This paper presents compact micro actuators that achieve large output force and stroke utilizing paraffin as phase change material (PCM) activated via joule heat. A glass chamber, sealed with a double-layered composite diaphragm, constrains the deflection induced by phase change in the out-of-plane direction. Each individual actuator has a volume of less than 1 mm(3). Devices with 700 mu m and 300 mu m chamber depths achieved maximum deflection of 233.51 mu m and 90.27 mu m, and work density up to 26.08x10(3) J/m(3) and 3.50x10(3) J/m(3), respectively. A 3-actuator array (5 mmx2 mmx1 mm) was integrated with a microchannel to form a micropump, delivering ultra-low flowrate of 4-6 nL/min.

Parylene C is a common substrate and encapsulation material used in implantable microelectrodes. Its reliability and failure are of great significance in the research and application of microelectrodes. In this study, three different failure stages of Parylene C thin-film electrodes were modeled using equivalent circuits, and the electric impedance spectroscopy of the electrodes were rapidly analyzed 9 different machine learning algorithms to identify the failure stages. The results showed that the three equivalent circuit models (ECMs) can represent the dynamics of the three failure stages of the Parylene C thin-film electrodes. The support vector machine (SVM) algorithm achieves more than 93% accuracy in identifying the ECMs from electric impedance spectroscopy data with an average time of 0.0273 s. The SVM algorithm has great potential in fast analysis of electric impedance spectroscopy for the endurability study and application of implantable microelectrodes.

A weak signal detection circuit based on AC modulation is designed to address issues such as zero drift and low-frequency noise interference in DC amplification circuits. The system employs the AD630 phase-sensitive detection circuit as its core, along with amplification and filtering circuits, signal generation circuits, and sensitive components. By utilizing AC modulation technique, the DC signal is converted into an AC signal, eliminating the zero drift issue in the DC amplification circuit and suppressing low-frequency interference to improve system accuracy. The circuit adopts cross-correlation detection method, leveraging the uncorrelated nature of noise and reference signals to enhance anti-interference capability, ensuring that the circuit output voltage is zero when the sensor has zero input. Experimental results demonstrate that the adoption of modulation-demodulation technique can improve circuit accuracy by 3.74 times compared to the original DC amplification circuit. The AC amplification circuit exhibits significantly superior anti-interference performance over the DC amplification circuit, enabling the extraction of the tested signal with an amplitude smaller than 30mV in the presence of 1V noise, with an error of only 0.408%.

This paper presents a self-sensing soft pneumatic micro actuator based on a liquid metal capacitive sensor, which provides a prospect of realizing more precise and controllable fully flexible haptic interface. The flexible pneumatic actuator is less than 500 μm in radius, close to the haptic spatial resolution of human skin. It features large displacement (about 140 μm), large output force (over 160 mN) and wide frequency bandwidth (at least 60 Hz). A completely soft capacitive sensor is integrated with the actuator for self-sensing capability. Its liquid-metal electrodes can withstand large local strain brought by the actuator. The self-sensing feature is stable and repeatable under continuous stimulus over 500 cycles and exhibits good frequency characteristics.

Morphology control can enable multifunction micro devices, but the fabrication of 3D structures is challenging. This paper presents a planar fabrication compatible pneumatic actuator based on corrugation structure. The actuator exhibits large deformation and high actuation force on sub-millimeter scale. Furthermore, the deformation is self-sustainable after actuation, and the deformation morphology is programable by tuning the actuator pattern. The actuator enables a novel access to achieve 3D morphology control in micro-devices.

Piezoelectric micromachined ultrasonic transducers (pMUTs) have attracted widespread attentions on account of their vital applications in medical imaging, fingerprint identification, range-finding, gesture recognition, and so on. The piezoelectric layer, which is the key component for pMUTs, has been dominated by ferroelectric Pb(Zr, Ti)O-3 (PZT) because of its outstanding piezoelectric performance, especially in the case of pMUT transmitters. However, the toxicity of lead restricts their medical-related applications, so developing lead-free pMUTs should be of great significance but remains challenging. As a precedent, the present work pioneered the processing of lead-free (K,Na)NbO3 (KNN)-based pMUTs, and obtained an ultrahigh transmitting sensitivity of 1250 nm V-1 at 66.2 kHz, which is superior to the majority of PZT-based pMUTs reported so far. The excellent displacement sensitivity can be ascribed to the dense and homogeneous feature, (100)-textured structure, orthorhombic -tetragonal phase coexistence, nanosized domains, great electrical insulation and high piezoresponse of the KNN-based film as well as the the elaborate micromachining process. This work initiated a successful precedent for applying lead-free ferroelectric films to pMUTs also as a breakthrough step in promoting KNN-based films toward practical applications.

A novel coronavirus disease 2019 (COVID-19) was detected and has spread rapidly across various countries around the world since the end of the year 2019. Computed Tomography (CT) images have been used as a crucial alternative to the time-consuming RT-PCR test. However, pure manual segmentation of CT images faces a serious challenge with the increase of suspected cases, resulting in urgent requirements for accurate and automatic segmentation of COVID-19 infections. Unfortunately, since the imaging characteristics of the COVID-19 infection are diverse and similar to the backgrounds, existing medical image segmentation methods cannot achieve satisfactory performance. In this article, we try to establish a new deep convolutional neural network tailored for segmenting the chest CT images with COVID-19 infections. We first maintain a large and new chest CT image dataset consisting of 165,667 annotated chest CT images from 861 patients with confirmed COVID-19. Inspired by the observation that the boundary of the infected lung can be enhanced by adjusting the global intensity, in the proposed deep CNN, we introduce a feature variation block which adaptively adjusts the global properties of the features for segmenting COVID-19 infection. The proposed FV block can enhance the capability of feature representation effectively and adaptively for diverse cases. We fuse features at different scales by proposing Progressive Atrous Spatial Pyramid Pooling to handle the sophisticated infection areas with diverse appearance and shapes. The proposed method achieves state-of-the-art performance. Dice similarity coefficients are 0.987 and 0.726 for lung and COVID-19 segmentation, respectively. We conducted experiments on the data collected in China and Germany and show that the proposed deep CNN can produce impressive performance effectively. The proposed network enhances the segmentation ability of the COVID-19 infection, makes the connection with other techniques and contributes to the development of remedying COVID-19 infection.

Liver vessel segmentation is fast becoming a key instrument in the diagnosis and surgical planning of liver diseases. In clinical practice, liver vessels are normally manual annotated by clinicians on each slice of CT images, which is extremely laborious. Several deep learning methods exist for liver vessel segmentation, however, promoting the performance of segmentation remains a major challenge due to the large variations and complex structure of liver vessels. Previous methods mainly using existing UNet architecture, but not all features of the encoder are useful for segmentation and some even cause interferences. To overcome this problem, we propose a novel deep neural network for liver vessel segmentation, called LVSNet, which employs special designs to obtain the accurate structure of the liver vessel. Specifically, we design Attention-Guided Concatenation (AGC) module to adaptively select the useful context features from low-level features guided by high-level features. The proposed AGC module focuses on capturing rich complemented information to obtain more details. In addition, we introduce an innovative multi-scale fusion block by constructing hierarchical residual-like connections within one single residual block, which is of great importance for effectively linking the local blood vessel fragments together. Furthermore, we construct a new dataset containing 40 thin thickness cases (0.625 mm) which consist of CT volumes and annotated vessels. To evaluate the effectiveness of the method with minor vessels, we also propose an automatic stratification method to split major and minor liver vessels. Extensive experimental results demonstrate that the proposed LVSNet outperforms previous methods on liver vessel segmentation datasets. Additionally, we conduct a series of ablation studies that comprehensively support the superiority of the underlying concepts.

A novel coronavirus disease 2019 (COVID-19) was detected and has spread rapidly across various countries around the world since the end of the year 2019, Computed Tomography (CT) images have been used as a crucial alternative to the time-consuming RT-PCR test. However, pure manual segmentation of CT images faces a serious challenge with the increase of suspected cases, resulting in urgent requirements for accurate and automatic segmentation of COVID-19 infections. Unfortunately, since the imaging characteristics of the COVID-19 infection are diverse and similar to the backgrounds, existing medical image segmentation methods cannot achieve satisfactory performance. In this work, we try to establish a new deep convolutional neural network tailored for segmenting the chest CT images with COVID-19 infections. We firstly maintain a large and new chest CT image dataset consisting of 165,667 annotated chest CT images from 861 patients with confirmed COVID-19. Inspired by the observation that the boundary of the infected lung can be enhanced by adjusting the global intensity, in the proposed deep CNN, we introduce a feature variation block which adaptively adjusts the global properties of the features for segmenting COVID-19 infection. The proposed FV block can enhance the capability of feature representation effectively and adaptively for diverse cases. We fuse features at different scales by proposing Progressive Atrous Spatial Pyramid Pooling to handle the sophisticated infection areas with diverse appearance and shapes. We conducted experiments on the data collected in China and Germany and show that the proposed deep CNN can produce impressive performance effectively.

The sudden outbreak of novel coronavirus 2019 (COVID-19) increased the diagnostic burden of radiologists. In the time of an epidemic crisis, we hope artificial intelligence (AI) to reduce physician workload in regions with the outbreak, and improve the diagnosis accuracy for physicians before they could acquire enough experience with the new disease. In this paper, we present our experience in building and deploying an AI system that automatically analyzes CT images and provides the probability of infection to rapidly detect COVID-19 pneumonia. The proposed system which consists of classification and segmentation will save about 30%–40% of the detection time for physicians and promote the performance of COVID-19 detection. Specifically, working in an interdisciplinary team of over 30 people with medical and/or AI background, geographically distributed in Beijing and Wuhan, we are able to overcome a series of challenges (e.g. data discrepancy, testing time-effectiveness of model, data security, etc.) in this particular situation and deploy the system in four weeks. In addition, since the proposed AI system provides the priority of each CT image with probability of infection, the physicians can confirm and segregate the infected patients in time. Using 1,136 training cases (723 positives for COVID-19) from five hospitals, we are able to achieve a sensitivity of 0.974 and specificity of 0.922 on the test dataset, which included a variety of pulmonary diseases.

Objective: To improve the existing manually assembled cochlear implant electrode arrays, a thin-film electrode array (TFEA) was microfabricated having a maximum electrode density of 15 sites along an 8-mm length, with each site having a 75 μm × 1.8 μm (diameter × height) disk electrode. Methods: The microfabrication method adopted photoresist transferring, lift-off, two-step oxygen plasma etching, and fuming nitric acid release to reduce lift-off complexity, protect the metal layer, and increase the release efficiency. Results: Systematic in vitro characterization showed that the TFEA's bending stiffness was 6.40 × 10−10 N·m2 near the base and 1.26 × 10−10 N·m2 near the apex. The TFEA electrode produced an average impedance of 16 kΩ and a maximum current limit of 800 μA, measured with 1-kHz sinusoidal current using monopolar stimulation in saline. A TFEA prototype was implanted in a cat cochlea to obtain in vivo measurements of electrically evoked auditory brainstem and inferior colliculus responses to monopolar stimulation with 41-μs/phase biphasic pulses. Both physiological responses produced a threshold of ∼300 μA and a dynamic range of 5–8 dB above the threshold. Compared with existing arrays, the present TFEA had 104 times less bending stiffness, 97% less electrode area, and comparable physiological thresholds. Conclusion: Using a simplified structure and stable fabrication method, the present TEFA produced physical and physiological performance comparable to existing commercial devices. Significance: The present TFEA represents a step closer toward an automated process replacing the labor-intensive and expensive manual assembly of the cochlear implant electrode arrays.

This paper reports an intracochlear electric and acoustic stimulator (EAS) prototype, which adopted the board-level integration of a piezoelectric micro-actuator and a thin film electrode array. The EAS prototype meets the cochlear size limitations, with its implant part having a maximum diagonal dimension of <1.5 mm, and a 21-mm-long electrode array complying with the first cochlear turn. When the electric and acoustic stimulators work simultaneously, the amplitude of the interfering stimulus' frequency component accounts for <2% of the base frequency component. Compared to the state of the art in the hybrid cochlear implant, for the first time, the prototype proved the feasibility of intracochlear hybrid stimulation with an integrated stimulator. The EAS prototype represents a promising solution to improve the patients' wearing comfort and application scenarios.

This paper presents a flexible Parylene-based micro pneumatic actuator featuring a large displacement, which suggests potential applications in micro robotic activation and deformable implanted transducers. Due to the inelastic mechanical property of Parylene, corrugated Parylene film was fabricated to introduce stretchable areas, which grants the proposed large displacement and designable deformation. The fabrication process involves sacrificial photoresist molds and titanium adhesion promoting layer, which avoid silicon etching or Parylene bonding process, for better compatibility with other fabrication processes and possible integrated design. An over 400 μm displacement of an actuator of 500 μm in radius and 20 μm in thickness at an inflating pressure of 200 kPa was demonstrated. A finite element model (FEM) was also constructed and the simulation tallies well with the experiment.

A piezoelectric thin-film microactuator in the form of an asymmetrically laminated diaphragm is developed as an intracochlear hearing aid. Experimentally, natural frequencies of the microactuator bifurcate with respect to an applied bias voltage. To qualitatively explain the findings, we model the lead-zirconate-titanate (PZT) diaphragm as a doubly curved, asymmetrically laminated, piezoelectric shallow shell defined on a rectangular domain with simply supported boundary conditions. The von Karman type nonlinear strain–displacement relationship and the Donnell–Mushtari–Vlasov theory are used to calculate the electric enthalpy and elastic strain energy. Balance of virtual work between two top electrodes is also considered to incorporate an electric-induced displacement field that has discontinuity of in-plane strain components. A set of discretized equations of motion are obtained through a variational approach.

Designing an electrode array with a high stimulation resolution (SR) is the main challenge in cochlear implant development. In this work, a thin-film electrode array (TFEA) and partial tripolar (pTP) mode were combined in the design stage to optimize the SR. A finite-element model of the intracochlear electric potential Ve incorporating a TFEA and pTP mode was built and validated using previous experimental measurements. Based on this model, the SR was analyzed by using a defined stimulation factor Vs, which takes both the amplitude and bandwidth of Ve into account. A co-simulation method integrating the model and genetic algorithm was employed to maximize Vs with an optimized parameter set including the electrode diameter d, electrode interval g, and compensation coefficient σ. The results indicated that a TFEA combined with pTP mode outperforms their individual utilization to improve the SR and that d has an independent negative correlation with the SR, but it is more effective and feasible to consider all three parameters in the design stage with the proposed model and co-simulation optimization method. In our design, the optimized parameters were d = 150 μm, g = 200.5 μm, and σ = 0.746.

The micro-fabricated thin film electrode array (TFEA) has been a promising design for cochlear implants (CIs) because of its cost-effectiveness and fabrication precision. The latest polymer-based cochlear TFEAs have faced difficulties for cochlear insertion due to the lack of structural stiffness. To stiffen the TFEA, dissolvable stiffening materials, TFEAs with different structures, and TFEAs with commercial CIs as carriers have been invested. In this work, the concept of enhancing a Parylene TFEA with Kapton tape as a simpler carrier for cochlear insertion has been proved to be feasible. The bending stiffness of the Kapton-aided TFEA was characterized with an analytical model, a finite element model, and a cantilever bending experiment, respectively. While the Kapton tape increased the bending stiffness of the Parylene TFEA by 103 times, the 6-μm-thick TFEA with a similar Young’s modulus, as a polyimide, in turn significantly increased the bending stiffness of the 170-μm-thick Kapton carrier by 60%. This result indicated that even the TFEA is ultra-flexible and that its bending stiffness should not be neglected in the design or selection of its carrier.

An intracochlear lead-zirconate-titanate (PZT) microactuator integrated with a cochlear implant electrode array could be a feasible strategy to implement combined electric and acoustic stimulation inside the cochlea. The purpose of this paper is to characterize in vitro a prototype PZT microactuator for intracochlear applications, including service life, failure mechanisms, and lead leaching. PZT microactuators were driven sinusoidally to failure in air and in artificial perilymph. Frequency response functions (FRFs) and electrical impedance were monitored. After the PZT microactuators failed, the amount of leached lead was measured via inductive coupled plasma mass spectrometry (ICP-MS). Two failure mechanisms are identified: electrical breakdown and structural failure. The electrical breakdown, possibly from loss of parylene encapsulation, is evidenced by a sudden and significant drop of the actuators' electrical resistance. The structural failure, possibly from electrode delamination, is evidenced by a sudden and significant drop of FRFs. The amount of lead leached from the PZT microactuator is well below published safety guidelines from federal agencies.

Piezoelectric thin-film micro-sensors and actuators often appear in the form of a diaphragm anchored around its entire boundary. When a micro-device is scaled down in size, its sensitivity reduces and natural frequency increases, drastically lowering its performance. The purpose of the paper is to study the feasibility of introducing through-etched slots to partially release the diaphragm at its anchor. As a result, a micro-device can be scaled down in size without significantly altering its sensitivity and natural frequency. In this paper, we first present a finite element simulation proving the concept. We then describe processing steps to fabricate a lead-zirconate-titanate (PZT) thin-film diaphragm sensor/actuator with a partially released boundary. Challenges encountered in the fabrications, such as cat ear and electrode non-uniformity, are explained and overcome. As a case study, we demonstrate the feasibility to design, fabricate, and test an intra-cochlear micro-actuator probe that employs three partially released PZT diaphragms at the tip of a cantilever. Experimental measurements indicate that the sensitivity is dominated by the diaphragm deflection, while the first natural frequency is dominated by the cantilever structure. Finite element simulations not only confirm the experimental measurements but also optimize diaphragm dimensions for sensor/actuator performance. (C) 2017 Elsevier B.V. All rights reserved.

In this paper, we present a nano-composite thin-film sensor that consists of numerous lead-zirconate-titanate (PZT) nanoparticles embedded in a silane matrix. Our main efforts include fabrication, characterization, and demonstration of the thin-film sensor. The fabrication includes the following steps. First, PZT nanoparticles, with a size distribution ranging from 300 to 800 nm, are fabricated via a hydrothermal synthesis. The PZT nanoparticles are then suspended in a silane-based fluid to form PZT ink. The PZT ink can then be printed, sprayed, or dropped onto a substrate. The deposited PZT ink is subsequently cured at low temperature (e.g., 120 degrees C) to form the PZT-silane thin-film sensor. A similar ink and thin-film sensor using crushed bulk PZT are also fabricated for reference. The characterization of the PZT-silane films includes the following efforts: (a) measurements of dielectric properties via an impedance analyzer, (b) measurements of piezoelectric charge from the PZT-silane films under an impulsive load, and (c) extraction of piezoelectric constant d(33) via a finite element analysis. To demonstrate its validity as a vibration sensor, the PZT-silane thin film is attached to a square aluminum plate supported by four pillars. The frequency response of charge measured from the PZT-silane thin-film sensor replicates the vibration measurements from a laser Doppler vibrometer. (C) 2016 Elsevier B.V. All rights reserved.