-
Precise measurement of blood flow is of vital importance in studying the formation of thrombus and atherosclerotic plaque. However, conventional color Doppler methods are limited to obtaining the velocity component along the ultrasound beam and have poor accuracy. Several Doppler flow imaging methods based on the plane wave emission can estimate the blood velocity vectors and visualize hemodynamic parameters, which provide more detailed blood flow information and effectively improve the capability of clinical diagnosis treatment. Considering the low accuracy of the Doppler flow methods for measuring velocity in complex flow fields, an optimization technique is used to improve the imaging quality and the accuracy of velocity estimation. In this study we propose a modified vector Doppler method through combining multi-angle compound technique, to reconstruct blood velocity vectors of carotid bifurcations obtained from 3D printing. Since the multi-angle compound technology can effectively improve the quality of imaging, this technology is applied to Doppler imaging to achieve high-accuracy velocity estimation. It can significantly reduce the velocity estimation errors. Comparing the velocity estimation accuracy of different angle compound numbers (n = 1, 3, 5, and 7) in the simulation, it is found that the accuracy of velocity estimation increases with angle compound increasing. Beside, the 5-angle compound method is more robust for velocity estimation and can obtain higher frames. The experiments were carried out using a programmable ultrasonic array system and a high-frequency linear array transducer L12-5c with a central frequency of 8.125 MHz. The sample rate is set to be 31.25 MHz. The imaging results of carotid bifurcation also show that the vector Doppler based on 5-angle compound can obtain a clear image of intravascular vector flow, which is beneficial to the identifying of complex flow state, and realize intravascular dynamic imaging. Especially, it can capture the vortex phenomenon in the blood stream. The quantitative results indicate that this method significantly reduces the error between the flow calculation results and the reference results, making the estimation results more accurate. In conclusion, the vector Doppler method based on multi-angle compound has the good performance of visualizing complex blood flow and calculating hemodynamic parameters. It also provides the reference for the diagnosis of cardiovascular disease and the research of flow imaging methods.
-
Keywords:
- multi-angle compound /
- ultrasonic Doppler /
- carotid bifurcation /
- vector flow imaging
- PACS:
89.20.Hh (World Wide Web, Internet) 1. 引 言
脉冲光纤激光在工业加工、生物成像、医学检测、现代通信等领域需求广泛, 一直受到国内外广泛关注[1-3]. 自从2010年石墨烯可饱和吸收(saturable absorber, SA)调制器件用于掺铒光纤环形腔, 产生了1.5 μm的脉冲激光以来[4,5], 二维材料因其制备简便、非线性吸收可控、带隙可调, 已广泛应用于光纤脉冲激光产生[6-10]. 二维材料中, 含磷材料是重要成员, 含磷一元二维材料黑磷、二元二维材料砷磷等已广泛应用到半导体行业, 并作为可饱和吸收体在不同波长上获得了不同类型的脉冲激光[11-14]. 相比于一元和二元的含磷二维材料, 三元的过渡金属硫代亚磷酸盐(MPS3)二维材料化学多样性高和结构复杂, 表现出新颖的电、光和磁特性[15-17]. 作为含磷家族成员, MPS3不仅具备黑磷优异的工作性能, 还解决了稳定性难题, 能充分发挥磷元素电子施主、化学亲和力、高晶格匹配能力等优势, 且随着过渡金属M的变化, MPS3拥有丰富的光学带隙, 广泛应用于低维材料非线性光学和器件[17-21].
作为二维层状结构低维材料, MPS3由三种元素组成[15,22], 其中过渡金属元素M主要为Fe, Mn, Ni, Zn, Cd, V等. MnPS3作为典型的MPS3: 1)具有层状单斜晶体结构, 每个晶胞中含有2个Mn2+离子, 1个
P2S4−6 基团, 为六方晶格; 2)各向异性, 且层间有较强的磁相互作用; 3)拥有中等的半导体带隙, 在光电探测、二维磁性、光催化等领域应用广泛, 是当前MPS3家族的研究热点之一[23,24]. 20世纪80年代, MnPS3的材料结构和磁性首次报道后, MnPS3在光催化、磁性已开展许多研究工作[19,25-27]. 关于MnPS3低维尺度如少层纳米片的研究还处于起步阶段, 特别是MnPS3可饱和吸收特性较少应用到脉冲激光产生, 目前还没有MnPS3纳米片作为可饱和吸收体产生锁模脉冲的报道.本文采用化学气相传输方法制备MnPS3单晶, 并优化机械剥离方法, 制备MnPS3可饱和吸收体(MnPS3-SA)光纤调制器件. 以MnPS3-SA 为调制器件, 掺铒光纤为增益光纤, 实现稳定的自启动双波长锁模输出, 实现双波长输出对激光器的实际应用有重要意义[28].
2. MnPS3可饱和吸收体的制备和表征
与黑砷磷的制备方法类似, 使用化学气相传输法生长三元含磷二维材料MnPS3单晶[17,29], 但区别在于矿化剂和输运剂的选择, 化学气相传输法原理跟化学气相沉积类似, 不同于一般化学气相沉积方法有裂解反应物过程, 化学气相传输方法则是把粉末原料蒸发后跟随载气沉积在低温区的衬底. 实验使用的纯锰粉(Mn, 99.99%)、红磷(RP, 99.999%)、硫(S, 99.99%)和碘(I2, 1 mg/mL)按照特定比例混合均匀并称重2 g作为反应物(均购于Sigma-Aldrich公司), 以适量的碘作为输运剂, 放置于尖头石英安瓿内, 并在1 × 10–3 Pa的真空条件下真空封管. 放置在双温区管式炉中的石英安瓿, 处于700 ℃/650 ℃的双区温度梯度环境中, 在I2和温度梯度的共同作用下通过输运作用在石英管冷端获得干净的厘米级绿色六角形态片状MnPS3单晶(如图1(a)和图1(b)所示).
图 1 MnPS3晶体生长及表征 (a)化学气相传输法制备MnPS3晶体的工艺流程示意图; (b) MnPS3单晶的照片; (c) MnPS3单晶的拉曼光谱Fig. 1. Characteristics of MnPS3 crystals: (a) Chemical vapor transport method; (b) picture of MnPS3; (c) Raman spectrum for MnPS3可饱和吸收体的制备方法目前主要有三明治型、倏逝波型、可饱和吸收镜型等[30-32]. 使用优化的胶带法机械剥离MnPS3单晶, 而后转移到光纤跳线端帽上, 制备光纤脉冲激光所需的类三明治式结构MnPS3-SA调制器件. 采用同样的机械剥离方法, 将MnPS3-SA转移到特制Si衬底上(表面有285 nm SiO2)表征. 图1(c)给出了使用共聚焦激光拉曼光谱仪(Raman, LabRAM HR Evolution)测得的MnPS3-SA拉曼光谱, 155, 225, 274和568 cm–1处的拉曼峰对应Eg振动模式, 在246, 384和582 cm–1处观察到
A1g 振动模式, 与之前报道MnPS3典型特征峰相符[33,18].图2为使用Quanta 400 FEG场发射扫描电子显微镜(scanning electron microscope, SEM)测得的MnPS3-SA微观结构. 图2(a)为MnPS3-SA的场发射电子显微镜形貌图像, 图像显示样品表面没有杂质、光滑平整, 样品具有MnPS3低维材料特有的六边形和层状形貌. 利用SEM配套的能量散射X射线(energy dispersive X-ray spectroscopy, EDX)表征MnPS3选定区域的元素成分和原子含量, 并做元素面扫描, EDX能谱分析得到Mn原子、P原子和S原子在均匀分布且原子比为19.97%, 20.89%和59.13% (约1∶1∶3), 满足化学式比例(如图2所示).
图 2 MnPS3-SA的SEM表征 (a)随机选取的样品SEM图像和元素分析表; (b)−(d) Mn, P和S的EDX元素面扫描Fig. 2. SEM characteristics of MnPS3 -SA: (a) SEM image of a randomly selected MnPS3 flake, and elemental analysis of this sample; (b)−(d) EDX element mappings for Mn, P, and S.采用透射电子显微镜(transmission electron microscope, TEM)表征MnPS3样品的晶体结构, 结果如图3所示. MnPS3透射样品如图3(a)所示. 图3(b)和图3(c)分别展示了高分辨率透射电子显微镜(high-resolution transmission electron microscope, HRTEM)图像和区域电子衍射(selected area electron diffraction, SAED)图, 图示平面距离为0.29 nm的清晰晶格条纹对应于(
13−1 )晶面. 这些表征, 证明了本方法制备MnPS3样品结构均一、组分准确.
图 3 MnPS3纳米片的TEM表征 (a) MnPS3纳米片形貌; (b) MnPS3纳米片的HRTEM像; (c) SAED图Fig. 3. TEM characterization of MnPS3 nanosheets: (a) TEM image of a MnPS3 nanosheet on a copper grid; (b) the HRTEM image of the MnPS3 nanosheet; (c) the corresponding SAED showing its single crystal nature.3. 光纤激光器环形腔结构
将MnPS3-SA调制器件接入光纤环形腔构建如图4所示实验装置: 环形腔使用976 nm稳波长激光二极管(laser diode, LD)作为抽运源, 并通过980/1550波分复用器(wavelength division multiplexing, WDM)接入环形腔; 增益介质使用3.2 m长的掺铒光纤(6/125); 为了保证环形腔中能量单向传输, 在掺铒光纤后端接入偏振无关隔离器(polarization independent isolator, PI-ISO); 为了管理腔内色散, 在PI-ISO后连接了约20 m长的单模光纤(SMF-28); 采用偏振控制器(polarization controller, PC)对偏振态进行调节; 采用20∶80的光耦合器(optical coupler, OC)从激光腔中耦合出20%的腔内能量. 光纤环形腔的总长度约为35 m, 在耦合器的输出端, 使用2 GHz的光电探头将光信号转化为电信号, 并使用1 GHz的带射频分析功能示波器记录输出激光的时域和频域特性; 利用最高分辨率0.02 nm的精密光谱仪记录输出激光的光谱; 使用毫瓦功率计记录激光功率.
基于光纤环形腔实验装置, 随着抽运光功率增加, 当输入抽运功率增加到70 mW时, 在示波器上观测到激光输出的脉冲时域型号信号, 并在70—270 mW的抽运范围内连续变化时可以观测到被动锁模激光输出. 输出激光功率和抽运功率的关系如图5(a)所示, 输出功率随抽运光功率的增加而线性增加, 最大输出功率为27.2 mW.
图 5 基于MnPS3-SA的脉冲光纤激光器的性能 (a)输出功率与抽运功率的关系; (b)输出光谱; (c)脉冲序列; (d)脉冲脉宽; (e) 0−10 MHz射频信号; (f)射频基频信号Fig. 5. Performances of the pulse fiber laser based on MnPS3-SA: (a) The output power versus the pump power; (b) output optical spectrum; (c) the pulse trace; (d) the duration of single pulse; (e) the radio frequency spectrum from 0−10 MHz; (f) the radio frequency spectrum with ～64 dB (inset).利用高精度光谱仪记录输出激光的光谱, 显示输出脉冲激光为双波长. 记录抽运功率为180 mW时典型输出光谱如图5(b)所示, 两个波长峰值分别为1565.19 nm和1565.63 nm, 双波长间隔为0.44 nm. 光谱展宽不明显, 且没有克利边带. 同时记录此抽运功率下激光器输出激光的时域特性如图5(c)和图5(d)所示, 输出锁模脉冲脉冲间隔为196.1 ns (图5(c)), 脉冲宽度为3.8 ns (图5(d)). 利用示波器的频谱分析功能分析脉冲激光的频谱特性如图5(e)和图5(f)所示, 图5(e)为射频基频信号, 输出脉冲射频信号基频信噪比超过64 dB, 显示脉冲具有较高的稳定性.
在相同条件下记录抽运功率为70, 120, 170, 220和270 mW时激光器输出光谱和功率. 图6(a)给出了不同抽运功率下激光器的发射光谱, 双波长稳定在1565.18 nm (± 0.01 nm)和1565.64 nm (± 0.02 nm), 双波长间隔0.46 nm (± 0.02 nm), 如图6(b)所示. 记录不同抽运功率下脉冲激光的重复频率(图6(c)), 输出频率稳定在5.109 MHz (± 0.011 MHz). 随抽运功率的增加, 长波长强度呈非线性减弱, 双波长效果逐渐减弱, 在270 mW抽运功率下, 下降到第一个峰值的29.7%. 因为抽运功率的增加会导致线性增益的增加, 使得腔内模式更容易达到可饱和吸收体的吸收光强. 因此双波长锁模输出需要在70—270 mW功率区间内实现, 且功率越低, 双波长效果越好. 此双波长光纤脉冲激光在不同抽运功率下发射激光具有稳定的输出波长和重复频率.
图 6 基于MnPS3-SA的脉冲光纤激光器在70, 120, 170, 220和270 mW抽运功率下的(a)光谱、(b)波长和(c)频率特性Fig. 6. Performances of the pulse fiber laser based on MnPS3-SA with the pump power at 70, 120, 170, 220, and 270 mW pump power: (a) Spectrum; (b) wavelength; (c) frequency.实验过程中为了测试被动锁模激光输出的长时间稳定性, 在相同的环境条件下, 使用相同测试方法记录第1, 7, 8, 11, 12天光纤激光器在抽运功率为180 mW时的出光特性. 实验显示, 光纤激光正当其可以实现自启动锁模, 输出激光的波长和输出功率特性如图7所示. 从图7(a)可看出, 输出光谱的2个输出波长稳定在1565.19 nm (±0.01 nm)和1565.58 nm (±0.03 nm), 双波长间隔0.39 nm (± 0.04 nm), 输出功率为16.67 mW (± 0.10 mW, ± 0.6%). 实验数据证实该激光器长时间工作时, 发射波长、波长间隔和频率均保持高稳定性. 分析误差增大的主要原因是谐振腔在实验环境变化(温度、振动等)的影响下, 等价腔长可能发生漂移以及温度导致折射率的微小变化.
图 7 基于MnPS3-SA的脉冲激光长时间工作稳定性 (a)第1, 7, 8, 11, 12 天的输出光谱; (b)中心波长; (c)输出功率Fig. 7. Output spectrum of the EDFL based on MnPS3-SA: (a) Output spectrum recorded on 1st, 7th, 8th, 11th, 12th day; (b) wavelength peak position; (c) output power.实验中为验证光纤环形腔是否有类似的可饱和吸收特性, 将MnPS3 可饱和吸收体从腔内移除, 改变抽运功率和调节偏振控制器都没有观察到脉冲和双波长现象, 验证MnPS3是产生双波长锁模的惟一因素. 同时, 去除MnPS3 可饱和吸收体后, 激光器输出为较好的单波长光纤激光.
MnPS3作为一种可饱和吸收材料, 插入在谐振腔中可被动地周期性调制谐振腔的内部吸收损耗, 来实现激光器的锁模运转, 其具体作用机理是: 初始激光脉冲包含了所有模式, 彼此之间相位无规则分布, 此时输出光强随机, 未实现锁模. MnPS3的吸收特性导致对特定波长的吸收弱. 光强大于可饱和吸收光强时脉冲被吸收, 在增益介质的线性放大下, 对强脉冲的吸收弱, 对弱脉冲吸收强, 弱脉被抑制, 强脉冲高速增长, 最终得到双波长的锁模输出.
4. 结 论
本文采用化学气相传输方法制备了MnPS3单晶, 并利用优化胶带法机械剥离制备光纤脉冲激光所需的MnPS3-SA调制器件, 并对其进行材料表征和分析, MnPS3-SA调制器件光纤激光振荡器实现了稳定的全光纤被动锁模激光自启动输出. 基于该环形腔, 可实现最大输出功率为27.2 mW, 双波长稳定在1565.18 nm (± 0.01 nm)和1565.64 nm (± 0.02 nm), 双波长间隔0.46 nm (± 0.02 nm), 基频信噪比约为64 dB, 频率为5.109 MHz (± 0.011 MHz)的脉冲输出. 此外, MnPS3-SA光纤激光振荡器可实现280 h以上稳定自启动锁模, 输出双波长稳定在1565.19 nm (± 0.01 nm)和1565.58 nm (± 0.03 nm), 双峰间隔0.39 nm (± 0.04 nm), 输出的功率为16.67 mW (± 0.10 mW, ± 0.6%). 实验表明, 二维材料MnPS3作为可饱和吸收体, 可以提供一种新型方法实现低成本、稳定性好的双波长的锁模脉冲输出, 在光纤通信、工业加工、医疗器械等领域都有潜在的应用价值.
[1] Stein J H, Korcarz C E, Hurst R T, Lonn E, Kendall C B, Mohler E R, Najjar S S, Rembold C M and Post W S 2008 J. Am. Soc. Echocardiogr. 21 93
Google Scholar
[2] Kornblum H I, Araujo D M, Annala A J, Tatsukawa K J, Phelps M E, Cherry S R 2000 Nat. Biotechnol. 18 655
Google Scholar
[3] Ogawa S, Lee T M, Kay A R, Tank D W 1990 Proc. Natl. Acad. Sci. U. S. A. 87 9868
Google Scholar
[4] Denarie B, Tangen T A, Ekroll I K, Rolim N, Torp H, Bjastad T, Lovstakken L 2013 IEEE Trans. Med. Imaging 32 1265
Google Scholar
[5] Cloutier G, Zhao Q, Durand L G, Teh B G 1996 IEEE Trans. Biomed. Eng. 43 441
Google Scholar
[6] Tanter M, Bercoff J, Sandrin L, Fink M 2002 IEEE Trans. Ultrason. Ferroelectr. Freq. Control 49 1363
Google Scholar
[7] Mace E, Montaldo G, Cohen I, Baulac M, Fink M, Tanter M 2011 Nat. Methods 8 662
Google Scholar
[8] 臧佳琦, 许凯亮, 韩清见, 陆起涌, 梅永丰, 他得安 2020 物理学报 70 114303
Google Scholar
Zang J Q, Xu K L, Han Q J, Lu Q Y, Mei Y F, Ta D A 2020 Acta. Phys. Sin 70 114303
Google Scholar
[9] Fredriksen T D, Avdal J, Ekroll IK, Dahl T, Løvstakken L, Torp H 2014 IEEE Trans. Ultrason. Ferroelect. Freq. Control 61 1161
Google Scholar
[10] Tortoli P, Dallai A, Boni E, Francalanci L, Ricci S 2010 Ultrasound Med. Biol. 36 488
Google Scholar
[11] Jensen J, Hoyos C A V, Stuart M B, Ewertsen C, Nielsen M B, Jensen J A 2017 IEEE Trans. Ultrason. Ferroelect. Freq. Control 64 1050
Google Scholar
[12] Behar V, Adanm D, Friedman Z 2003 Ultrasonics 41 377
Google Scholar
[13] Lovstakken L, Torp H 2010 IEEE International Ultrasonics Symposium San Diego, CA, USA, Oct 11–14, 2010 p1198
[14] Wan M X, Gong X Z, Qian M 1999 IEEE Trans. Biomed. Eng. 46 1074
Google Scholar
[15] Steel R, Fish P J 2002 IEEE Trans. Ultrason. Ferroelect. Freq. Control 49 1375
Google Scholar
[16] Girault J M, Kouame D, Ouahabi A 2000 Ultrasonics 38 682
Google Scholar
[17] Peronneau P, Bournat J P, Bugnon A, Barbet A, Xhaard M 1974 Cardiovascular Applications of Ultrasound Netherlands, North Holland, 1974 p66
[18] Tortoli P, Bambi G, Ricci S 2006 IEEE Trans. Ultrason. Ferroelect. Freq. Control 53 1425
Google Scholar
[19] Dunmire B, Beach K W, Labs K H, Plett M, Strandness D E 2000 Ultrasound Med. Biol. 26 1213
Google Scholar
[20] Phillips P J, Kadi A P, Von Ramm O T 1995 Ultrasound Med. Biol. 21 217
Google Scholar
[21] Scabia M, Calzolai M, Capineri L, Masotti L, Fort A 2000 Ultrasound Med. Biol. 26 121
Google Scholar
[22] Tsang I K H, Yiu B Y S, Yu A C H 2009 IEEE International Ultrasonics Symposium Rome, Italy, Sept 20–23, 2009 p1387
[23] Ricci S, Ramalli A, Bassi L, Boni E, Tortoli P 2017 IEEE Trans. Ultrason. Ferroelect. Freq. Control 65 201
Google Scholar
[24] Bjaerum S, Torp H, and Kristoffersen K, 2002 IEEE Trans. Ultrason. Ferroelect. Freq. Control 49 204
Google Scholar
[25] Revellin R, Rousset F, Baud D, Bonjour J 2009 Theor. Biol. Med. Modell. 6 1
Google Scholar
-
图 1 方法流程图
Figure 1. Flow chart of the proposed method.
图 2 (a)平面波传输示意图; (b)无偏转角的平面波传输路径; (c)偏转角为
α 时声波传输的路径Figure 2. (a) Plane wave transmission; (b) time delays for a plane wave emission and reception; (c) time delays for a plane wave of the angle
α .
图 3 交叉波速示意图.
Figure 3. Schematic diagram of crossed-beam vector Doppler.
图 4 实验系统原理图
Figure 4. Schematic diagram of the experimental system.
图 5 颈动脉分叉实验装置图
Figure 5. Experimental setup of carotid bifurcation.
图 6 多角度复合多普勒成像结果 (a) 单角度复合; (b) 3个角度复合; (c) 5个角度复合; (d) 7个角度复合
Figure 6. Multi-angle compound Doppler imaging results: (a) Single angle compound; (b) 3 angles compound; (c) 5 angles compound; (d) 7 angles compound.
图 7 多角度复合多普勒估算速度与标准参考值的散点分布图 (a)单角度; (b) 3个角度复合; (c) 5个角度复合; (d) 7个角度复合
Figure 7. Scatter plots showing the relation between reference and estimated velocities: (a) Single angle; (b) 3 angles compound; (c) 5 angles compound; (d) 7 angles compound.
图 8 血管沿深度方向的速度分布曲线.
Figure 8. Velocity distribution along the depth direction of the blood vessel.
图 9 单角度平面波发射的颈动脉分叉血流成像结果 (a)偏转接收角度为12°时三种流量入口条件的彩色多普勒图像; (b)偏转接收角度为–12°的彩色多普勒图像; (c)矢量合成得到的速度矢量图像; (d)分叉处放大图
Figure 9. Blood flow imaging of carotid bifurcation by single angle plane wave composite imaging: (a) Color Doppler imaging of three flow inlet conditions at 12° deflection receiving angle; (b) color Doppler imaging with a deflected reception angle of –12°; (c) the dual-mode imaging of velocity vector and B mode; (d) Enlarged view of partial bifurcation.
图 10 单角度平面波成像和5个角度复合的颈动脉分叉成像结果对比 (a), (b)和(c)分别为单角度平面波发射情况下流量为200, 250和300 mL/min的血流矢量成像结果; (e), (f)和(d)为5个角度复合的成像结果. 白色线框表示用于估算平均速度的区域
Figure 10. Blood flow imaging of carotid bifurcation by single plane wave and 5-angle plane wave compound: (a), (b), (c) The imaging results of 200, 250 and 300 mL/min under the condition of single plane wave emission; (e), (f), (d) the imaging results of 5-angle composite. The white box represents the area used to estimate the average speed.
表 1 声场仿真的参数设置
Table 1. Simulation parameters setting.
参数 值 发射阵元数 128 接收阵元数 128 中心频率/MHz 5 阵元宽度/mm 0.208 阵元高度/mm 4.5 阵元间距/mm 0.35 声速/(m·s–1) 1540 幅度变迹 Hanning 激励脉冲 4-period sinusoid 最大脉冲重复频率/kHz 15 角度复合数 3, 5, 7 采样频率/MHz 100 直径/mm 10 峰值速度/(m·s–1) 1 
DownLoad: CSV
表 2 多普勒速度估算的标准偏差
Table 2. Standard deviation of Doppler velocity estimation.
Nangles 1 3 5 7 归一化标准差 0.164 0.073 0.067 0.0659
DownLoad: CSV
表 3 矢量多普勒方法计算流量与设定值的误差
Table 3. The error between the estimated flow rate and the reference value.
Vvolume/(mL·min–1) 200 250 300 Error(Nangles = 1) 0.125 0.208 0.107 Error(Nangles = 5) 0.120 0.080 0.053
DownLoad: CSV
-
[1] Stein J H, Korcarz C E, Hurst R T, Lonn E, Kendall C B, Mohler E R, Najjar S S, Rembold C M and Post W S 2008 J. Am. Soc. Echocardiogr. 21 93
Google Scholar
[2] Kornblum H I, Araujo D M, Annala A J, Tatsukawa K J, Phelps M E, Cherry S R 2000 Nat. Biotechnol. 18 655
Google Scholar
[3] Ogawa S, Lee T M, Kay A R, Tank D W 1990 Proc. Natl. Acad. Sci. U. S. A. 87 9868
Google Scholar
[4] Denarie B, Tangen T A, Ekroll I K, Rolim N, Torp H, Bjastad T, Lovstakken L 2013 IEEE Trans. Med. Imaging 32 1265
Google Scholar
[5] Cloutier G, Zhao Q, Durand L G, Teh B G 1996 IEEE Trans. Biomed. Eng. 43 441
Google Scholar
[6] Tanter M, Bercoff J, Sandrin L, Fink M 2002 IEEE Trans. Ultrason. Ferroelectr. Freq. Control 49 1363
Google Scholar
[7] Mace E, Montaldo G, Cohen I, Baulac M, Fink M, Tanter M 2011 Nat. Methods 8 662
Google Scholar
[8] 臧佳琦, 许凯亮, 韩清见, 陆起涌, 梅永丰, 他得安 2020 物理学报 70 114303
Google Scholar
Zang J Q, Xu K L, Han Q J, Lu Q Y, Mei Y F, Ta D A 2020 Acta. Phys. Sin 70 114303
Google Scholar
[9] Fredriksen T D, Avdal J, Ekroll IK, Dahl T, Løvstakken L, Torp H 2014 IEEE Trans. Ultrason. Ferroelect. Freq. Control 61 1161
Google Scholar
[10] Tortoli P, Dallai A, Boni E, Francalanci L, Ricci S 2010 Ultrasound Med. Biol. 36 488
Google Scholar
[11] Jensen J, Hoyos C A V, Stuart M B, Ewertsen C, Nielsen M B, Jensen J A 2017 IEEE Trans. Ultrason. Ferroelect. Freq. Control 64 1050
Google Scholar
[12] Behar V, Adanm D, Friedman Z 2003 Ultrasonics 41 377
Google Scholar
[13] Lovstakken L, Torp H 2010 IEEE International Ultrasonics Symposium San Diego, CA, USA, Oct 11–14, 2010 p1198
[14] Wan M X, Gong X Z, Qian M 1999 IEEE Trans. Biomed. Eng. 46 1074
Google Scholar
[15] Steel R, Fish P J 2002 IEEE Trans. Ultrason. Ferroelect. Freq. Control 49 1375
Google Scholar
[16] Girault J M, Kouame D, Ouahabi A 2000 Ultrasonics 38 682
Google Scholar
[17] Peronneau P, Bournat J P, Bugnon A, Barbet A, Xhaard M 1974 Cardiovascular Applications of Ultrasound Netherlands, North Holland, 1974 p66
[18] Tortoli P, Bambi G, Ricci S 2006 IEEE Trans. Ultrason. Ferroelect. Freq. Control 53 1425
Google Scholar
[19] Dunmire B, Beach K W, Labs K H, Plett M, Strandness D E 2000 Ultrasound Med. Biol. 26 1213
Google Scholar
[20] Phillips P J, Kadi A P, Von Ramm O T 1995 Ultrasound Med. Biol. 21 217
Google Scholar
[21] Scabia M, Calzolai M, Capineri L, Masotti L, Fort A 2000 Ultrasound Med. Biol. 26 121
Google Scholar
[22] Tsang I K H, Yiu B Y S, Yu A C H 2009 IEEE International Ultrasonics Symposium Rome, Italy, Sept 20–23, 2009 p1387
[23] Ricci S, Ramalli A, Bassi L, Boni E, Tortoli P 2017 IEEE Trans. Ultrason. Ferroelect. Freq. Control 65 201
Google Scholar
[24] Bjaerum S, Torp H, and Kristoffersen K, 2002 IEEE Trans. Ultrason. Ferroelect. Freq. Control 49 204
Google Scholar
[25] Revellin R, Rousset F, Baud D, Bonjour J 2009 Theor. Biol. Med. Modell. 6 1
Google Scholar
下载:
DownLoad:
Catalog
Metrics
- Abstract views: 7991
- PDF Downloads: 205
