross-section within a three-dimensional field of view, their implementation presents a [...] Read more.
Skin-attachable ultrasound probes face challenges in imaging the intended cross-section due to the difficulty in precisely adjusting the position and angle of attachment. While matrix element arrays are capable of imaging any cross-section within a three-dimensional field of view, their implementation presents a challenge due to the significant number of required ultrasound elements. We propose a method for optimizing the coordinates and shapes of elements based on the focusing quality onto the imaging points under the positional and angular errors in the element array. A 128-element array was optimized through the proposed method and its imaging performance was evaluated with simulated phantoms. The optimized array demonstrated the ability to clearly visualize the simulated wires, cysts, and blood vessels even with the positional error of 3 mm and the angular error of 20°. These results indicate the feasibility of developing a skin-attachable ultrasound probe that can be easily used in daily life without requiring precise positional and angular accuracy. Full article
(This article belongs to the Special Issue Current Updates on Ultrasound for Biomedical Applications)
Show Figures

Figure 1

Figure 1
<p>The arrangement of the focus point, nearby points, and peripheral points. The focus point for evaluating the main lobe level was randomly placed within the imaging region. The nearby points for evaluating the main lobe width were arranged at equal intervals on an ellipse centered on the focus point. The peripheral points for evaluating the side lobe level were randomly distributed within the region where the depth is close to the focus point.</p>
Full article ">Figure 2
<p>Simulated scatterer distributions used in the ultrasound imaging experiments. (<b>a</b>) Simulated wire scatterers. Nine wires, each 5 mm in length, are arranged with 10 mm intervals. Scattering points are placed on the wires at 1/10 wavelength intervals. In regions other than the wires, scattering points are randomly placed at a density of one per cubic volume with sides of one wavelength. (<b>b</b>) Simulated cyst scatterers. Four cysts, each with a diameter of 10 mm, are arranged with 15 mm intervals. There are no scattering points inside the cysts, while scattering points are randomly placed outside the cysts at a density of one per cubic volume with sides of one wavelength. (<b>c</b>) Simulated blood vessel scatterer. A blood vessel with an inner diameter of 6 mm and an outer diameter of 7 mm is positioned at a depth of 25 mm. There are no scattering points inside the blood vessel, while scattering points are randomly placed in the blood vessel wall and outside the blood vessel at densities of four and one per cubic volume with sides of one wavelength, respectively. (<b>d</b>–<b>f</b>) Ground truth maps of the US images for the wire, cyst, and blood vessel scatterers, respectively.</p>
Full article ">Figure 3
<p>Imaging performance comparison of the optimized element arrays with different cost function weights. (<b>a</b>) Representative examples of the optimized element arrays. The one on the right achieved the highest signal-to-noise ratio (SNR) in the imaging of wire scatterers. (<b>b</b>) Ultrasound images of the wire scatterers obtained with the optimized element arrays shown in (<b>a</b>) with an angle error of 20° around the <math display="inline"><semantics> <mrow> <mo> </mo> <mi>z</mi> </mrow> </semantics></math>-axis and a positional error of 3 mm in the <math display="inline"><semantics> <mrow> <mi>y</mi> </mrow> </semantics></math>-direction. The dynamic range is 25 dB. (<b>c</b>) Average SNRs of the nine wires, each imaged under nine different positional or angular error conditions: three angular errors (−20°, 0°, 20°) and three positional errors (−3 mm, 0 mm, 3 mm). The optimized array with the cost function weights of <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>β</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msup> <mrow> <mn>10</mn> </mrow> <mrow> <mo>−</mo> <mn>2</mn> </mrow> </msup> <mo>,</mo> <mo> </mo> <msub> <mrow> <mi>β</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> <mo>=</mo> <msup> <mrow> <mn>10</mn> </mrow> <mrow> <mn>0</mn> </mrow> </msup> </mrow> </semantics></math> achieved the highest average SNR of the wires.</p>
Full article ">Figure 4
<p>Evaluation of the robustness of the element arrays against positional and angular errors in the imaging of wire scatterers. (<b>a</b>,<b>b</b>) The average SNR of the nine wires in US images obtained under various positional and angular errors, respectively.</p>
Full article ">Figure 5
<p>Examples of ultrasound images of wire scatterers obtained with the linear, matrix, and optimized arrays. The left column shows the element arrays, and the top row shows the positional relationship between the imaging plane and the element arrays. The dynamic range of the ultrasound images is 25 dB. The average SNR of the wires are displayed on each image.</p>
Full article ">Figure 6
<p>Evaluation of the robustness of the element arrays against positional and angular errors in the imaging of cyst scatterers. (<b>a</b>,<b>b</b>) The average CNR of the four cysts in US images obtained under various positional and angular errors, respectively.</p>
Full article ">Figure 7
<p>Examples of ultrasound images of cyst scatterers obtained with the linear, matrix, and optimized arrays. The left column shows the element arrays, and the top row shows the positional relationship between the imaging plane and the element arrays. The dynamic range of the ultrasound images is 25 dB. The average CNR of the cysts are displayed on each image.</p>
Full article ">Figure 8
<p>Examples of US images of blood vessel scatterers obtained with the linear, matrix, and optimized arrays. Before the image reconstruction, white noise was added to the US signals so that the SNR became 20 dB. The left column shows the element arrays, and the top row shows the positional relationship between the imaging plane and the element arrays. The dynamic range of the ultrasound images is 25 dB. The average measurement error from the ground truth blood vessel inner diameter of 6 mm is displayed on each image.</p>
Full article ">
20 pages, 4646 KiB  
Article
Comparative Approach to Performance Estimation of Pulsed Wave Doppler Equipment Based on Kiviat Diagram
by Giorgia Fiori, Andrea Scorza, Maurizio Schmid, Silvia Conforto and Salvatore Andrea Sciuto
Sensors 2024, 24(19), 6491; https://doi.org/10.3390/s24196491 - 9 Oct 2024
Viewed by 803
Abstract
Quality assessment of ultrasound medical systems is a demanding task due to the high number of parameters to quantify their performance: in the present study, a Kiviat diagram-based integrated approach was proposed to effectively combine the contribution of some experimental parameters and quantify [...] Read more.
Quality assessment of ultrasound medical systems is a demanding task due to the high number of parameters to quantify their performance: in the present study, a Kiviat diagram-based integrated approach was proposed to effectively combine the contribution of some experimental parameters and quantify the overall performance of pulsed wave Doppler (PWD) systems for clinical applications. Four test parameters were defined and assessed through custom-written measurement methods based on image analysis, implemented in the MATLAB environment, and applied to spectral images of a flow phantom, i.e., average maximum velocity sensitivity (AMVS), velocity measurements accuracy (VeMeA), lowest detectable signal (LDS), and the velocity profile discrepancy index (VPDI). The parameters above were scaled in a standard range to represent the four vertices of a Kiviat plot, whose area was considered the overall quality index of the ultrasound system in PWD mode. Five brand-new ultrasound diagnostic systems, equipped with linear array probes, were tested in two different working conditions using a commercial flow phantom as a reference. The promising results confirm the robustness of AMVS, VeMeA, and LDS parameters while suggesting further investigations on the VPDI. Full article
Show Figures

Figure 1

Figure 1
<p>Example of sample volume positioning at six different radial distances (<span class="html-italic">R</span> = radius) from the flow axis of a sketched straight vessel of a reference device for the velocity profile measurement.</p>
Full article ">Figure 2
<p>Block diagram of the measurement method for the velocity profile discrepancy index assessment.</p>
Full article ">Figure 3
<p>Block diagram of the measurement method for the average maximum velocity sensitivity assessment.</p>
Full article ">Figure 4
<p>Block diagram of the measurement method for the velocity measurements accuracy assessment.</p>
Full article ">Figure 5
<p>Block diagram of the measurement method for the lowest detectable signal assessment.</p>
Full article ">Figure 6
<p>Example of ROIn and ROIv positioning on the spectral images to determine the maximum Doppler gain <span class="html-italic">G<sub>max</sub></span> (<b>a</b>) and the minimum Doppler gain <span class="html-italic">G<sub>min</sub></span> (<b>b</b>).</p>
Full article ">Figure 7
<p>Kiviat diagrams for UDS1 (<b>a</b>,<b>b</b>), UD2 (<b>c</b>,<b>d</b>) according to the working condition: pre-set A (<b>a</b>,<b>c</b>) and pre-set B (<b>b</b>,<b>d</b>).</p>
Full article ">Figure 8
<p>Kiviat diagrams for UDS3 (<b>a</b>,<b>b</b>), UDS4 (<b>c</b>,<b>d</b>), and UD5 (<b>e</b>,<b>f</b>) according to the working condition: pre-set A (<b>a</b>,<b>c</b>,<b>e</b>) and pre-set B (<b>b</b>,<b>d</b>,<b>f</b>).</p>
Full article ">
26 pages, 7912 KiB  
Article
Investigation of Sonication Parameters for Large-Volume Focused Ultrasound-Mediated Blood–Brain Barrier Permeability Enhancement Using a Clinical-Prototype Hemispherical Phased Array
by Dallan McMahon, Ryan M. Jones, Rohan Ramdoyal, Joey Ying Xuan Zhuang, Dallas Leavitt and Kullervo Hynynen
Pharmaceutics 2024, 16(10), 1289; https://doi.org/10.3390/pharmaceutics16101289 - 30 Sep 2024
Viewed by 1293
Abstract
Background/Objectives: Focused ultrasound (FUS) and microbubble (MB) exposure is a promising technique for targeted drug delivery to the brain; however, refinement of protocols suitable for large-volume treatments in a clinical setting remains underexplored. Methods: Here, the impacts of various sonication parameters on blood–brain [...] Read more.
Background/Objectives: Focused ultrasound (FUS) and microbubble (MB) exposure is a promising technique for targeted drug delivery to the brain; however, refinement of protocols suitable for large-volume treatments in a clinical setting remains underexplored. Methods: Here, the impacts of various sonication parameters on blood–brain barrier (BBB) permeability enhancement and tissue damage were explored in rabbits using a clinical-prototype hemispherical phased array developed in-house, with real-time 3D MB cavitation imaging for exposure calibration. Initial experiments revealed that continuous manual agitation of MBs during infusion resulted in greater gadolinium (Gd) extravasation compared to gravity drip infusion. Subsequent experiments used low-dose MB infusion with continuous agitation and a low burst repetition frequency (0.2 Hz) to mimic conditions amenable to long-duration clinical treatments. Results: Key sonication parameters—target level (proportional to peak negative pressure), number of bursts, and burst length—significantly affected BBB permeability enhancement, with all parameters displaying a positive relationship with relative Gd contrast enhancement (p < 0.01). Even at high levels of BBB permeability enhancement, tissue damage was minimal, with low occurrences of hypointensities on T2*-weighted MRI. When accounting for relative Gd contrast enhancement, burst length had a significant impact on red blood cell extravasation detected in histological sections, with 1 ms bursts producing significantly greater levels compared to 10 ms bursts (p = 0.03), potentially due to the higher pressure levels required to generate equal levels of BBB permeability enhancement. Additionally, albumin and IgG extravasation correlated strongly with relative Gd contrast enhancement across sonication parameters, suggesting that protein extravasation can be predicted from non-invasive imaging. Conclusions: These findings contribute to the development of safer and more effective clinical protocols for FUS + MB exposure, potentially improving the efficacy of the approach. Full article
(This article belongs to the Section Drug Delivery and Controlled Release)
Show Figures

Figure 1

Figure 1
<p>Clinical-prototype FUS array system and PCI-based cavitation feedback control example. (<b>a</b>) Experimental setup. Animals were positioned supine with scalps coupled directly to degassed/deionized water within the FUS array. Ultraharmonic receivers were used for PCI-based control and subharmonic receivers were used for ICD. (<b>b</b>) Transmit/receive module. Each module contains 64 transducer elements (8 × 8 grid, 2.5 mm inter-element spacing, 60 transmit and 4 receive elements). (<b>c</b>) Transmit and receive array layouts. (<b>d</b>) PCI-based cavitation feedback control example from in vivo data. PNP was iteratively increased until the detection of coherent MB activity via PCI. On the burst prior to detection (t = 105 s, magenta; PNP = 0.49 MPa) there was no evidence of coherent MB activity on PCI and the SPTA intensity remained below the maximum levels observed during baseline pressure ramps without MBs in circulation (blue dotted line). Spatially coherent MB activity observed in PCI MIPs (t = 110 s, red; calibration PNP = 0.51 MPa) was accompanied by a large spike in the SPTA intensity. The driving voltage was reduced to the minimum system output until the calibration phase was completed at all targets. In this example, a target level of 50% was set (Tx phase PNP = 0.25 MPa for 60 bursts). There was no evidence of MB activity on PCI throughout the Tx phase during which the SPTA intensity remained below the maximum levels observed during baseline pressure ramps, as seen at t = 250 s (green). White scale bar = 4 mm. FUS: focused ultrasound; ICD: inertial cavitation detection; MB: microbubble; PCI: passive cavitation imaging; PNP: peak negative pressure; SPTA: spatial peak temporal average; Tx: treatment.</p>
Full article ">Figure 2
<p>Calibration phase cavitation feedback control scheme example. (<b>a</b>) An in vivo example of the cavitation feedback control scheme used during the calibration phase of FUS + MB exposures. PNP was increased each burst until satisfying PCI detection thresholds and/or exceeding the threshold for ICD ratio (<b>top panel</b>). In this example, PCI detection thresholds were satisfied on the 24th burst of the calibration phase (time = 115 s, green marker), corresponding to a divergence in PCI SPTA intensity versus the baseline pressure ramp (i.e., no MBs in circulation; <b>middle panel</b>). The ICD threshold was exceeded during the same burst (<b>bottom panel)</b>. (<b>b</b>) Frequency spectrum of filtered (8th order digital Butterworth filter, 380–400 kHz bandpass) RF data delay-and-summed to the voxel of maximum PCI SPTA intensity (green marker) for the calibration pressure burst (t = 115 s, blue line; PNP = 0.51 MPa), as well as that of the same voxel and sonicating PNP during a baseline pressure ramp without MBs in circulation (t = 115 s, red line; PNP = 0.51 MPa). The filtered delay-and-summed frequency spectrum for the burst prior to the calibration pressure with MBs in circulation is also shown (t = 110 s, light blue line; PNP = 0.49 MPa). (<b>c</b>) The mean unfiltered frequency spectrum across 4 subharmonic receivers used for ICD is shown for the calibration pressure burst (t = 115 s; blue line; PNP = 0.51 MPa), the same sonicating PNP during a baseline pressure ramp without MBs in circulation (t = 115 s; red line; PNP = 0.51 MPa), and the burst prior to the calibration pressure with MBs in circulation (t = 110 s; light blue line; PNP = 0.49). Light green rectangles indicate the bandwidth used for ICD calculations. White scale bar = 4 mm.</p>
Full article ">Figure 3
<p>Preliminary comparison of MB infusion methods. In a subset of rabbits (n = 4; aka Cohort #1), the impact of MB infusion method on BBB permeability enhancement was evaluated. (<b>a</b>–<b>d</b>) Target layout and sonication parameters are displayed in relation to T2w targeting scans for gravity drip (n = 24 targets) and continuous manual agitation (n = 32 targets) infusion methods. (<b>e</b>–<b>h</b>) T1w MRI highlights differences in relative Gd contrast enhancement between infusion methods and various sonication parameters. (<b>i</b>–<b>l</b>) No evidence of hypointensities in T2*w MRI were observed. (<b>m</b>) Relative Gd contrast enhancement is plotted for gravity drip and continuous manual agitation infusion; for each infusion method, only targets for which target level ≥ 70% and Tx phase bursts = 120, were considered (n = 8 targets for continuous manual infusion; n = 18 targets for gravity drip infusion). A significant difference was detected between infusion methods (<span class="html-italic">p</span> &lt; 0.01). White scale bars = 1 cm. A: anterior; L: left; P: posterior; R: right.</p>
Full article ">Figure 4
<p>Relative Gd contrast enhancement across sonication parameters. T1w MRI was performed approximately 10 min following the end of sonication. (<b>a</b>) Target-wise (n = 237) relative Gd contrast enhancement is plotted for each set of sonication parameters investigated. Burst length (<span class="html-italic">p</span> &lt; 0.01), target level (<span class="html-italic">p</span> &lt; 0.01), and number of Tx phase bursts (<span class="html-italic">p</span> &lt; 0.01) had significant effects on relative Gd contrast enhancement. Targets sonicated with 75% target level exhibited significantly greater levels of relative Gd contrast enhancement vs. 50% target level (<span class="html-italic">p</span> &lt; 0.01). At 75% target level, burst length had a significant impact on relative Gd contrast enhancement (<span class="html-italic">p</span> &lt; 0.01), with a significant difference between 1 ms vs. 10 ms burst lengths (<span class="html-italic">p</span> = 0.048). Number of Tx phase bursts also had a significant effect on relative Gd contrast enhancement at 75% target level (<span class="html-italic">p</span> &lt; 0.01). TL = target level. Representative examples of the targeting scheme (<b>b</b>,<b>e</b>), Gd contrast enhancement in T1w MRI (<b>c</b>,<b>f</b>), and T2*w MRI (<b>d</b>,<b>g</b>) are shown for sonications performed with 5 ms bursts and either 75% (<b>b</b>–<b>d</b>) or 50% (<b>e</b>–<b>g</b>) target levels. Number of Tx phase bursts range from 0 to 240 within each animal. White scale bars = 1 cm. A: anterior; L: left; P: posterior; R: right.</p>
Full article ">Figure 5
<p>RBC extravasation area across sonication parameters. In a subset of rabbits, the area of RBC extravasation was quantified in H&amp;E-stained tissue sections (66 targets, 5 rabbits). The animals were perfused 1.5 h following the start of FUS + MB exposure. (<b>a</b>) A significant correlation between relative Gd contrast enhancement and RBC area was observed across all sonication parameters (r<sup>2</sup> = 0.24, <span class="html-italic">p</span> &lt; 0.01). For targets sonicated with 5 ms bursts a significant correlation was also observed (r<sup>2</sup> = 0.33, <span class="html-italic">p</span> &lt; 0.01). At 75% target level, when relative Gd contrast enhancement was considered as a covariate, burst length (<span class="html-italic">p</span> &lt; 0.01) had significant effects on RBC extravasation. Post-hoc Tukey’s HSD test revealed a significant difference between 1 ms and 10 ms burst lengths (<span class="html-italic">p</span> = 0.03). (<b>b</b>) The target displaying the largest area of RBC extravasation observed (yellow border) was sonicated with 1 ms bursts, 75% target level, and 120 Tx phase bursts. This target displayed hypointense signal intensity on T2*w imaging (<a href="#pharmaceutics-16-01289-f0A3" class="html-fig">Figure A3</a>). The data point corresponding to this target is circled (yellow) in panel (<b>a</b>). (<b>c</b>) A histological image representative of 5 or 10 ms burst lengths and 75% target level is displayed (cyan border). Low levels of RBC extravasation are observed across the sonicated volume. The data point corresponding to this target is circled (cyan) in panel (<b>a</b>).</p>
Full article ">Figure 6
<p>Albumin and IgG immunofluorescence following sonication. In a subset of animals, immunofluorescent staining for albumin and IgG was performed on tissue sections from rabbits perfused for 1.5 h following the end of FUS + MB exposure (5 rabbits, 66 targets). A strong linear correlation was observed between relative Gd contrast enhancement and relative immunofluorescent signal intensity for both (<b>a</b>) albumin–GFP (r<sup>2</sup> = 0.63, <span class="html-italic">p</span> &lt; 0.01) and (<b>b</b>) IgG-DsRed (r<sup>2</sup> = 0.46, <span class="html-italic">p</span> &lt; 0.01). Across all sonication parameters, the target level had a significant effect on relative immunofluorescent signal intensity for both (<b>a</b>) albumin–GFP (<span class="html-italic">p</span> &lt; 0.01) and (<b>b</b>) IgG-DsRed (<span class="html-italic">p</span> &lt; 0.01). (<b>a</b>,<b>b</b>) When relative Gd contrast enhancement was considered as a covariate, no significant effect of target level or number of Tx phase bursts on relative immunofluorescent signal intensity of either protein were observed. (<b>c</b>) A representative example of albumin–GFP immunofluorescence for 8 posterior targets sonicated with 1 ms burst lengths and 75% target level is displayed. Areas of relatively homogeneous signal intensity across the target volume (cyan border) is contrasted with a more heterogeneous signal intensity (yellow border). Evidence of perivascular transport of albumin–GFP is shown in an area distant from any targeted volume (magenta border); IgG-DsRed signal intensity is not higher than background levels in this ROI.</p>
Full article ">Figure A1
<p>Comparison of estimated plasma concentration of Definity<sup>TM</sup> MBs over time. One compartment model of MB concentration in circulation over time for bolus (4 μL/kg) and infusion (0.8 μL/kg/min and 1.6 μL/kg/min) administration. Half-life of Definity<sup>TM</sup> was assumed to be 79 s [<a href="#B57-pharmaceutics-16-01289" class="html-bibr">57</a>]. For bolus administration, t = 0 represents the peak concentration in circulation. For infusion administration, t = 0 represents the start of delivery. Sonication durations are set to 120 s.</p>
Full article ">Figure A2
<p>Relative Gd contrast enhancement analysis. (<b>a</b>) An example of contrast enhanced T1w MRI collected following FUS + MB exposure. (<b>b</b>) ROIs selected for quantification of T1w signal intensity at targets (#1–16), as well as regions used as non-sonicated control tissue (C1–C4). Mean signal intensity within each targeted ROI was divided by mean signal intensity across the non-sonicated control ROIs to obtain relative Gd contrast enhancement values. White scale bars = 1 cm. A: anterior; L: left; P: posterior; R: right.</p>
Full article ">Figure A3
<p>T2*w hypointensity induced by FUS + MB exposure. (<b>a</b>,<b>b</b>) Sequential coronal slices from contrast enhanced T1w MRI collected following FUS + MB exposure. White arrows highlight a single target of increased BBB permeability. (<b>c</b>,<b>d</b>) Red arrows highlight a small area of hypointensity in T2*w MRI corresponding to the same target highlighted above. This target displayed the largest area of RBC extravasation in H&amp;E sections across all targets processed for histology. White scale bars = 1 cm. D: dorsal; L: left; R: right; V: ventral.</p>
Full article ">Figure A4
<p>Correlations between RBC extravasation area and relative Gd contrast enhancement for different numbers of Tx phase bursts. In a subset of rabbits, area of RBC extravasation was quantified in H&amp;E stained tissue sections (66 targets, 5 rabbits). Animals were perfused 1.5 h following the end of FUS + MB exposure. A significant correlation between relative Gd contrast enhancement and RBC area was observed across all sonication parameters (r<sup>2</sup> = 0.24, <span class="html-italic">p</span> &lt; 0.01). For targets sonicated with 60, 120, and 240 Tx phase bursts, significant correlations were also observed (60 Tx phase bursts: r<sup>2</sup> = 0.49, <span class="html-italic">p</span> &lt; 0.01; 120 Tx phase bursts: r<sup>2</sup> = 0.51, <span class="html-italic">p</span> &lt; 0.01; 240 Tx phase bursts: r<sup>2</sup> = 0.3, <span class="html-italic">p</span> = 0.03). For targets sonicated with 75% target level, when relative Gd contrast enhancement was considered as a covariate, number of Tx phase bursts (<span class="html-italic">p</span> &lt; 0.01) and burst length (<span class="html-italic">p</span> &lt; 0.01; <a href="#pharmaceutics-16-01289-f005" class="html-fig">Figure 5</a>a) had significant effects on RBC extravasation. Post-hoc analysis revealed no significant differences between any paired comparison of Tx phase burst numbers.</p>
Full article ">Figure A5
<p>Correlations between protein immunofluorescence and relative Gd contrast enhancement for different numbers of Tx phase bursts. In a subset of animals, immunofluorescent staining for albumin and IgG was performed on tissue sections from rabbits perfused 1.5 h following the end of FUS + MB exposure (5 rabbits, 66 targets). A strong linear correlation was observed between relative Gd contrast enhancement and relative immunofluorescent signal intensity for both (<b>a</b>) albumin-GFP (r<sup>2</sup> = 0.63, <span class="html-italic">p</span> &lt; 0.01) and (<b>b</b>) IgG-DsRed (r<sup>2</sup> = 0.46, <span class="html-italic">p</span> &lt; 0.01). When targets sonicated with an equal number of Tx phase bursts were considered, significant correlations were observed for 60, 120, and 240 Tx phase bursts for both albumin-GFP (60 Tx phase bursts: r<sup>2</sup> = 0.57, <span class="html-italic">p</span> &lt; 0.01; 120 Tx phase bursts: r<sup>2</sup> = 0.79, <span class="html-italic">p</span> &lt; 0.01; 240 Tx phase bursts: r<sup>2</sup> = 0.73, <span class="html-italic">p</span> &lt; 0.01) and IgG-DsRed (60 Tx phase bursts: r<sup>2</sup> = 0.31, <span class="html-italic">p</span> = 0.05; 120 Tx phase bursts: r<sup>2</sup> = 0.76, <span class="html-italic">p</span> &lt; 0.01; 240 Tx phase bursts: r<sup>2</sup> = 0.60, <span class="html-italic">p</span> &lt; 0.01). (<b>a</b>,<b>b</b>) When relative Gd contrast enhancement was considered as a covariate, no significant effect of the number of Tx phase bursts on relative immunofluorescent signal intensity of either protein were observed.</p>
Full article ">
18 pages, 7717 KiB  
Article
Machine Learning-Empowered Real-Time Acoustic Trapping: An Enabling Technique for Increasing MRI-Guided Microbubble Accumulation
by Mengjie Wu and Wentao Liao
Sensors 2024, 24(19), 6342; https://doi.org/10.3390/s24196342 - 30 Sep 2024
Viewed by 868
Abstract
Acoustic trap, using ultrasound interference to ensnare bioparticles, has emerged as a versatile tool for life sciences due to its non-invasive nature. Bolstered by magnetic resonance imaging’s advances in sensing acoustic interference and tracking drug carriers (e.g., microbubble), acoustic trap holds promise for [...] Read more.
Acoustic trap, using ultrasound interference to ensnare bioparticles, has emerged as a versatile tool for life sciences due to its non-invasive nature. Bolstered by magnetic resonance imaging’s advances in sensing acoustic interference and tracking drug carriers (e.g., microbubble), acoustic trap holds promise for increasing MRI-guided microbubbles (MBs) accumulation in target microvessels, improving drug carrier concentration. However, accurate trap generation remains challenging due to complex ultrasound propagation in tissues. Moreover, the MBs’ short lifetime demands high computation efficiency for trap position adjustments based on real-time MRI-guided carrier monitoring. To this end, we propose a machine learning-based model to modulate the transducer array. Our model delivers accurate prediction of both time-of-flight (ToF) and pressure amplitude, achieving low average prediction errors for ToF (−0.45 µs to 0.67 µs, with only a few isolated outliers) and amplitude (−0.34% to 1.75%). Compared with the existing methods, our model enables rapid prediction (<10 ms), achieving a four-order of magnitude improvement in computational efficiency. Validation results based on different transducer sizes and penetration depths support the model’s adaptability and potential for future ultrasound treatments. Full article
(This article belongs to the Special Issue Multi-sensor Fusion in Medical Imaging, Diagnosis and Therapy)
Show Figures

Figure 1

Figure 1
<p>Overview of trapping drug carriers using acoustic trapping. (<b>a</b>) Acoustic trap generation within human liver using phased array. MR-compatible robotic manipulator (Image source #1) [<a href="#B27-sensors-24-06342" class="html-bibr">27</a>] positions the array towards the liver. The emitted beams pass through tissues (e.g., skin, fat, muscle, and ribs) to generate a trap at the given target zone. MBs are injected via radial artery of the forearm [<a href="#B45-sensors-24-06342" class="html-bibr">45</a>]. (<b>b</b>) Example of focal spot visualization in <span class="html-italic">ex vivo</span> porcine kidney via MR-ARFI (Image source #2 [<a href="#B29-sensors-24-06342" class="html-bibr">29</a>]). (<b>c</b>) Close-up illustration of MB accumulation in a microvessel due to acoustic trapping. Two finger-like high-pressure (warmer color) regions locate MBs around the tumor cells. (<b>d</b>) Segmentation of MR T2 image for FE modeling in both data acquisition and performance validation process. Letters “A” and “P” in black color indicate the anterior and posterior abdominal walls, respectively.</p>
Full article ">Figure 2
<p>ANN-based workflow for acoustic trap generation. Step 1 illustrates the architecture of a learning-based model for predicting ToF or amplitude. The variable <span class="html-italic">n</span> is the element number. The proposed two ANNs have 2<span class="html-italic">n</span> + 2 input variables, five hidden layers and <span class="html-italic">n</span> outputs but diverge in hidden layers’ node layout (i.e., <span class="html-italic">N</span><sub>1</sub>, <span class="html-italic">N</span><sub>2</sub>, …, <span class="html-italic">N</span><sub>5</sub>) for <b><span class="html-italic">T</span></b> and <b><span class="html-italic">A</span></b> prediction. Step 2 depicts the phase–amplitude modulation process applied to <span class="html-italic">n</span> elements (e<sub>1</sub>, e<sub>2</sub>, …, e<span class="html-italic"><sub>n</sub></span>), including phase modulation (PA) and amplitude modulation (AM). Circles’ radii are proportional to the element’s emission pressure, and the grey scale of elements represents the phase pattern. A blue–red color bar is used to characterize the normalized pressure (NP) across all acoustic fields.</p>
Full article ">Figure 3
<p>Overview of FE-based data collection process. (<b>a</b>) Planning of array position and workspace for data collection. An MR T2 image (330 mm × 240 mm) was segmented to build a 2D geometry model for FE modeling. Boundaries <span class="html-italic">l</span><sub>2</sub> and <span class="html-italic">l</span><sub>3</sub> outlined the workspace, and these samples were distributed around the vessel with a spacing of <span class="html-italic">d</span>. (<b>b</b>) Pressure attenuation in relation to spread angle (2σ). At the critical angle of σ = 24.2°, the pressure attenuates by half (−6 dB) over the same propagating distance <span class="html-italic">r</span>. (<b>c</b>) Simulated acoustic field in COMSOL. Eight elements acted as receivers to capture wave signals from one representative sound source S* (<b>d</b>) Signal emitted by sound source S* and received signals by the array. (<b>e</b>) Close-up illustration of the received wave signals. Eight peak amplitudes and their timestamps denote, respectively, <b><span class="html-italic">A</span></b> and <b><span class="html-italic">T</span></b>.</p>
Full article ">Figure 4
<p>ToF prediction performance of the model trained separately using datasets with four sample sizes. (<b>a</b>) Prediction errors of eight elements (i.e., e<sub>1</sub>, e<sub>2</sub>, …, e<sub>8</sub>). The higher sample density (<span class="html-italic">D</span>) corresponds to the larger sample size. When <span class="html-italic">D</span> = 1/λ, the errors remain range from −0.45 µs to 0.67 µs, with few outliers. Below this density, the errors surge, and many outliers occur. (<b>b</b>) Average model training time over 10 runs using datasets with four sample sizes.</p>
Full article ">Figure 5
<p>Visualization of focal beam based on two phased arrays with different apertures. The size of the (<b>a</b>) small elements is 3.7 mm, and the (<b>b</b>) large element is 7.0 mm. When <span class="html-italic">D</span> was not less than 1/λ, both kinds of phased arrays formed a focal beam at the given position. Letters “A” and “P” in yellow color indicate the anterior and posterior abdominal walls, respectively.</p>
Full article ">Figure 6
<p>Visualization of twin traps based on two-phased arrays with different apertures. The size of the (<b>a</b>) small elements was 3.7 mm, and the (<b>b</b>) large elements was 7.0 mm. When <span class="html-italic">D</span> was not less than 1/λ, both arrays formed twin traps at the given positions. A pair of solid green circles represented twin trap’s two control points.</p>
Full article ">Figure 7
<p>Performance evaluation on focal beam formed without and with AM. (<b>a</b>) Average prediction errors of eight elements at the validated sample densities (i.e., 1/λ). Their maximum errors were less than 1.75%. (<b>b</b>) Average model training time over 10 runs. (<b>c</b>) Comparison of two focal beams formed without and with AM. The left one formed without AM served as a baseline. Two stacked figures showed the normalized pressure changes at target’s lateral direction across a 3.5λ span, as elements were activated in sequential order. The bottom blue layer represents the beam profile formed when five elements (i.e., e<sub>2</sub>, e<sub>3</sub>, e<sub>4</sub>, e<sub>7</sub>, and e<sub>8</sub>) were simultaneously activated. The turquoise, orange, and red layers represent the beam profiles of e<sub>1</sub>, e<sub>5</sub>, and e<sub>6</sub>, respectively. Their full-width-at-half-maximum of the acoustic intensity profile at the target was 4.59 mm and 4.04 mm, respectively.</p>
Full article ">Figure 8
<p>Numerical simulation on the microbubble trapping in pulsatile flow. (<b>a</b>) Curve of inflow velocity. Its period is 1 s, and the maximum velocity is up to 1.54 mm/s. It was used to drive the fluid inside microvessel. (<b>b</b>) Acoustic field pattern of the twin traps around the predefined microvessel. The maximum pressure is about 340 KPa. The microvessel diameter is 40 μm, and the fluid enters from the left side and flows to the right. (<b>c</b>) MB transient distribution at t = 0 s, 0.08 s, 0.12 s, and 0.18 s in COMSOL simulation. At t = 0 s, all MBs do not experience ARF and move due to fluid dynamics. After triggering the twin trap, the MBs (ACF &lt; 0) gradually accumulated in two highest pressure spots.</p>
Full article ">
26 pages, 1219 KiB  
Article
Array Comparative Genomic Hybridization (aCGH) Results among Patients Referred to Invasive Prenatal Testing after First-Trimester Screening: A Comprehensive Cohort Study
by Anna Wójtowicz, Katarzyna Kowalczyk, Katarzyna Szewczyk, Anna Madetko-Talowska, Wojciech Wójtowicz, Hubert Huras, Mirosław Bik-Multanowski and Nowakowska Beata
Diagnostics 2024, 14(19), 2186; https://doi.org/10.3390/diagnostics14192186 - 30 Sep 2024
Viewed by 804
Abstract
Introduction: Invasive prenatal testing with chromosomal microarray analysis after first-trimester screening is a relevant option but there is still debate regarding the indications. Therefore, we evaluated the prevalence of numerical chromosomal aberrations detected by classic karyotype and clinically relevant copy number variants (CNVs) [...] Read more.
Introduction: Invasive prenatal testing with chromosomal microarray analysis after first-trimester screening is a relevant option but there is still debate regarding the indications. Therefore, we evaluated the prevalence of numerical chromosomal aberrations detected by classic karyotype and clinically relevant copy number variants (CNVs) in prenatal samples using array comparative genomic hybridization (aCGH) stratified to NT thickness: <the 95th percentile, the 95th percentile–2.9 mm, 3.0–3.4 mm, 3.5–3.9 mm, 4.0–4.5 mm, and >4.5 mm, and by the presence/absence of associated structural anomalies detected by ultrasonography. Materials and Methods: Retrospective cohort study carried out at two tertiary Polish centers for prenatal diagnosis (national healthcare system) in central and south regions from January 2018 to December 2021. A total of 1746 prenatal samples were received. Indications for invasive prenatal testing included high risk of Down syndrome in the first-trimester combined test (n = 1484) and advanced maternal age (n = 69), and, in 193 cases, other reasons, such as parental request, family history of congenital defects, and genetic mutation carrier, were given. DNA was extracted directly from amniotic fluid (n = 1582) cells and chorionic villus samples (n = 164), and examined with classic karyotype and aCGH. Results: Of the entire cohort of 1746 fetuses, classical karyotype revealed numerical chromosomal aberrations in 334 fetuses (19.1%), and aCGH detected CNV in 5% (n = 87). The frequency of numerical chromosomal aberrations increased with NT thickness from 5.9% for fetuses with NT < p95th to 43.3% for those with NT > 4.5 mm. The highest rate of numerical aberrations was observed in fetuses with NT > 4.5 mm having at least one structural anomaly (50.2%). CNVs stratified by NT thickness were detected in 2.9%, 2.9%, 3.5%, 4.3%, 12.2%, and 9.0% of fetuses with NT < 95th percentile, 95th percentile–2.9 mm, 3.0–3.4 mm, 3.5–3.9 mm, 4.0–4.5 mm, and >4.5 mm, respectively. After exclusion of fetuses with structural anomalies and numerical aberrations, aCGH revealed CNVs in 2.0% of fetuses with NT < 95th percentile, 1.5% with NTp95–2.9 mm, 1.3% with NT 3.0–3.4 mm, 5.4% with NT 3.5–3.9 mm, 19.0% with NT 4.0–4.5 mm, and 14.8% with NT > 4.5 mm. Conclusions: In conclusion, our study indicates that performing aCGH in samples referred to invasive prenatal testing after first-trimester screening provides additional clinically valuable information over conventional karyotyping, even in cases with normal NT and anatomy. Full article
(This article belongs to the Special Issue Diagnosis and Management in Prenatal Medicine, 3rd Edition)
Show Figures

Figure 1

Figure 1
<p>Prenatal diagnosis algorithm according to the Polish recommendations.</p>
Full article ">Figure 2
<p>Distribution of numerical chromosomal aberrations depending on nuchal translucency (NT) thickness.</p>
Full article ">Figure 3
<p>Distribution of copy number variants (CNVs) depending on the nuchal translucency (NT) thickness.</p>
Full article ">
14 pages, 7528 KiB  
Article
Fine-Tuning of Optical Resonance Wavelength of Surface-Micromachined Optical Ultrasound Transducer Arrays for Single-Wavelength Light Source Readout
by Zhiyu Yan, Cheng Fang and Jun Zou
Micromachines 2024, 15(9), 1111; https://doi.org/10.3390/mi15091111 - 31 Aug 2024
Viewed by 1046
Abstract
This article reports the fine-tuning of the optical resonance wavelength (ORW) of surface-micromachined optical ultrasound transducer (SMOUT) arrays to enable ultrasound data readout with non-tunable interrogation light sources for photoacoustic computed tomography (PACT). Permanent ORW tuning is achieved by material deposition onto or [...] Read more.
This article reports the fine-tuning of the optical resonance wavelength (ORW) of surface-micromachined optical ultrasound transducer (SMOUT) arrays to enable ultrasound data readout with non-tunable interrogation light sources for photoacoustic computed tomography (PACT). Permanent ORW tuning is achieved by material deposition onto or subtraction from the top diaphragm of each element with sub-nanometer resolution. For demonstration, a SMOUT array is first fabricated, and its ORW is tuned for readout with an 808 nm laser diode (LD). Experiments are conducted to characterize the optical and acoustic performances of the elements within the center region of the SMOUT array. Two-dimensional and three-dimensional PACT (photoacoustic computed tomography) is also performed to evaluate the imaging performance of the ORW-tuned SMOUT array. The results show that the ORW tuning does not degrade the optical, acoustic, and overall imaging performances of the SMOUT elements. As a result, the fine-tuning method enables new SMOUT-based PACT systems that are low cost, compact, powerful, and even higher speed, with parallel readout capability. Full article
Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) The cross-section of a SMOUT element, the top DBR diaphragm (constructed by multi-layer SiN/SiO) of which is vibrated by the impinging ultrasound wave; (<b>b</b>) the SMOUT reflectance spectrum shifted by the top diaphragm vibration, where <math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mn>0</mn> </mrow> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">b</mi> <mi mathvariant="normal">i</mi> <mi mathvariant="normal">a</mi> <mi mathvariant="normal">s</mi> </mrow> </msub> </mrow> </semantics></math> are the ORW and interrogation wavelength, respectively.</p>
Full article ">Figure 2
<p>SMOUT readout (<b>a</b>) in serial with a single beam (with low optical power) and (<b>b</b>) in parallel with multiple beams (with high overall optical power).</p>
Full article ">Figure 3
<p>(<b>a</b>) Four representative SMOUT arrays (a, b, c, and d) fabricated on different substrates in one batch or different batches; (<b>b</b>) reflectance spectra and ORWs (<math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">a</mi> </mrow> </msub> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">b</mi> </mrow> </msub> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">c</mi> </mrow> </msub> </mrow> </semantics></math>, and <math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">d</mi> </mrow> </msub> </mrow> </semantics></math> for a, b, c, and d array, respectively) before the tuning; (<b>c</b>) uniform reflectance spectra and ORWs after the tuning (<math display="inline"><semantics> <mrow> <msub> <mrow> <mi mathvariant="sans-serif">λ</mi> </mrow> <mrow> <mi mathvariant="normal">L</mi> <mi mathvariant="normal">D</mi> </mrow> </msub> </mrow> </semantics></math>: the output wavelength of the high-power non-tunable LD).</p>
Full article ">Figure 4
<p>(<b>a</b>) Pressure-induced deflection and cavity length changes and (<b>b</b>) the corresponding ORW shifts due to coating or etching the top diaphragm of the SMOUT element.</p>
Full article ">Figure 5
<p>COMSOL simulation of ORW tuning by coating or etching the top DBR diaphragm: (<b>a</b>) the FEM model built in COMSOL Multiphysics; (<b>b</b>) cross-section view of the simulated pressure-deflected diaphragm before tuning; (<b>c</b>) spectrum shifts with a −10 nm overall ∆ORW and (<b>d</b>) an average tuning rate of −0.8 nm per 100 nm by LTO removal; (<b>e</b>) spectrum shifts with a +6.2 nm overall ∆ORW and (<b>f</b>) an average tuning rate of +1.6 nm per 100 nm by SiN deposition.</p>
Full article ">Figure 5 Cont.
<p>COMSOL simulation of ORW tuning by coating or etching the top DBR diaphragm: (<b>a</b>) the FEM model built in COMSOL Multiphysics; (<b>b</b>) cross-section view of the simulated pressure-deflected diaphragm before tuning; (<b>c</b>) spectrum shifts with a −10 nm overall ∆ORW and (<b>d</b>) an average tuning rate of −0.8 nm per 100 nm by LTO removal; (<b>e</b>) spectrum shifts with a +6.2 nm overall ∆ORW and (<b>f</b>) an average tuning rate of +1.6 nm per 100 nm by SiN deposition.</p>
Full article ">Figure 6
<p>Fabrication process of a SMOUT array: (<b>a</b>) PECVD of the bottom DBR on a glass substrate; (<b>b</b>) RF sputtering and lithography patterning of a sacrificial layer on the bottom DBR; (<b>c</b>) PECVD of the top DBR; (<b>d</b>) etching holes opened through the top DBR by RIE to partially expose the sacrificial layer; (<b>e</b>) wet etching of the sacrificial layer to release the top DBR as the diaphragm, which is still linked to the bottom DBR at the edges; (<b>f</b>) LPCVD of LTO for cavity sealing in a vacuum; ORW tuning by (<b>g</b>) LTO etching or (<b>h</b>) SiN PECVD; (<b>i</b>) a photo (under the microscope) of the tuned SMOUT array after LTO wet etching; and (<b>j</b>) a zoom-in view of one SMOUT element with four etching holes at the corners.</p>
Full article ">Figure 7
<p>(<b>a</b>) Reflectance spectrum shifts with overall −12 nm ∆ORW and (<b>b</b>) an average tuning rate of −1.0 nm per 100 nm by LTO removal. (<b>c</b>) The spectrum shifts with overall +7 nm ∆ORW and (<b>d</b>) an average tuning rate of +1.7 nm per 100 nm by SiN deposition. The error bars in (<b>b</b>,<b>d</b>) indicate the ORW deviation among the five tested elements.</p>
Full article ">Figure 8
<p>The LD-based setup for characterizing the ORW-tuned SMOUT array (Circ: fiber circulator; FC: fiber collimator; DM: dichroic mirror; BS: beam sampler; PD: photodetector; P/R: pulser/receiver; DAQ: data acquisition board; OL: objective lens; MR: mirror).</p>
Full article ">Figure 9
<p>(<b>a</b>) The map and (<b>b</b>) corresponding histogram showing spatial distribution of signal amplitude from elements in the 1.2 × 1.2 <math display="inline"><semantics> <mrow> <mi mathvariant="normal">c</mi> <msup> <mrow> <mi mathvariant="normal">m</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msup> </mrow> </semantics></math> center region of the SMOUT array.</p>
Full article ">Figure 10
<p>(<b>a</b>) NEP map of nine SMOUT elements within the central 12 × 12 <math display="inline"><semantics> <mrow> <mi mathvariant="normal">m</mi> <msup> <mrow> <mi mathvariant="normal">m</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msup> </mrow> </semantics></math> area of the ORW-tuned array; (<b>b</b>) the measured linearity of the ultrasound response of the SMOUT array. Error bars in (<b>b</b>) indicate the deviation in the signal amplitude among different elements.</p>
Full article ">Figure 11
<p>(<b>a</b>) A typical PA response of an element from black human hair as the target; (<b>b</b>) the FFT spectrum of the PA signal in (<b>a</b>) to characterize the frequency response of the tuned SMOUT array.</p>
Full article ">Figure 12
<p>Fluctuations in the normalized signal amplitude received by the tuned SMOUT array due to (<b>a</b>) ambient temperature drift from 25 °C to 55 °C and (<b>b</b>) continuous immersion in water for one week. Error bars indicate the deviation in the signal amplitude among different elements.</p>
Full article ">Figure 13
<p>Imaging setup for PACT experiments.</p>
Full article ">Figure 14
<p>(<b>a</b>) Imaging setup for 2D PACT with a strand of black hair as the target, where the scanning path of the interrogation laser spot is marked by the yellow dashed arrow; (<b>b</b>) reconstructed B-scan image from the 1D scanning; (<b>c</b>) reconstructed 2D image of the black hair, the profile of which along (<b>d</b>) the Z and (<b>e</b>) X axes is used to evaluate the axial and lateral resolution, respectively. Red arrows in (<b>a</b>) represent 10 mm length in the corresponding axes.</p>
Full article ">Figure 14 Cont.
<p>(<b>a</b>) Imaging setup for 2D PACT with a strand of black hair as the target, where the scanning path of the interrogation laser spot is marked by the yellow dashed arrow; (<b>b</b>) reconstructed B-scan image from the 1D scanning; (<b>c</b>) reconstructed 2D image of the black hair, the profile of which along (<b>d</b>) the Z and (<b>e</b>) X axes is used to evaluate the axial and lateral resolution, respectively. Red arrows in (<b>a</b>) represent 10 mm length in the corresponding axes.</p>
Full article ">Figure 15
<p>(<b>a</b>) Three dot-shaped pencil leads as imaging targets (marked by red dashed circles) for 3D PACT; (<b>b</b>) the reconstructed 3D image of the targets (marked by white dashed circles) after 2D scanning the central 3 × 3 <math display="inline"><semantics> <mrow> <mi mathvariant="normal">c</mi> <msup> <mrow> <mi mathvariant="normal">m</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msup> </mrow> </semantics></math> region of the tuned SMOUT array. Red arrows in (<b>a</b>,<b>b</b>) indicate 10 mm length in the corresponding axes.</p>
Full article ">
Back to TopTop