翻译：徐祗彪 徐医2022级麻醉学研究生
审校：赵林林 徐医附院麻醉科
A wearable cardiac ultrasound imager
一种可穿戴式心脏超声成像仪
Continuous imaging of cardiac functions is highly desirable for the assessment of long-term cardiovascular health, detection of acute cardiac dysfunction and clinical management of critically ill or surgical patients 1–4 . However, conventional non-invasive approaches to image the cardiac function cannot provide continuous measurements owing to device bulkiness 5–11 , and existing wearable cardiac devices can only capture signals on the skin 12–16 . Here we report a wearable ultrasonic device for continuous, real-time and direct cardiac function assessment. We introduce innovations in device design and material fabrication that improve the mechanical coupling between the device and human skin, allowing the left ventricle to be examined from different views during motion. We also develop a deep learning model that automatically extracts the left ventricular volume from the continuous image recording, yielding waveforms of key cardiac performance indices such as stroke volume, cardiac output and ejection fraction. This technology enables dynamic wearable monitoring of cardiac performance with substantially improved accuracy in various environments.
心功能连续成像对于长期心血管健康评估、急性心功能不全检测以及危重患者或外科患者的临床管理非常有价值1-4。然而，传统的非侵入性心功能成像方法由于设备的体积较大，无法提供连续的测量5-11，现有的可穿戴心脏设备只能捕获皮肤上的信号12-16。该研究报道了一种可穿戴超声设备，用于连续、实时和直接的心功能评估。该研究介绍了设备设计和材料制造方面的创新，改善了设备和人体皮肤之间的机械耦合，允许在运动过程中从不同的角度检查左心室。该研究还开发了一个深度学习模型，从连续图像记录中自动提取左心室容量，生成关键心脏性能指标的波形，如每搏量、心输出量和射血分数。该技术能够在各种环境中对心脏性能进行动态可穿戴监测，显著提高了准确性。
The device features piezoelectric transducer arrays, liquid metal composite electrodes and triblock copolymer encapsulation, as shown by the exploded schematics (Fig. 1a, left, Extended Data Fig. 1 and Supplementary Discussion 3). The device is built on styrene–ethylene–butylene–styrene (SEBS). To provide a comprehensive view of the heart, standard clinical practice is to image it in two orthogonal orientations by rotating the ultrasound probe 17 . To eliminate the need for manualrotation, we designed the device with an orthogonal configuration(Fig. 1a, right and Supplementary Videos 1 and 2). Each transducer element consisted of an anisotropic 1-3 piezoelectric composite and a silver-epoxy-based backing layer 18,19 . To balance the penetration depth and spatial resolution, we chose a centre resonant frequency of 3 MHz for deep tissue imaging 19 (Supplementary Fig. 1). The array pitch was 0.4 mm (that is, 0.78 ultrasonic wavelengths), which enhances lateral resolutions and reduces grating lobes 20 .
该装置具有压电换能器阵列、液态金属复合电极和三嵌段共聚物封装功能，如分解图所示（图1a，左，扩展数据图1和补充讨论3）。该装置是用苯乙烯-乙烯-丁烯-苯乙烯（SEBS）建造。为了提供一个全面的心脏视图，标准的临床实践是通过旋转超声探头在两个正交的方向上成像17。为了消除手动旋转的需要，我们设计了一个正交配置的设备（图1a，右和补充视频1和2）。每个换能器元件由一个各向异性的1-3压电复合材料和一个银-环氧基背衬层18、19组成。为了平衡穿透深度和空间分辨率，我们选择了一个3 MHz的中心共振频率用于深度组织成像19（补充图1）。阵列间距为0.4mm（即0.78个超声波波长），提高了横向分辨率，减少了光栅叶20。
To individually address each element in such a compact array, we made high-density multilayered stretchable electrodes based on a composite of eutectic gallium–indium liquid metal and SEBS 21 . The composite is highly conductive and easy to pattern (Fig. 1b,c, Supplementary Figs. 2–4 and Methods). Lap shear measurements show that the interfacial bonding strength is about 250 kPa between the transducer element and the SEBS substrate, and about 236 kPa between the transducer element and the composite electrode (Fig. 1d and Supplementary Fig. 5), which are both stronger than typical commercial adhesives 22 (Supplementary Table 2). The resulting electrode has a thickness of only about 8 μm (Supplementary Figs. 6 and 7). Electromagnetic shielding, also made of the composite, can mitigate the interference of ambient electromagnetic waves, which reduces the noise in the ultrasound radiofrequency signals and enhances the image quality 23 (Supplementary Fig. 8 and Supplementary Discussion 4). The device has excellent electromechanical properties, as determined by its high electromechanical coupling coefficient, low dielectric loss, wide bandwidth and negligible crosstalk (Supplementary Fig. 1 and Methods). The entire device has a low Young’s modulus of 921 kPa, comparable with the human skin modulus 24 (Supplementary Fig. 9). The device exhibits a high stretchability of up to approximately 110% (Fig. 1e and Supplementary Fig. 10) and can withstand various deformations (Fig. 1f). Considering that the typical strain on the human skin is within 20% (ref. 19 ), these mechanical properties allow the wearable imager to maintain intimate contact with the skin over a large area, which is challenging for rigid ultrasound devices 25 .
为了单独处理这样一个紧凑阵列中的每个元件，我们基于共晶镓-铟液态金属和SEBS 21的复合材料制作了高密度多层可拉伸电极。该复合材料具有高导电性，易于形成图案(图1b、c，补充图。2-4和方法)。搭接剪切测量表明，传感器元件和SEBS衬底之间的界面结合强度约250kPa，传感器元件和复合电极之间约236 kPa（图1d和补充图5），都比典型的商业粘合剂强22（补充表2）。所得到的电极厚度仅约为8μm(补充图。6和7）。电磁屏蔽也由该复合材料制成，可以减轻环境电磁波的干扰，降低超声射频信号中的噪声，增强图像质量23（补充图8和补充讨论4）。该装置具有良好的机电性能，包括其高机电耦合系数、低介电损耗、宽带宽和可忽略的串扰（补充图1和方法）。整个装置的杨氏模量较低，为921 kPa，可与人类皮肤模量相当24（补充图9）。该装置具有高达约110%的高拉伸能力（图1e和补充图10)，并能承受各种变形（图1f)。考虑到人类皮肤上的典型应变在20%以内(参考文献。19 )，这些力学特性允许可穿戴成像仪在大范围内与皮肤保持亲密接触，然而这对刚性超声设备来说确是一个挑战25。
Imaging strategies and characterizations
We evaluated the quality of the generated images based on the five most crucial metrics for anatomical imaging: spatial resolutions (axial,lateral and elevational), signal-to-noise ratio, location accuracies (axial and lateral), dynamic range and contrast-to-noise-ratio 26.
成像策略和特征分析
我们基于解剖成像的五个最关键的指标评估了生成图像的质量：空间分辨率（轴向、横向和高程）、信噪比、位置精度（轴向和横向）、动态范围和对比度噪比26。
The transmit beamforming strategy is critical for image quality. Therefore, we compared three distinct strategies: plane-wave, mono-focus and wide-beam compounding. Phantoms containing mono-filament wires were used for this comparison (Supplementary Fig.11, position 1). Among the three strategies, the wide-beam compounding implements a sequence of divergent acoustic waves with a series of transmission angles, and the generated images of each transmission are coherently combined to create a compounding image, which has the best quality with an expanded sonographic window 27 (Fig. 2a,band Supplementary Figs. 12–14). We also used a receive beamforming strategy to further improve the image quality (Supplementary Fig. 15 and Methods). The wide-beam compounding achieves a synthetic focusing effect and, therefore, a high acoustic intensity across the entire insonation area (Fig. 2c and Supplementary Fig. 13), which leads to the best signal-to-noise ratio and spatial resolutions (Fig. 2a, third column, Fig. 2b and Supplementary Fig. 12).
发射波束形成策略对图像质量至关重要。因此，我们比较了三种不同的策略：平面波、单聚焦和宽光束复合。使用含有单灯丝导线的幻像进行比较（补充图11，位置1）。三种策略中，宽波束复合实现一系列不同的声波与一系列的传输角度，和生成的图像每个传输相干组合创建一个复合图像，具有最好的质量与扩大超声窗口27(图2，补充图12–14).我们还使用了接收波束形成策略来进一步提高图像质量（补充图15和方法）。宽光束复合实现了合成聚焦效果，因此在整个共振区域内具有较高的声学强度（图2c和补充图13），从而获得了最佳的信噪比和空间分辨率（图2a，第三列，图2b和补充图12）。
To quantify the device spatialresolutions using the wide-beam compounding strategy, we measured full widths at half maximum from the point spread function curves 28 extracted from the images (Fig. 2a, third and fourth columns and the bottom row and Supplementary Fig. 11, positions 1 and 2). As the depth increases, the elevational resolution deteriorates (Fig. 2d) because the beam becomes more divergent in the elevational direction. Therefore, we integrated six small elements into a long element (Extended Data Fig. 1) to offer better acoustic beam convergence and elevational resolution. The lateral resolution deteriorates only slightly with depth (Fig. 2d) owing to the process of receive beamforming (Methods). The axial resolution remains almost constant with depth (Fig. 2d) because it depends only on the frequency and bandwidth of the transducer array. Similarly, at the same depth, the axial resolution remains consistent with different lateral distances from the central axis of the device, whereas the lateral resolution is the best at the centre, where there is a high overlap of acoustic beams after compounding (Fig. 2e and Methods).
为了使用宽光束复合策略量化设备的空间分辨率，我们从图像中提取的点扩散函数曲线28（图2a，第三和第四列和底部行和补充图11，位置1和2）测量了全宽度。随着深度的增加，高程分辨率下降（图2d）。因此，我们将6个小元件集成到一个长元件中（扩展数据图1），以提供更好的声束收敛性和高程分辨率。由于接收波束形成的过程（方法），横向分辨率仅随着深度的增加而略有下降（图2d)。轴向分辨率几乎与深度保持不变（图2d），因为它只取决于换能器阵列的频率和带宽。同样，在相同的深度下，轴向分辨率与距离器件中心轴的不同横向距离保持一致，而横向分辨率在中心位置最好，在复合后声束有很高的重叠（图2e和方法）。
Another critical metric for imaging is thelocation accuracies. The agreements between the imaging results and the ground truths (the red dots in Fig. 2a) in the axial and lateral directions are 96.01% and 95.90%, respectively, indicating excellent location accuracies (Methods).
成像的另一个关键度量标准是定位精度。成像结果和地面实况（图2a中的红点）在轴向和横向上的一致性分别为96.01%和95.90%，表明具有良好的定位精度（方法）。
Finally, we evaluated the dynamic range and contrast-to-noise ratio of the device using the wide-beam compounding strategy. Phantoms containing cylindrical inclusions with different acoustic impedances were used for the evaluation (Supplementary Fig. 11, position 3). A high acoustic impedance mismatch results in images with high contrast (Fig. 2f). We extracted the average grey values of the inclusion images and performed a linear regression 29 , and determined the dynamic range to be 63.2 dB (Fig. 2g, Supplementary Fig. 16 and Methods), which is well above the 60-dB threshold typically used in medical diagnosis 30 .
最后，我们使用宽光束复合策略评估了该设备的动态范围和对比度噪比。使用含有具有不同声阻抗的圆柱形内含物的幻像进行评估（补充图11，位置3）。高声阻抗失匹配产生高对比度的图像（图2f)。我们提取纳入图像的平均灰度值，并进行线性回归29，确定动态范围为63.2 dB（图2g，补充图16和方法），远高于医学诊断中通常使用的60-dB阈值30。
We selected two regions of interest, one inside and the other outside each inclusion area, to derive the contrast-to-noise ratios 31 , which range from 0.63 to 2.07 (Fig. 2h and Methods). A higher inclusion contrast leads to a higher contrast-to-noise ratio of the image. The inclusions with the lowest contrast (+3 dB or −3 dB) can be clearly visualized, demonstrating the outstanding sensitivity of this device 20 . The performance of the wearable imager is comparable with that of the commercial device (Supplementary Figs. 17 and 18, Extended Data Table 1 and Supplementary Discussion 5).
我们选择了两个感兴趣的区域，一个在每个包含区域内，另一个在每个包含区域外，以得出对比度-噪声比31，其范围从0.63到2.07（图2h和方法）。较高的包含对比度会导致图像的对比度与噪比更高。对比度最低的夹杂物（+3 dB或−3dB）可以清晰地显示出来，显示了该设备突出的灵敏度20。可穿戴成像仪的性能可与商业设备相媲美(补充图。17和18，扩展数据表1和补充讨论5)。
Echocardiography from several views
Echocardiography is commonly used to examine the structural integrity and blood-delivery capabilities of the heart. Uniquely for soft devices, the contours of the human chest cause a non-planar distribution of the transducer elements, which leads to phase distortion and therefore image artefacts 32 . We used a three-dimensional scanner to collect the chest curvature to compensate for element position shifts within the wearable imager and thus correct phase distortion during transmit and receive beamforming (Supplementary Fig. 19, Extended Data Fig. 2 and Supplementary Discussion 6).
超声心动图的几个观点
超声心动图通常用于检查心脏的结构完整性和供血能力。对于软器件来说，人类胸部的轮廓会导致传感器元件的非平面分布，从而导致相位失真，图像伪影32。我们使用三维扫描仪来收集胸部曲率，以补偿可穿戴成像仪内的元件位置偏移，从而纠正发射和接收波束形成过程中的相位失真（补充图19，扩展数据图2和补充讨论6）。
We compared the performance of the wearabledevice with a commercial device in four primary views of echocardiography, in which critical cardiac features can be identified (Extended Data Fig. 3). Figure 3a shows the schematics and corresponding B-mode images of these four views, including apical four-chamber view, apical two-chamber view, parasternal long-axis view and parasternal short-axis view. The difference between the results from the wearable and commercial devices is negligible. The parasternal short-axis view is particularly useful for evaluating the contractile function of the myocardium based on its motion in the radial direction and its relative thickening, as both are easily seen from this view. During contraction and relaxation, healthy myocardium undergoes strain and the wall thickness changes accordingly: thickening during contraction and thinning during relaxation. The strength of the left ventricle’s contractile function can be directly reflected on the ultrasound image through the magnitude of the myocardial strain. Abnormalities in the contractile function, such as akinesia, can be indicative of ischaemic heart disease and myocardial infarction 33 .
我们比较了可穿戴设备与超声心动图的四个主要视图中的商业设备的性能，在超声心动图中可以识别关键的心脏特征（扩展数据图3）。图3a显示了这四个视图的示意图和相应的b型图像，包括心尖四腔心视图、心尖两腔心视图、胸骨旁长轴视图和胸骨旁短轴视图。来自可穿戴设备和商用设备的测试结果之间的差异可以忽略不计。胸骨旁短轴视图对于基于心肌的径向运动和相对增厚来评估心肌的收缩功能特别有用，从这个视图中很容易看到。在收缩和舒张过程中，健康心肌发生应变，壁厚发生相应变化：收缩时增厚，舒张时变薄。左心室收缩功能的强度可以通过心肌应变的大小直接反映在超声图像上。收缩功能的异常，如运动障碍，可提示缺血性心脏病和心肌梗死33。
To better localize the specific segment of the left ventricular wall that is potentially pathological, the 17-segment model can be adopted as in standard clinical practice 33 (Fig. 3b). We took the basal, mid-cavity and apical slices of the parasternal short-axis view from the left ventricular wall, and divided them into segments according to the model. Each segment is linked to a certain coronary artery, allowing ischaemia in the coronary arteries to be localized on the basis of akinesia in the corresponding myocardial segment 33 . We then recorded the displacement waveforms of the myocardium boundaries (Fig. 3c and Supplementary Discussion 6). The two peaks in each cardiac cycle in the displacement curves correspond to the two inflows into the left ventricle during diastole. The wall displacements, as measured in the basal, mid-cavity and apical views, become sequentially smaller owing to the decreasing radius of the myocardium along the conical shape of the left ventricle.
为了更好地定位可能存在病理变化的左心室壁的特定节段，可以采用标准临床实践中的17节段模型33（图3b）。我们从左心室壁取胸骨旁短轴的基底、中腔和根尖切片，并根据模型将其分成节段。每个节段都与某一冠状动脉相连，允许冠状动脉的缺血基于相应的心肌节段的运动障碍进行定位33。然后，我们记录了心肌边界的位移波形（图3c和补充讨论6）。位移曲线中每个心脏周期的两个峰值对应于舒张期流入左心室的两个峰值。在基底、中腔和根尖测量的壁位移，由于心肌沿左心室锥形半径减小，壁位移逐渐减小。
Motion-mode (M-mode) images track activities over time in a one-dimensional target region 34,35 . We extracted M-mode images from parasternal long-axis view B-mode images (Fig. 3d). Primary targets include the left ventricular chamber, the septum and the mitral/aortic valves. In M-mode, structural information, such as the myocardial thickness and the left ventricular diameter, can be tracked according to the distances between the boundaries of features. Valvular functions, for example, their opening and closing velocities, can be evaluated on the basis of the distance between the leaflet and septal wall (Supplementary Discussion 1). Moreover, we can correlate the mechanical activities in the M-mode images with the electrical activities in the electrocardiogram measured simultaneously during different phases in a cardiac cycle (Fig. 3d and Supplementary Discussions 1 and 6).
运动模式（M-模式）图像随时间的活动跟踪一维目标区域34,35。我们从胸骨旁长轴视图B-模式图像中提取M-模式图像（图3d）。主要目标包括左心室、室间隔和二尖瓣/主动脉瓣。在M-型模式下，可以根据特征边界之间的距离来跟踪心肌厚度和左心室直径等结构信息。瓣膜的功能，例如，它们的开启和关闭速度，可以根据小叶和间隔壁之间的距离来评估（补充讨论1）。此外，我们还可以将M-型图像中的机械活动与心脏周期不同阶段同时测量的心电图电活动联系起来（图3d和补充讨论1和6）。
Monitoring during motion
Stress echocardiography assesses cardiac responses to stress induced by exercise or pharmacologic agents, which may include new or worsened ischaemia presenting as wall-motion abnormalities, and is crucial in the diagnosis of coronary artery diseases 36 . Furthermore, subjects with heart failure may sometimes seem asymptomatic at rest, as the heart sacrifices its efficiency to maintain the same cardiac output 37,38 . Thus, by pushing the heart towards its limits during exercise, the lack of efficiency becomes apparent. However, in current procedures, ultra-sound images are obtained only before and after exercise. With the cumbersome apparatus, it is impossible to acquire data during exercise, which may contain invaluable real-time insights when new abnormalities initiate 39 (Supplementary Discussion 7). Also, because images are traditionally obtained after exercise, a quick recovery can mask any transient pathologic response during stress and lead to false-negative examinations 40 . In addition, the end point for terminating the exercise is subjective, which may result in suboptimal testing.
运动期间的监测
应激超声心动图评估了由运动或药物引起的应激心脏反应，其中可能包括新的或恶化的缺血，表现为壁运动异常，在冠状动脉疾病的诊断中至关重要36。此外，心力衰竭的受试者在休息时有时似乎无症状，因为心脏牺牲了其效率来维持相同的心输出量37,38。因此，通过运动将心脏推向极限，效率的缺乏就变得明显。然而，在目前的程序中，超声图像只能在运动前后获得。由于繁琐的设备，不可能在运动过程中获得数据，当新的异常开始时，可能会包含宝贵的实时见解39（补充讨论7）。而且，因为图像传统上是在运动后获得的，快速恢复可以掩盖压力期间任何短暂的病理反应，并导致假阴性检查40。此外，终止测试的终点是主观的，这可能会导致次优测试。
The wearable ultrasonic patch is ideal for overcoming these challenges. The device can be attached to the chest with minimal constraint to the movement of the subject, providing a continuous recording of cardiac activities before, during and after exercise with negligible motion artefacts (Extended Data Fig. 4). This not only captures the real-time responses during the test but also offers objective data to standardize the end point and enhances patient safety throughout the test (Supplementary Discussion 7). We used liquidus silicone as the couplant to achieve images of stable quality instead of water-based ultrasound gels that evaporate over time (Supplementary Figs. 20 and 21 and Supplementary Discussion 8). We observed no skin irritation or allergy after 24 h of continuous wear (Supplementary Fig. 22). The heart rate of the subject remained stable with a constant device temperature of about 32 °C after the device continuously worked for 1 h (Supplementary Fig. 23). In addition, one device was tested on different subjects (Supplementary Fig. 24). The reproducible results indicate the stable and reliable performance of the wearable imager.
可穿戴的超声波贴片是克服这些挑战的理想选择。该装置可以附着在胸部，对受试者运动的限制最小，可以连续记录运动前、运动期间和运动之后的心脏活动，而运动伪影可以忽略不计（扩展数据图4）。这不仅捕获了测试期间的实时响应，而且还提供了客观的数据来标准化终点，并提高了整个测试过程中患者的安全性（补充讨论7）。我们使用液相线硅胶作为偶联体，以获得稳定质量的图像，而不是随着时间蒸发的水基超声凝胶(补充图20和21和补充讨论8)。我们没有观察到皮肤刺激或持续磨损24小时后的过敏（补充图22）。在设备连续工作1小时后，受试者的心率保持稳定，设备温度在32 ℃左右，保持不变（补充图23）。此外，有一个设备在不同的受试者上进行了测试（补充图24）。可重复性结果表明，该可穿戴成像仪的性能稳定、可靠。
We performed stress echocardiography to demonstrate the performance of the device during exercise (Supplementary Discussion7). The device was attached to the subject for continuous recording along the parasternal long axis during the entire process, which consisted of three main stages (Fig. 4a). In the rest stage, the subject laid in the supine position. In the exercise stage, the subject exercised on a stationary bike with several intervals until a possible maximal heart rate was reached. In the recovery stage, the subject was placed in the supine position again. The device demonstrated uninterrupted tracking of the left ventricular activities, including the corresponding M-mode echocardiography and synchronized heart-rate waveform (Fig. 4b,c, Extended Data Fig. 5 and Supplementary Video 3). We examined a representative section of each testing stage and extracted the left ventricular internal diameter end systole (LVIDs) and left ventricular internal diameter end diastole (LVIDd) (Fig. 4d). The LVIDs and LVIDd of the subject remained stable during the rest stage (Fig. 4e). In the exercise stage, the interventricular septum and left ventricular posterior wall of the subject moved closer to the skin surface, with the latter moving more than the former, resulting in a decrease in LVIDs and LVIDd. In the recovery stage, the LVIDs and LVIDd slowly returned to normal. The variation in fractional shortening, a measure of the cardiac muscular contractility, reflects the changing demand for blood supply in different stages of stress echocardiography (Fig. 4e). Particularly, section 4 in Fig. 4b includes periods of exercise and intervals for rest, when patterns of a deep breath can also be seen from the left ventricular posterior wall motions (Fig. 4f).
我们进行了应力超声心动图来演示该设备在运动过程中的性能（补充讨论7）。该装置被连接在受试者身上，在整个过程中沿着胸骨旁长轴进行连续记录，其中包括三个主要阶段（图4a)。在休息阶段，受试者位为仰卧位。在运动阶段，受试者在固定自行车上间隔几次运动，直到达到可能的最大心率。在恢复阶段，受试者再次处于仰卧位。该装置显示了对左心室活动的不间断跟踪，包括相应的M-型超声心动图和同步心率波形（图4b、c、扩展数据图5和补充视频3）。我们检查了每个测试阶段的代表性切片，并提取了左室内径收缩期（LVIDs）和左室内径舒张期（LVIDd）（图4d）。受试者的LVIDs和LVIDd在休息阶段保持稳定（图4e）。在运动阶段，受试者的室间隔和左心室后壁更靠近皮肤表面，后者移动得大于前者，导致LVIDs和LVIDd的减少。在恢复阶段，LVIDs和LVIDd缓慢恢复正常。缩短分数是衡量心肌收缩力的变化，反映了应力超声心动图不同阶段对血液供需求的变化（图4e）。特别是，图4b中的第4节包括运动时间和休息时间，从左心室后壁运动也可以看到深呼吸的模式（图4f）。
Automatic image processing
Cardiovascular diseases are often associated with changes in the pumping capabilities of the heart, which could be measured by stroke volume, cardiac output and ejection fraction. Non-invasive, continuous monitoring of these indices are valuable for the early detection and surveillance of cardiovascular conditions (Supplementary Discussion 9). Critical information embodied in these waveforms may help precisely determine potential risk factors and track the health state 41 (Supplementary Discussion 10). On the other hand, processing of the unprecedented image data streams, if done manually, can be overwhelming for clinicians, which potentially introduces inter-observer variability or even errors 42 .
自动图像处理
心血管疾病通常与心脏泵血能力的变化有关，这种能力可以通过每搏出量、心输出量和射血分数来测量。对这些指标进行无创、持续的监测对于心血管疾病的早期发现和监测很有价值（补充讨论9）。这些波形中所包含的关键信息可能有助于精确地确定潜在的风险因素，并跟踪健康状态41（补充讨论10）。另一方面，如果手动处理前所未有的图像数据流，对临床医生来说可能是压倒性的，这可能会导致观察者间的可变性甚至错误42。
Automatic image processing can overcome the challenges. We applied a deep learning neural network to extract key information (for example, the left ventricular volume in apical four-chamber view) from the continuous stream of images (Fig. 5a, Supplementary Fig. 25 and Supplementary Discussion 11). We evaluated different types of deep learning models 43 through the output images and waveforms of the left ventricular volume (Extended Data Figs. 6 and 7, Supplementary Table 3 and Supplementary Video 4). The FCN-32 model outperforms others based on qualitative and quantitative analyses (Supplementary Fig. 26, Supplementary Tables 4 and 5 and Supplementary Discussion 11). We also applied data augmentation to expand the dataset and improve the performance (Supplementary Fig. 27 and Supplementary Discussion 11).
自动图像处理可以克服这些挑战。我们应用深度学习神经网络从连续的图像流中提取关键信息（例如，根尖四室视图）（图5a，补充图25和补充讨论11）中的左心室容积)。我们通过输出图像和左心室容积的波形，评估了不同类型的深度学习模型43(扩展数据图。6和7，补充表3和补充视频4)。基于定性和定量分析的FCN-32模型优于其他模型（补充图26、补充表4和5以及补充讨论11）。我们还应用了数据增强技术来扩展数据集，并提高了性能（补充图27和补充讨论11）。
The output left ventricular volumes for the wearable and commercial imagers show similar waveform morphologies (Fig. 5b, left). From the waveforms, corresponding phases of a cardiac cycle can be identified (Fig. 5b, right and Extended Data Fig. 8). Bland–Altman analysis gives a quantitative comparison between the model-generated and manually labelled left ventricular volumes, indicating a stable and reliable performance of the FCN-32 model 44 (Fig. 5c and Supplementary Discussion 11). The mean differences in the left ventricular volume are both approximately −1.5 ml, which is acceptable for standard medical diagnosis 45 . We then derived stroke volume, cardiac output and ejection fraction from the left ventricular volume waveforms. No marked difference is observed in the averages or standard deviations between the two devices (Fig. 5d). The results verified the comparable performance of the wearable imager to the commercial imager.
可穿戴式设备和商用成像仪的输出左心室容积显示出相似的波形形态（图5b，左）。从波形中，可以识别出心脏周期的相应阶段（图5b，右图和扩展数据图8）。Bland-Altman分析给出了一个定量比较模型生成的和人工标记的左心室容积，表明FCN-32模型的性能稳定和可靠（图5c和补充讨论11）44。左心室容积的平均差异均约为−1.5ml，这是标准医学诊断中可接受的45。然后，我们从左心室容积波形中推导出每搏量、心输出量和射血分数。两种器件之间的平均值或标准差均无明显差异（图5d）。结果验证了可穿戴成像仪与商用成像仪相当的性能。
The left ventricular volume is constantly changing and generally follows a steady-state pattern at rest in healthy subjects. Therefore, stroke volume, cardiac output and ejection fraction also tend to remain constant. However, cardiac pathologies or ordinary daily activities such as exercise may dynamically change those indices. To validate the performance of the wearable imager under dynamic situations, we extracted the left ventricular volume from recordings in the recovery stage of stress echocardiography (Fig. 5e). The dimensions of the left ventricle cannot be accurately determined when the images are collected in the standing position, owing to anatomical limitations of the human body (Supplementary Fig. 28 and Supplementary Discussion 9). Owing to the deep breathing after exercise, the heart was sometimes blocked by the lungs in the image. We used an image-imputation algorithm to complement the blocked part (Supplementary Fig. 29 and Supplementary Discussion 11). The acquired waveform shows an increasing trend in the left ventricular volume. Figure 5f illustrates three representative sections of the recording taken from the beginning, middle and end of the recovery stage. In the initial section, the diastasis stage is barely noticeable because of the high heart rate. In the middle section, the diastasis stage becomes visible. In the end section, the heart rate decreases notably. The end-diastolic and end-systolic volumes are increasing, because the slowing heartbeat during recovery allows more time for blood to fill the left ventricle 46 (Fig. 5g). The stroke volume gradually increases, indicating that the end-diastolic volume increases slightly faster than the end-systolic volume (Fig. 5g). The ejection fraction decreases as heart contraction decreases during the recovery (Fig. 5g). The cardiac output reduces, indicating a larger influence brought about by the decreasing heart rate than the increasing stroke volume (Supplementary Discussion 9).
健康受试者的左心室容积不断变化，在休息时一般遵循稳定状态模式。因此，每搏量、心输出量和射血分数也倾向于保持不变。然而，心脏病理状态或日常活动，如运动，可能会动态地改变这些指标。为了验证可穿戴成像仪在动态情况下的性能，我们从应力超声心动图恢复阶段的记录中提取了左心室容积（图5e）。由于人体解剖学的限制，在站立位置采集图像时无法准确确定左心室的尺寸（补充图28和补充讨论9）。由于运动后的深呼吸，图像中心脏有时被肺部阻塞。我们使用了一种图像计算算法来补充阻塞部分（补充图29和补充讨论11）。获得性波形的左心室容积呈增加趋势。图5f显示了从恢复阶段的开始、中期和结束开始的三个有代表性的记录部分。在初期，由于心率高，分流期几乎不明显。在中间部分，可以看到分流期。在最后的部分中，心率明显下降。舒张末期和收缩期末期容积正在增加，因为恢复期间心跳减慢使血液有更多的时间填充左心室46（图5g）。冲程容积逐渐增加，说明舒张末期容积的增长速度略快于收缩期末期容积（图5g)。在恢复过程中，射血分数随着心脏收缩的减少而降低（图5g）。心排血量减少，表明心率下降的影响大于增加的每搏量（补充讨论9）。
Discussion
Echocardiography is crucial in the diagnosis of cardiac diseases, but the current implementation in clinics is cumbersome and limits its application in continuous monitoring. Emerging technologies based on wearable rigid modules 25 or flexible patches 47 lack one or more of the ideal properties of wearable ultrasound technologies (Extended Data Table 2). In this work, we provided uninterrupted frame-by-frame acquisitions of cardiac images even when the subject was undertaking intensive exercise. In addition, the wearable imager with deep learning gave actionable information by automatically and continuously out-putting curves of critical cardiac metrics, such as myocardial displacement, stroke volume, ejection fraction and cardiac output, which are highly desirable in critical care, cardiovascular disease management and sports science 48 . This capability is unprecedented in conventional clinical practice 9 and the non-invasiveness can extend potential benefits to the outpatient and athletic populations.
讨论
超声心动图在心脏病的诊断中至关重要，但目前在临床的实施很繁琐，限制了其在持续监测中的应用。基于可穿戴式刚性模块25或柔性贴片47的新兴技术缺乏可穿戴式超声波技术的一种或多种理想特性（扩展数据表2）。在这项工作中，即使受试者正在进行高强度运动，我们也能提供不间断的逐帧获取心脏图像。此外，具有深度学习的可穿戴成像仪通过自动和连续地输出关键心脏指标的曲线，如心肌位移、每搏量、射血分数和心输出量，提供可操作的信息，这在重症照护、心血管疾病管理和运动科学中是非常可取的48。这种能力在传统的临床实践中是前所未有的，非侵入性可以扩展到门诊和运动人群。
The implications of this technology go far beyond imaging the heart, as it can be generalized to image other deep tissues, such as the inferior vena cava, abdominal aorta, spine and liver (Supplementary Fig. 30).For example, as demonstrated in an ultrasound-guided biopsy procedure on a cyst phantom (Supplementary Fig. 31), the two orthogonal imaging sections present the entire biopsy process simultaneously, freeing up one hand of the operator (Supplementary Video 5). The uniquely enabling capability of this technology forgoes the need for an operator to constantly hold the device.
这项技术的意义远远超出了对心脏的成像，因为它可以推广到对其他深层组织的成像，如下腔静脉、腹主动脉、脊柱和肝脏（补充图30）。例如，如超声引导下的囊肿幻像活检程序所示（补充图31），两个正交的成像切片同时显示了整个活检过程，解放了操作者的一只手（补充视频5）。这项技术的独特启用能力缓解了操作员不断持有设备的需要。
Other future efforts could ensue by further improving spatial resolutions (Supplementary Fig.32). A three-dimensional scanner can only provide the curvature of a static human chest. To accommodate the dynamic chest curvature, advanced imaging algorithms need to be developed to compensate for the phase distortion and thus improve spatial resolutions. In addition, the wearable imager is connected to the back-end system for data processing by means of a flexible cable (Supplementary Fig. 33) and future work needs to focus on system miniaturization and integration. Besides, the FCN-32 neural network can only be applied to subjects in the training dataset at present. Its generalizability could potentially be improved by expanding the training dataset or optimizing the network with few-shot-learning 49 or reinforcement-learning 50 strategies, which will allow the model to adapt to a larger population
其他未来的努力也可以通过进一步提高空间分辨率来实现（补充图32）。一个三维扫描仪只能提供一个静态的人体胸部的曲率。为了适应动态胸部曲率，需要开发先进的成像算法来补偿相位畸真，从而提高空间分辨率。此外，可穿戴成像仪通过柔性电缆连接到后端系统进行数据处理（补充图33），未来的工作需要集中在系统的小型化和集成上。此外，FCN-32神经网络目前只能应用于训练数据集中的主题。它的普遍性可以通过扩展训练数据集或使用少镜头学习或强化学习策略优化网络来提高，这将允许模型适应更大的人群。
Methods
方法学
Materials
Gallium–indium eutectic liquid metal, toluene, ethyl alcohol, acetone and isopropyl alcohol were purchased from Sigma-Aldrich. SEBS (G1645) was obtained from Kraton. Silicone (Ecoflex 00-30) was bought from Smooth-On as the encapsulation material of the device. Silicone (Silbione) was obtained from Elkem Silicones as the ultrasound couplant. Aquasonic ultrasound transmission gel was bought from Parker Laboratories. 1-3 composite (PZT-5H) was purchased from Del Piezo Specialties. Silver epoxy (Von Roll 3022 E-Solder) was obtained from EIS. Anisotropic conductive film cable was purchased from Elform.
材料
镓-铟共晶液态金属、甲苯、乙醇、丙酮和异丙醇均购自西格玛-奥尔德里奇股份有限公司。SEBS（G1645）来自Kraton。硅胶（Ecoflex 00-30）来自Smooth-On，作为该设备的封装材料。硅酮（硅联酮）以Elkem硅酮作为超声偶合体。水波超声传输凝胶购自帕克实验室。1-3复合材料（PZT-5H）购自Del压电专业公司。环氧银（Von Roll 3022 E-Solder）从EIS获得。各向异性导电薄膜电缆购自Elform。
Design and fabrication of the wearable imager
We designed the transducer array in an orthogonal geometry, similar to a Mills cross array (Supplementary Fig. 34), to achieve biplane standard views simultaneously. For the transducers, we chose the 1-3 composite for transmitting and receiving ultrasound waves because it possesses superior electromechanical coupling 18 . In addition, the acoustic impedance of 1-3 composites is close to that of the skin, maximizing the acoustic energy propagating into human tissues 19 . The backing layer dampens the ringing effect, broadens the bandwidth and thus improves the spatial resolution 18,51 .
可穿戴成像仪的设计与制造
我们设计了一个正交几何形状的传感器阵列，类似于磨坊交叉阵列（补充图34），以同时实现双平面标准视图。对于换能器，我们选择了1-3复合材料来发射和接收超声波，因为它具有优越的机电耦合18。此外，1-3复合材料的声阻抗接近于皮肤，最大限度地提高了传播到人体组织中的声能19。背衬层抑制了振铃效应，拓宽了带宽，从而提高了空间分辨率18,51。
We used an automatic alignment strategy to fabricate the orthogonal array. The existing method of bonding the backing layer to the 1-3 composite was to first dice many small pieces of backing layer and 1-3 composite, and then bond each pair together one by one. A template was needed to align the small pieces. This method was of very low efficiency. In this study, we bond a large piece of backing layer with a large piece of 1-3 composite and then dice them together into small pieces with designed configurations. The diced array is then automatically aligned on adhesive tape with high uniformity and perfect alignment.
我们使用了一种自动对齐策略来制造正交阵列。现有的将衬层与1-3复合材料结合的方法是先将许多小块衬层和1-3复合材料切碎，然后一对一对粘在一起。需要一个模板来对齐这些小块。这种方法的效率很低。在这项研究中，我们将一大块的背衬层与一大块的1-3复合材料结合起来，然后将它们切成小块。然后，切丁的阵列自动对齐，以高均匀性和完美对齐。
Electrodes based on eutectic gallium–indium liquid metal are fabricated to achieve better stretchability and higher fabrication resolution than existing electrodes based on serpentine-shaped copper thin film. Eutectic gallium–indium alloys are typically patterned through approaches such as stencil lithography 52 , masked deposition 53 , inkjet printing 54 , microcontact printing 55 or microfluidic channelling 56 . Although these approaches are reliable, they are either limited in patterning resolution or require sophisticated photolithography or printing hardware. The sophisticated hardware makes fabrication complicated and time-consuming, which presents a challenge in the development of compact, skin-conformal wearable electronics.
与现有的基于蛇形铜薄膜的电极相比，基于共晶镓-铟液态金属的电极具有更好的拉伸性和更高的制备分辨率。共晶镓-铟合金通常通过模板光刻52、掩模沉积53、喷墨印刷54、微接触印刷55或微流控通道56等方法进行图案。虽然这些方法是可靠的，但它们要么在图案分辨率上有限，要么需要复杂的光刻或打印硬件。复杂的硬件使制造变得复杂和耗时，这给紧凑的、皮肤适形的可穿戴电子产品的开发带来了挑战。
In this study, we exploited a new technology for patterning. We first screen-printed a thin layer of liquid metal on a substrate. A key consideration before screen printing was how to get the liquid metal to wet the substrate. To solve this problem, we dispersed big liquid metal particles into small microparticles using a tip sonicator (Supplementary Fig. 2). When microparticles contacted air, their outermost layer generated an oxide coating, which lowered the surface tension and prevented those microparticles from aggregating. In addition, we used 1.5 wt.% SEBS as a polymer matrix to disperse the liquid metal particles because SEBS could wet well on the liquid metal surface. We also used SEBS as the substrate. Therefore, the SEBS in the matrix and the substrate could merge and cure together after screen printing, allowing the liquid metal layer to adhere to the substrate efficiently and uniformly. Then we used laser ablation to selectively remove the liquid metal from the substrate to form patterned electrodes.
在这项研究中，我们利用了一种新的模式形成技术。我们首先在基板上用丝网印刷了一层薄薄的液体金属。在丝网印刷之前的一个关键考虑因素是如何让液态金属弄湿基板。为了解决这个问题，我们使用针尖超声器将大的液体金属颗粒分散成小的微粒（补充图2）。当微粒与空气接触时，它们的最外层产生了一层氧化物涂层，这降低了表面张力，防止了这些微粒的聚集。此外，我们使用1.5 wt.%的SEBS作为聚合物基质来分散液态金属颗粒，因为SEBS可以很好地润湿液态金属表面。我们还使用了SEBS作为底物。因此，在丝网印刷后，基质和基板中的SEBS可以合并并固化，使液态金属层能够有效而均匀地粘附在基板上。然后，我们使用激光烧蚀，选择性地从基底中去除液态金属，形成图案电极。
The large number of piezoelectric transducer elements in the array requires many such electrodes to address each element individually. We designed a four-layered top electrode and a common ground electrode. There are SEBS layers between different layers of liquid metal electrodes as insulation. To expose all electrode layers to connect to transducer elements, we used laser ablation to drill vertical interconnect accesses 21 . Furthermore, we created a stretchable shielding layer using liquid metal and grounded it through a vertical interconnect access, which effectively protected the device from external electromagnetic noises (Supplementary Fig. 8).
阵列中大量的压电换能器元件需要许多这样的电极来单独处理每个元件。我们设计了一个四层的顶部电极和一个公共接地电极。在不同层的液体金属电极之间有SEBS层作为绝缘层。为了暴露所有的电极层以连接到传感器元件，我们使用激光烧蚀来钻取垂直的互连通道21。此外，我们利用液态金属创建了一个可伸缩的屏蔽层，并通过垂直互连接入将其接地，有效地保护了设备免受外部电磁噪声的影响（补充图8）。
Before we attached the electrodes to the transducer array, we spin-coated toluene–ethanol solution (volume ratio 8:2) on the top of the multilayered electrode to soften the liquid-metal-based elastomer, also known as ‘solvent-welding’. The softened SEBS provided a sufficient contact surface, which could help form a relatively strong van der Waals force between the electrodes and the metal on the transducer surface. After bonding the electrodes to the transducer array, we left the device at room temperature to let the solvent evaporate. The final bonding strength of more than 200 kPa is stronger than many commercial adhesives 22 .
在我们将电极连接到换能器阵列之前，我们在多层电极的顶部旋转涂层甲苯-乙醇溶液（体积比8：2），以软化液体金属基弹性体，也被称为“溶剂焊接”。软化的SEBS提供了一个足够的接触面，这有助于在电极和传感器表面上的金属之间形成一个相对较强的范德华力。在将电极连接到传感器阵列后，我们将设备放在室温下，让溶剂蒸发。这个超过200 kPa的最终粘合强度比许多商业粘合剂都要强22。
To encapsulate the device, we irrigated the device in a petri dish with uncured silicone elastomer (Ecoflex 00-30, Smooth-On) to fill the gap between the top and bottom electrodes and the kerf among the transducer elements. We then cured the silicone elastomer in an oven for 10 min at 80 °C. As the filling material, it suppresses spurious shear waves from adjacent elements, effectively isolating crosstalk between the elements 18,19 . With that being said, we think the main reason for the suppressed spurious shear waves is because of the epoxy in the 1-3 composite, which limits the lateral vibration of the piezoelectric materials. The Ecoflex as the filling material may have contributed but not played a chief role because the kerf is not too wide, only 100 to 200 μm. We lifted off the glass slide on the top electrode and directly covered the top electrode with a shielding layer. Then we lifted off the glass slide on the bottom electrode to release the entire device. Finally, screen-printing an approximately 50-μm layer of silicone adhesive on the device surface completed the entire fabrication.
为了封装该设备，我们在一个使用未固化的硅酮弹性体（Ecoflex 00-30，Smooth-On）的培养皿中冲洗该设备，以填补顶部和底部电极以及传感器元件之间的间隙。然后，我们将硅弹性体在80℃下固化10 min。作为填充材料，它抑制来自相邻元件的杂波，有效地隔离元件18、19之间的串扰。综上所述，我们认为杂散剪波被抑制的主要原因是1-3复合材料中的环氧树脂，它限制了压电材料的横向振动。Ecoflex作为填充材料可能有贡献，但没有发挥主要作用，因为角不是太宽，只有100到200 μm。我们卸下顶部电极上的载玻片，直接用屏蔽层覆盖顶部电极。然后我们松开底部电极上的载玻片，释放整个装置。最后，在设备表面丝网印刷大约50μm的硅胶粘合剂完成了整个制造过程。
Characterization of the liquid metal electrode
Existing wearable ultrasound arrays can achieve excellent stretchability by serpentine-shaped metal thin films as electrodes 19,26 . The serpentine geometry, however, severely limits the filling ratio of functional components, precluding the development of systems that require a high integration density or a small pitch. In this study, we chose to use liquid metal as the electrode owing to its large intrinsic stretchability, which makes the high-density electrode possible. The patterned liquid metal electrode had a minimum width of about 30 μm with a groove of about 24 μm (Supplementary Fig. 3), an order of magnitude finer than other stretchable electrodes 18,26,57 . The liquid metal electrode is ideal for connecting arrays with a small pitch 58 .
液态金属电极的表征
现有的可穿戴超声波阵列可以通过蛇形金属薄膜作为电极19,26来实现优异的拉伸性。然而，蛇形的几何形状严重限制了功能组件的填充比例，排除了需要高集成密度或小间距的系统的发展。在本研究中，我们选择使用液体金属作为电极，因为它具有较大的固有拉伸性，这使得高密度电极成为可能。该图案化的液态金属电极的最小宽度约为30 μm，凹槽约为24μm（补充图3），比其他可拉伸电极18、26、57更细一个数量级。液体金属电极理想地用于连接具有小间距58的阵列。
This liquid metal electrode exhibited high conductivity, exceptional stretchability and negligible resistance change under tensile strain (Fig. 1b and Supplementary Fig. 4). The initial resistance at 0% strain was 1.74 Ω (corresponding to a conductivity of around 11,800 S m −1 ), comparable with reported studies 59,60 . The resistance gradually increased with strain until the electrode reached the approximately 750% failure strain (Fig. 1b and Supplementary Fig. 4). The relative resistance is a parameter widely used to characterize the change in the resistance of a conductor (that is, the liquid metal electrode in this case) under different strains relative to the initial resistance 58–60 . The relative resistance is unitless. When the strain was 0%, the initial resistance R 0 was 1.74 Ω. When the electrode was under 750% strain, the electrode was broken and the resistance R at the breaking point was measured to be 44.87 Ω. Therefore, the relative resistance (R/R 0 ) at the breaking point was 25.79.
这种液态金属电极具有高导电性、特殊的拉伸性、拉伸应变下的电阻变化可以忽略不计（图1b和补充图4）。与报道的研究59,60相比较，在0%应变下的初始抗性为1.74Ω（对应的电导率约为11,800 S m−1）。电阻随应变的增加而逐渐增大，直到电极达到约750%的失效应变（图1b和补充图4）。相对电阻是一个广泛用于表征导体（即在这种情况下的液体金属电极）在相对于初始电阻58-60的不同应变下的电阻变化的参数。相对电阻是无单位的。当应变为0%时，初始抗性R0为1.74 Ω。当电极在750%应变下时，电极断裂，断裂点处的电阻R为44.87 Ω。因此，在断裂点处的相对电阻（R/R 0）为25.79。
To investigate the electrode fatigue, we subjected them to 100% cyclic tensile strain (Fig. 1c). The initial 500 cycles observed a gradual increase in the electrode resistance because the liquid metal, when stretched, could expose more surfaces. These new surfaces were oxidized after contacting with air, leading to the resistance increase (Supplementary Fig. 4). After the initial 500 cycles, the liquid metal electrode exhibited stable resistance because, after a period of cycling, there were not many new surfaces exposed.
为了研究电极的疲劳情况，我们对它们施加了100%的循环拉伸应变（图1c）。最初的500次循环观察到电极电阻逐渐增加，因为液态金属被拉伸时，会暴露更多的表面。这些新的表面在与空气接触后被氧化，导致电阻增加（补充图4）。在最初的500次循环后，液态金属电极表现出稳定的电阻，因为经过一段时间的循环后，没有很多的新表面暴露出来。
This study is the first to use liquidmetal-based electrodes to connect ultrasound transducer elements. The bonding strength between them directly decides the robustness and endurance of the device. This is especially critical for the wearable patch, which will be subjected to repeated deformations during use. Therefore, we characterized the bonding strength of the electrode to the transducer element using a lap shear test. The liquid metal electrode was first bonded with the transducer element. The other sides of the electrode and the element were both fixed with stiff supporting layers. The supporting layer serves to be clamped by the tensile grips of the testing machine. Samples will be damaged if they are clamped by the grips directly. Then a uniaxial stretching was applied to the sample at a strain rate of 0.5 s −1 . The test was stopped when the electrode was delaminated from the transducer element. A SEBS film was bonded with a transducer element and we performed the lap shear test using the same method. The peak values of the curve were used to represent the lap shear strength (Fig. 1d). The bonding strength between the pure SEBS film and the transducer element was roughly 250 kPa, and that between the electrode and the transducer element was about 236 kPa, which were both stronger than many commercial adhesives (Supplementary Table 2). The results indicate the robust bonding between the electrode and the element, preventing the electrodes from delamination under various deformations. This robust bonding does not have any limitations on the ultrasound pressures that can be transduced.
本研究首次使用液体金属基电极连接超声换能器元件。它们之间的粘合强度直接决定了器件的坚固性和耐久性。这对于可穿戴贴片尤其重要，它在使用过程中会反复变形。因此，我们用搭接剪切试验来表征了电极与换能器元件的粘合强度。液体金属电极首先与传感器元件键合。电极和元件的另一边都用坚硬的支撑层固定。支撑层应由试验机的抗拉夹具夹紧。如果直接夹住，将会损坏。然后在0.5 s−1的应变速率下对样品进行单轴拉伸。当电极与传感器元件分层时，测试停止。将SEBS薄膜与传感器元件粘合，我们使用同样的方法进行了搭接剪切试验。用曲线峰值表示搭接剪切强度（图1d)。纯SEBS膜与换能器元件之间的粘合强度约为250 kPa，电极与换能器元件之间的粘合强度约为236 kPa，均比许多商业粘合剂都强（补充表2）。结果表明，电极与元件之间的结合牢固，防止了电极在各种变形下的分层。这种强大的键合对可传导的超声波压力没有任何限制。
Characterization of the transducer elements
The electromechanical coupling coefficient of the transducer elements was calculated to be 0.67, on par with that of commercial probes (0.58–0.69) 61 . This superior performance was largely owing to the technique for bonding transducer elements and electrodes at room temperature in this study, which protected the piezoelectric material from heat-induced damage and depolarization. The phase angle was >60°, substantially larger than most earlier studies 18,62 , indicating that most of the dipoles in the element aligned well after bonding 63 . The large phase angle also demonstrated the exceptional electromechanical coupling performance of the device. Dielectric loss is critical for evaluating the bonding process because it represents the amount of energy consumed by the transducer element at the bonding interface 20 . The average dielectric loss of the array was 0.026, on par with that of the reported rigid ultrasound probes (0.02–0.04) 64–66 , indicating negligible energy consumed by this bonding approach (Supplementary Fig. 1b). The response echo was characterized in time and frequency domains (Supplementary Fig. 1c), from which the approximately 35 dB signal-to-noise ratio and roughly 55% band-width were derived. The crosstalk values between a pair of adjacent elements and a pair of second nearest neighbours have been characterized (Supplementary Fig. 1d). The average crosstalk was below the standard −30 dB in the field, indicating low mutual interference between elements.
传感器元件的表征
换能器元件的机电耦合系数为0.67，与商业探针（0.58–0.69）61相当。这种优越的性能在很大程度上是由于本研究中在室温下键合传感器元件和电极的技术，它保护了压电材料免受热致损伤和去极化。相位角为>60°，大大大于大多数早期的研究18,62，表明元素中的大多数偶极子在键合后排列得很好。较大的相位角也证明了该器件优异的机电耦合性能。介电损耗对于评估键合过程至关重要，因为它代表了换能器元件在键合界面20处所消耗的能量量。该阵列的平均介电损耗为0.026，与所报道的刚性超声探针（0.02–0.04）64-66相当，表明这种键合方法所消耗的能量可以忽略不计（补充图1b）。响应回波在时域和频域进行表征（补充图1c），由此得到约35 dB的信噪比和约55%的带宽。对一对相邻元素和一对第二最近邻之间的串扰值进行了表征（补充图1d）。现场的平均串扰低于标准−30dB，表明元件间的相互干扰较低。
Characterization of the wearable imager
We characterized the wearable imager using a commercial multipurpose phantom with many reflectors of different forms, layouts and acoustic impedances at various locations (CIRS ATS 539, CIRS Inc.) (Supplementary Fig. 11). The collected data are presented in Extended Data Table 1. For most of the tests, the device was first attached to the phantom surface and rotated to ensure the best imaging plane. Raw image data were saved to guarantee minimum information loss caused by the double-to-int8 conversion. Then the raw image data were processed using the ‘scanConversion’ function provided in the k-Wave toolbox to restore the sector-shaped imaging window (restored data). We applied five times upsampling in both vertical and lateral directions. The upsampled data were finally converted to the dB unit using:
可穿戴成像仪的表征
我们使用一种商业多用途幻像对可穿戴成像仪进行了表征，它具有不同位置的许多反射形式、布局和声阻抗（CIRS ATS 539，CIRS Inc.）（补充图11）。收集到的数据见扩展数据表1。在大多数测试中，该设备首先附着在幻像表面，并旋转以确保最佳的成像平面。保存原始图像数据，以保证双-int8转换造成的信息损失最小。然后使用k-Wave工具箱中提供的“扫描转换”函数对原始图像数据进行处理，以恢复扇形成像窗口（恢复的数据）。我们在垂直方向和横向方向上都进行了5次上采样。上采样的数据最终用以下方法转换为dB单位：
The penetration depth was tested with a group of lines of higher acoustic impedance than the surrounding background distributed at different depths in the phantom. The penetration depth is defined as the depth of the deepest line that is differentiable from the background (6 dB higher in pixel value). Because the deepest line available in this study was at a depth of 16 cm and was still recognizable from the background, the penetration depth was determined as >16 cm.
用一组比分布在幻像中不同深度的周围背景更高的声阻抗线来测试穿透深度。穿透深度定义为可与背景区分的最深线的深度（像素值高出6 dB）。由于本研究中可用的最深线深度为16 cm，从背景中仍然可以识别，因此穿透深度为>16 cm。
The accuracy is defined as the precision of the measured distance. The accuracy was tested with the vertical and lateral groups of line phantoms. The physical distance between the two nearest pixels in the vertical and lateral directions was calculated as:
精度定义为测量距离的精度。用垂直组和横向组的线幻像来测试准确性。在垂直和横向方向上最近的两个像素之间的物理距离计算为：
We acquired the measured distance between two lines (shown as two bright spots in the image) by counting the number of pixels between the two spots and multiplying them by Δy or Δx, depending on the measurement direction. The measured distances at different depths were compared with the ground truth described in the data sheet. Then the accuracy can be calculated by:
我们通过计算两条线之间的像素数，并根据测量方向乘以Δy或Δx，获得了两条线之间的测量距离（以图像中的两个亮点表示）。将不同深度的测量距离与数据表中描述的地面真实值进行了比较。则可通过以下方法计算出精度：
The lateral accuracy was presented as the mean accuracy of the four neighbouring pairs of lateral lines at a depth of 50 mm in the phantom.
横向精度表示为在幻像中深度为50 mm处的四对相邻的横向线的平均精度。
The spatial resolutions were tested using the lateral and vertical groups of wires. For the resolutions at different depths, the full width at half maximum of the point spread function in the vertical or lateral directions for each wire was calculated. The vertical and lateral resolutions could then be derived by multiplying the number of pixels within the full width at half maximum by Δy or Δx, depending on the measurement direction. The elevational resolutions were tested by rotating the imager to form a 45° angle between the imager aperture and the lines. Then the bright spot in the B-mode images would reveal scatters out of the imaging plane. The same process as calculating the lateral resolutions was applied to obtain the elevational resolutions. The spatial resolutions at different imaging areas were also characterized with the lateral group of wires. Nine wires were located at ±4 cm, ±3 cm, ±2 cm, ±1 cm and 0 cm from the centre. The lateral and axial resolutions of the B-mode images from those wires were calculated with the same method.
使用横向和垂直的电线组来测试空间分辨率。对于不同深度的分辨率，计算了每根导线在垂直或横向方向上的点扩散函数的半最大值处的全宽。根据测量方向，可以通过将半最大值内的像素数乘以Δy或Δx，可以得到垂直和横向分辨率。通过旋转成像仪在成像仪孔径和线之间形成45°角的角度来测试高程分辨率。然后，B-模式图像中的亮点就会显示出成像平面外的散点。采用与计算横向分辨率相同的过程来获得高程分辨率。不同成像区域的空间分辨率也以侧组为特征。9根导线位于距中心±4 cm，±3 cm，±2 cm，±1 cm和0 cm处。用相同的方法计算了这些导线的B-模式图像的横向和轴向分辨率。
Note that the lateral resolution worsens with the depth, mainly because of the receive beamforming (Supplementary Fig. 15). There are two beamformed signals, A and B. The lateral resolution of the A point (x 1 ) is obviously better than that of the B point (x 2 ). The fact that lateral resolution becomes worse with depth is inevitable in all ultrasound imaging, as long as receive beamforming is used.
需要注意的是，横向分辨率随着深度的增加而恶化，这主要是因为接收波束的形成（补充图15）。有两个波束形成的信号，A和B。A点（x 1）的横向分辨率明显优于B点（x 2）。在所有超声成像中，只要使用接收波束形成，横向分辨率随深度而下降是不可避免的。
As for different transmit beamforming methods, the wide-beam compounding is the best because it can achieve a synthetic focusing effect in the entire insonation area. The better the focusing effect, the higher the lateral resolution, which is why the lateral resolution of the wide-beam compounding is better than the other two transmit methods at the same depth. Furthermore, the multiple-angle scan used in the wide-beam compounding can enhance the resolution at high-angle areas. The multiple-angle scan combines transmissions at different angles to achieve a global high signal-to-noise ratio, resulting in improved resolutions.
对于不同的发射波束形成方法，宽光束复合是最好的，因为它可以在整个共振区域实现合成聚焦效果。聚焦效果越好，横向分辨率就越高，这就是为什么在相同的深度下，宽光束复合方法的横向分辨率都优于其他两种发射方法。此外，在广光束复合中使用的多角度扫描可以提高高角度区域的分辨率。多角度扫描结合了不同角度的传输，以实现全球的高信噪比，从而提高了分辨率。
The elevational resolution can only be characterized when the imaging target is directly beneath the transducer. For those targets that are far away from the centre, they are difficult to be imaged, which makes their elevational resolutions challenging to calculate. When characterizing the elevational resolution, the device should rotate 45°. In this case, most of the reflected ultrasound waves from those wires cannot return to the device owing to the large incidence angles. Therefore, those wires cannot be captured in the B-mode images. One potential solution is to decrease the rotating angle of the device, which may help capture more wires distributed laterally in the B-mode image. However, a small rotating angle will cause the elevational image to merge with the lateral image, which increases the error of calculating the elevational resolution. Considering those reasons, we only characterized the elevational resolution of the imaging targets directly beneath the transducer array.
只有当成像目标直接在传感器下方时，才能确定高程分辨率。对于那些远离中心的目标，它们很难被成像，这使得它们的高程分辨率的计算具有挑战性。当确定高程分辨率时，设备应旋转45°。在这种情况下，由于入射角大，这些导线的反射超声波不能返回到设备。因此，这些导线不能在B-模式图像中被捕获。一个可能的解决方案是减少设备的旋转角度，这可能有助于捕获在B-模式图像中横向分布的更多的导线。但是，一个较小的旋转角度会导致高程图像与横向图像合并，从而增加了高程分辨率的计算误差。考虑到这些原因，我们只描述了在传感器阵列直接下方的成像目标的高度分辨率。
The contrast resolution, the minimum contrast that can be differentiated by the imaging system, was tested with greyscale objects. The collected B-mode images are shown in Fig. 2. Because the targets with +3 and −3 dB, the lowest contrast available in this study, could still be recognized in the images, the contrast resolution of the wearable imager is determined as <3 dB.
对比度分辨率，即成像系统可以分辨的最小对比度，用灰度物体进行测试。采集到的B-模式图像如图2所示。由于本研究中对比度最低的+3和−3dB的目标仍然可以在图像中被识别出来，因此可穿戴成像仪的对比度分辨率被确定为<3 dB。
The dynamic range in an ultrasound system refers to the contrast range that can be displayed on the monitor. The contrast between an object and the background is indicated by the average grey value of all pixels in the object in the display. The grey value is linearly proportional to the contrast. The larger the contrast, the larger the grey value. Because the display window was using the data type ‘uint8’ to differentiate the greyscale, the dynamic range was defined as the contrast range with a grey value ranging from 0 to 255.
超声系统中的动态范围是指显示器上显示的对比范围。物体与背景之间的对比度由显示器中物体所有像素的平均灰度值表示。灰度值与对比度成线性正比。对比度越大，灰色值就越大。因为显示窗口使用数据类型“uint8”来区分灰度，所以动态范围被定义为灰度值从0到255的对比度范围。
The object with −15 dB contrast has the lowest average grey value, whereas the object with +15 dB contrast has the highest (Supplementary Fig. 16). In our case, there are six objects with different contrasts to the background in the phantom. The highest grey value obtained from the object of +15 dB contrast was 159.8, whereas the lowest grey value from the object of −15 dB contrast was 38.7. We used a linear fit to extrapolate the contrasts when the corresponding average grey values were equal to 255 and 0, which corresponded to contrasts of 39.2 dB and −24.0 dB, respectively. Then the dynamic range was determined as:
−15dB对比度的物体平均灰度值最低，而+15dB对比度的物体平均灰度值最高（补充图16）。在我们的例子中，有六个物体与幻像中的背景有不同的对比。+15 dB对比度物体的最高灰度值为159.8，而−15dB对比度物体的最低灰度值为38.7。当相应的平均灰度值等于255和0时，我们使用线性拟合来推断对比度，它们对应的对比度分别为39.2 dB和−24.0dB。然后确定其动态范围为：
The dead zone is defined as the depth of the first line phantom that is not overwhelmed by the initial pulses. The dead zone was tested by imaging a specific set of wire phantoms with different depths right beneath the device (Supplementary Fig. 11, position 4) directly and measuring the line phantoms that were visible in the B-mode image.
死区被定义为不被初始脉冲淹没的第一行幻像的深度。通过直接成像设备下方一组不同深度的特定金属丝幻像（补充图11，位置4），并测量在B-模式图像中可见的线幻像，来测试死区。
The bandwidth of the imager is defined as the ratio between the full width at half maximum in the frequency spectrum and the centre frequency. It was measured by a pulse-echo test. A piece of glass was placed 4 cm away from the device and the reflection waveform was collected with a single transducer. The collected reflection waveform was converted to the frequency spectrum by a fast Fourier transform. The full width at half maximum was read from the frequency spectrum. We obtained the bandwidth using:
成像仪的带宽定义为频谱中半最大值的全宽与中心频率的比值。它是通过脉冲回波测试来测量的。将一块玻璃放置在距离设备4厘米的地方，用一个换能器收集反射波形。通过快速傅里叶变换将采集到的反射波形转换为频谱。从频谱中读取半最大值的全宽度。我们使用以下获取带宽：
Contrast sensitivity represents the capability of the device to differentiate objects with different brightness contrasts 20 . The contrast sensitivity was tested with the greyscale objects. The contrast sensitivity is defined as the contrast-to-noise ratio (CNR) of the objects having certain contrasts to the background in the B-mode image:
对比灵敏度表示该设备区分具有不同亮度对比度的物体的能力20。用灰度物体进行了对比灵敏度测试。对比灵敏度定义为与B-模式图像中背景有一定对比度的物体的对比噪声比（CNR）：
in which μ in and σ in are the mean and the standard deviation of pixel intensity within the object, and μ out and σ out are the mean and the standard deviation of pixel intensity of the background.
其中μ in和σ in为物体内像素强度的均值和标准差，μ out和σ out为背景像素强度的均值和标准差。
The insertion loss is defined as the energy loss during the transmission and receiving. It was tested in water with a quartz crystal, a function generator with an output impedance of 50 Ω and an oscilloscope (Rigol DS1104). First, the transducer received an excitation in the form of a tone burst of a 3-MHz sine wave from the function generator. Then the same transducer received the echo from the quartz crystal. Given the 1.9-dB energy loss of the transmission into the quartz crystal and the 2.2 × 10 −4 dB (mm MHz) −1 attenuation of water, the insertion loss could be calculated as:
插入损耗定义为传输和接收过程中的能量损失。用石英晶体、输出阻抗为50 Ω的函数发生器和示波器（Rigol DS1104）在水中进行了测试。首先，换能器从函数发生器接收到一个3mhz正弦波的音调突发形式的激励。然后，相同的传感器接收到来自石英晶体的回声。考虑到石英晶体的1.9dB能量损耗和2.2×10−4dB（mmMHz）−1衰减，插入损耗可计算为：
Simulation of theacoustic field
The simulation computes the root mean square of the acoustic pressure at each point in the defined simulation field. The root mean square is defined in the equation below and gives an average acoustic pressure over a certain time duration, which is pre-defined in a packaged function of the software. In the equation, x i is the simulated acoustic pressure at the ith time step.
声场的模拟
仿真计算了在定义的模拟场中每个点的声压的均方根。均方根定义在下面的方程中定义，并给出了在一定时间内的平均声压，这是在软件的打包函数中预先定义的。在式中，xi是第i个时间步长的模拟声压。
Figure2c is the simulated root mean square of the transmitted acoustic pressure field by the orthogonal transducers. The simulation was done using the MATLAB UltraSound Toolbox 67 . Each one-dimensional phased array in the orthogonal transducers gives a sector-shaped acoustic pressure field. The simulation merges two such sector-shaped acoustic pressure fields. The imaging procedure was done with the same parameters as the simulations.
图2c为正交传感器模拟的传输声压场的均方根。模拟是使用MATLAB超音工具箱67完成的。正交传感器中的每一个一维相控阵都给出了一个扇形的声压场。模拟合并了两个这样的扇形声压场。成像过程采用与模拟相同的参数。
In the simulation, we defined the transducer parameters first: the centre frequency of the transducers as 3 MHz, the width of the transducers as 0.3 mm, the length of the transducers as 2.3 mm, the pitch of the array as 0.4 mm, the number of elements as 32 and the bandwidth of the transducers as 55%. Then we defined wide-beam compounding (Supplementary Fig. 13) as the transmission method: 97 transmission angles, from −37.5° to +37.5°, with a step size of 0.78°. Then the acoustic pressure field was the overall effect of the 97 transmissions. Finally, we defined the computation area: −8 mm to +8 mm in the lateral direction, −6 mm to +6 mm in the elevational direction and 0 mm to 140 mm in the axial direction.
在模拟中，我们定义传感器参数：传感器的中心频率为3 MHz，传感器的宽度为0.3毫米，传感器的长度为2.3毫米，阵列的间距为0.4毫米，元素的数量为32和传感器的带宽为55%。然后我们将宽光束复合（补充图13）定义为传输方法：97个传输角，从−37.5到+37.5，步长为0.78。然后，声压场是97次传输的整体效应。最后，我们定义了计算面积：横向−8mm~+8mm，高程−−6mm~+6mm，轴向−0mm~140mm。
END
▼ 点击“阅读原文”查看更多精彩视频！
作者: Naci... 时间: 2023-8-26 20:19
不错，值得学一学
| 欢迎光临 新青年麻醉论坛 (https://xqnmz.com/) | Powered by Discuz! X3.2 |