FPGA-based Lightweight QDS-CNN System for sEMG Gesture and Force Level Recognition

Deep learning (DL) has been used for electromyographic (EMG) signal recognition and achieved high accuracy for multiple classification tasks. However, implementation in resource-constrained prostheses and human-computer interaction devices remains challenging. To overcome these problems, this paper implemented a low-power system for EMG gesture and force level recognition using Zynq architecture. Firstly, a lightweight network model structure was proposed by Ultra-lightweight depth separable convolution (UL-DSC) and channel attention-global average pooling (CA-GAP) to reduce the computational complexity while maintaining accuracy. A wearable EMG acquisition device for real-time data acquisition was subsequently developed with size of 36mm×28mm×4mm. Finally, a highly parallelized dedicated hardware accelerator architecture was designed for inference computation. 18 gestures were tested, including force levels from 22 healthy subjects. The results indicate that the average accuracy rate was 94.92% for a model with 5.0k parameters and a size of 0.026MB. Specifically, the average recognition accuracy for static and force-level gestures was 98.47% and 89.92%, respectively. The proposed hardware accelerator architecture was deployed with 8-bit precision, a single-frame signal inference time of 41.9μs, a power consumption of 0.317W, and a data throughput of 78.6 GOP/s.

Effects of Training and Calibration Data on Surface Electromyogram-Based Recognition for Upper Limb Amputees

Surface electromyogram (sEMG)-based gesture recognition has emerged as a promising avenue for developing intelligent prostheses for upper limb amputees. However, the temporal variations in sEMG have rendered recognition models less efficient than anticipated. By using cross-session calibration and increasing the amount of training data, it is possible to reduce these variations. The impact of varying the amount of calibration and training data on gesture recognition performance for amputees is still unknown. To assess these effects, we present four datasets for the evaluation of calibration data and examine the impact of the amount of training data on benchmark performance. Two amputees who had undergone amputations years prior were recruited, and seven sessions of data were collected for analysis from each of them. Ninapro DB6, a publicly available database containing data from ten healthy subjects across ten sessions, was also included in this study. The experimental results show that the calibration data improved the average accuracy by 3.03%, 6.16%, and 9.73% for the two subjects and Ninapro DB6, respectively, compared to the baseline results. Moreover, it was discovered that increasing the number of training sessions was more effective in improving accuracy than increasing the number of trials. Three potential strategies are proposed in light of these findings to enhance cross-session models further. We consider these findings to be of the utmost importance for the commercialization of intelligent prostheses, as they demonstrate the criticality of gathering calibration and cross-session training data, while also offering effective strategies to maximize the utilization of the entire dataset.

Miniature Fourier Transform Spectrometer Based on Thin-Film Lithium Niobate

A miniature Fourier transform spectrometer is proposed using a thin-film lithium niobate electro-optical modulator instead of the conventional modulator made by titanium diffusion in lithium niobate. The modulator was fabricated by a contact lithography process, and its voltage-length and optical waveguide loss were 2.26 V·cm and 1.01 dB/cm, respectively. Based on the wavelength dispersion of the half-wave voltage of the fabricated modulator, the emission spectrum of the input signal was retrieved by Fourier transform processing of the interferogram, and the analysis of the experimental data of monochromatic light shows that the proposed miniaturized FTS can effectively identify the input signal wavelength.

Low-loss, ultracompact n-adjustable waveguide bends for photonic integrated circuits

Countless waveguides have been designed based on four basic bends: circular bend, sine/cosine bend, Euler bend (developed in 1744) and Bezier bend (developed in 1962). This paper proposes an n-adjustable (NA) bend, which has superior properties compared to other basic bends. Simulations and experiments indicate that the NA bends can show lower losses than other basic bends by adjusting n values. The circular bend and Euler bend are special cases of the proposed NA bend as n equals 0 and 1, respectively. The proposed bend are promising candidates for low-loss compact photonic integrated circuits.

Wearable and flexible electrochemical sensors for sweat analysis: a review

Flexible wearable sweat sensors allow continuous, real-time, noninvasive detection of sweat analytes, provide insight into human physiology at the molecular level, and have received significant attention for their promising applications in personalized health monitoring. Electrochemical sensors are the best choice for wearable sweat sensors due to their high performance, low cost, miniaturization, and wide applicability. Recent developments in soft microfluidics, multiplexed biosensing, energy harvesting devices, and materials have advanced the compatibility of wearable electrochemical sweat-sensing platforms. In this review, we summarize the potential of sweat for medical detection and methods for sweat stimulation and collection. This paper provides an overview of the components of wearable sweat sensors and recent developments in materials and power supply technologies and highlights some typical sensing platforms for different types of analytes. Finally, the paper ends with a discussion of the challenges and a view of the prospective development of this exciting field.

Intelligent and highly sensitive strain sensor based on indium tin oxide micromesh with a high crack density

Cracks play an important role in strain sensors. However, a systematic analysis of how cracks influence the strain sensors has not been proposed. In this work, an intelligent and highly sensitive strain sensor based on indium tin oxide (ITO)/polyurethane (PU) micromesh is realized. The micromesh has good skin compatibility, water vapor permeability, and stability. Due to the color of the ITO/PU micromesh, it can be invisible on the skin. Based on the fragility of ITO, the density and resistance of cracks in the micromesh are greatly improved. Therefore, the ITO/PU micromesh strain sensor (IMSS) has an ultrahigh gauge factor (744.3). In addition, a finite element model based on four resistance layers is proposed to explain the performance of the IMSS and show the importance of high-density cracks. Compared with other strain sensors based on low-density cracks, the IMSS based on high-density cracks has larger sensitivity and better linearity. Physiological signals, such as respiration, pulse, and joint motion, can be monitored using the IMSS self-fixed on the skin. Finally, an invisible and artificial throat has been realized by combining the IMSS with a convolutional neural network algorithm. The artificial throat can translate the throat vibrations of the tester automatically with an accuracy of 86.5%. This work has great potential in health care and language function reconstruction.

Vertical MoS2 transistors with sub-1-nm gate lengths

Ultra-scaled transistors are of interest in the development of next-generation electronic devices^1-3. Although atomically thin molybdenum disulfide (MoS₂) transistors have been reported⁴, the fabrication of devices with gate lengths below 1 nm has been challenging⁵. Here we demonstrate side-wall MoS₂ transistors with an atomically thin channel and a physical gate length of sub-1 nm using the edge of a graphene layer as the gate electrode. The approach uses large-area graphene and MoS₂ films grown by chemical vapour deposition for the fabrication of side-wall transistors on a 2-inch wafer. These devices have On/Off ratios up to 1.02 × 10⁵ and subthreshold swing values down to 117 mV dec^-1. Simulation results indicate that the MoS₂ side-wall effective channel length approaches 0.34 nm in the On state and 4.54 nm in the Off state. This work can promote Moore's law of the scaling down of transistors for next-generation electronics.

Gate-Tunable Negative Differential Resistance Behaviors in a hBN-Encapsulated BP-MoS2 Heterojunction

Two-dimensional (2D) heterostructures show great potential in achieving negative differential resistance (NDR) effects by Esaki diodes and or resonant tunneling diodes. However, most of the reported Esaki diode-based NDR devices realized by bulk 2D films lack sufficient gate tunability, and the tuning of NDR behavior from appearing to vanishing remains elusive. Here, a gate-tunable NDR device is reported based on a vertically stacked black phosphorus (BP) and molybdenum disulfide (MoS₂) thin 2D heterojunction. At room temperature, a rectifying ratio of ∼6 orders of magnitude from a reverse rectifying diode to a forward rectifying diode by gate modulation is obtained. Through analyzing the temperature-dependent electrical properties, the tunneling mechanism at a certain gate voltage range is revealed. Moreover, the switchable and continuously gate-tunable NDR behavior is realized with a maximum peak-to-valley ratio of 1.23 at 77 K, as shown in the I_DS mappings by sweeping V_DS under different V_GS. In addition, a compact model for gate-tunable NDR behavior in 2D heterostructures is proposed. To our best knowledge, this is the first demonstration of NDR behavior in BP-MoS₂ heterostructures. Consequently, this work sheds light on the gate-tunable NDR devices and reconfigurable logic devices for realizing ternary and reconfigurable logic systems.

Multifunctional Graphene Microstructures Inspired by Honeycomb for Ultrahigh Performance Electromagnetic Interference Shielding and Wearable Applications

High-performance electromagnetic interference (EMI) shielding materials with ultralow density, excellent flexibility, and good mechanical properties are highly desirable for aerospace and wearable electronics. Herein, honeycomb porous graphene (HPG) fabricated by laser scribing technology is reported for EMI shielding and wearable applications. Due to the honeycomb structure, the HPG exhibits an EMI shielding effectiveness (SE) up to 45 dB at a thickness of 48.3 μm. The single-piece HPG exhibits an ultrahigh absolute shielding effectiveness (SSE/t) of 240 123 dB cm²/g with an ultralow density of 0.0388 g/cm³, which is significantly superior to the reported materials such as carbon-based, MXene, and metal materials. Furthermore, MXene and AgNWs are employed to cover the honeycomb holes of the HPG to enhance surface reflection; thus, the SSE/t of the HPG/AgNWs composite membrane can reach up to 292 754 dB cm²/g. More importantly, the HPG exhibits excellent mechanical stability and durability in cyclic stretching and bending, which can be used to monitor weak physiological signals such as pulse, respiration, and laryngeal movement of humans. Therefore, the lightweight and flexible HPG exhibits excellent EMI shielding performance and mechanical properties, along with its low cost and ease of mass production, which is promising for practical applications in EMI shielding and wearable electronics.

Substrate-Free Multilayer Graphene Electronic Skin for Intelligent Diagnosis

Current wearable sensors are fabricated with substrates, which limits the comfort, flexibility, stretchability, and induces interface mismatch. In addition, the substrate prevents the evaporation of sweat and is harmful to skin health. In this work, we have enabled the substrate-free laser scribed graphene (SFG) electronic skin (e-skin) with multifunctions. Compared with the e-skin with the substrate, the SFG has good gas permeability, low impedance, and flexibility. Only assisted using water, the SFG can be transferred to almost any objects including silicon and human skin and it can even be suspended. Many through-holes like stomas in leaf can be formed in the SFG, which make it breathable. After designing the pattern, the gauge factor (GF) of graphene electronic skin (GES) can be designed as the strain sensor. Physiological signals such as respiration, human motion, and electrocardiogram (ECG) can be detected. Moreover, the suspended SFG detect vibrations with high sensitivity. Due to the substrate-free structure, the impedance between SFG e-skin and the human body decreases greatly. Finally, an ECG detecting system has been designed based on the GES, which can monitor the body condition in real time. To analyze the ECG signals automatically, a convolutional neural network (CNN) was built and trained successfully. This work has high potential in the field of health telemonitoring.

Graphene-Based Thermoacoustic Sound Source

Thermoacoustic (TA) effect has been discovered for more than 130 years. However, limited by the material characteristics, the performance of a TA sound source could not be compared with magnetoelectric and piezoelectric loudspeakers. Recently, graphene, a two-dimensional material with the lowest heat capacity per unit area, was discovered to have a good TA performance. Compared with a traditional sound source, graphene TA sound sources (GTASSs) have many advantages, such as small volume, no diaphragm vibration, wide frequency range, high transparency, good flexibility, and high sound pressure level (SPL). Therefore, graphene has a great potential as a next-generation sound source. Photoacoustic (PA) imaging can also be applied to the diagnosis and treatment of diseases using the photothermo-acoustic (PTA) effect. Therefore, in this review, we will introduce the history of TA devices. Then, the theory and simulation model of TA will be analyzed in detail. After that, we will talk about the graphene synthesis method. To improve the performance of GTASSs, many strategies such as lowering the thickness and using porous or suspended structures will be introduced. With a good PTA effect and large specific area, graphene PA imaging and drug delivery is a promising prospect in cancer treatment. Finally, the challenges and prospects of GTASSs will be discussed.

High-performance humidity sensor using Schottky-contacted SnS nanoflakes for noncontact healthcare monitoring

Humidity sensors based on flexible sensitive nanomaterials are very attractive in noncontact healthcare monitoring. However, the existing humidity sensors have some shortcomings such as limited sensitivity, narrow relative humidity (RH) range, and a complex process. Herein, we show that a tin sulphide (SnS) nanoflakes-based sensor presents high humidity sensing behaviour both in rigid and flexible substrate. The sensing mechanism based on the Schottky nature of a SnS-metal contact endows the as-fabricated sensor with a high response of 2491000% towards a wide RH range from 3% RH to 99% RH. The response and recovery time of the sensor are 6 s and 4 s, respectively. Besides, the flexible SnS nanoflakes-based humidity sensor with a polyimide substrate can be well attached to the skin and exhibits stable humidity sensing performance in the natural flat state and under bending loading. Moreover, the first-principles analysis is performed to prove the high specificity of SnS to the moisture (H₂O) in the air. Benefiting from its promising advantages, we explore some application of the SnS nanoflakes-based sensors in detection of breathing patterns and non-contact finger tips sensing behaviour. The sensor can monitor the respiration pattern of a human being accurately, and recognize the movement of the fingertip speedily. This novel humidity sensor shows great promising application in physiological and physical monitoring, portable diagnosis system, and noncontact interface localization.

A Wearable Skinlike Ultra-Sensitive Artificial Graphene Throat

Most mute people cannot speak due to their vocal cord lesion. Herein, to assist mute people to "speak", we proposed a wearable skinlike ultrasensitive artificial graphene throat (WAGT) that integrated both sound/motion detection and sound emission in single device. In this work, the growth and patterning of graphene can be realized at the same time, and a thin poly(vinyl alcohol) film with laser-scribed graphene was obtained by a water-assisted transferring process. In virtue of the skinlike and low-resistant substrate, the WAGT has a high detection sensitivity (relative resistance changes up to 150% at 133 Ω) and an excellent sound-emitting ability (up to 75 dB at 0.38 W power and 2 mm distance). On the basis of the excellent mechanical-electrical performance of graphene structure, the sound detecting and emitting mechanisms of WAGT are realized and discussed. For sound detection, both the motion of larynx and vibration of vocal cord contribute to throat movements. For sound emission, a thermal acoustic model for WAGT was established to reveal the principle of sound emitting. More importantly, a homemade circuit board was fabricated to build a dual-mode system, combining the detection and emitting systems. Meanwhile, different human motions, such as strong and small throat movements, were also detected and transformed into different sounds like "OK" and "NO". Therefore, the implementation of these sound/motion detection acoustic systems enable graphene to achieve device-level applications to system-level applications, and those graphene acoustic systems are wearable for its miniaturization and light weight.

Artificial Synapses Based on in-Plane Gate Organic Electrochemical Transistors

Realization of biological synapses using electronic devices is regarded as the basic building blocks for neuromorphic engineering and artificial neural network. With the advantages of biocompatibility, low cost, flexibility, and compatible with printing and roll-to-roll processes, the artificial synapse based on organic transistor is of great interest. In this paper, the artificial synapse simulation by ion-gel gated organic field-effect transistors (FETs) with poly(3-hexylthiophene) (P3HT) active channel is demonstrated. Key features of the synaptic behaviors, such as paired-pulse facilitation (PPF), short-term plasticity (STP), self-tuning, the spike logic operation, spatiotemporal dentritic integration, and modulation are successfully mimicked. Furthermore, the interface doping processes of electrolyte ions between the active P3HT layer and ion gels is comprehensively studied for confirming the operating processes underlying the conductivity and excitatory postsynaptic current (EPSC) variations in the organic synaptic devices. This study represents an important step toward building future artificial neuromorphic systems with newly emerged ion gel gated organic synaptic devices.

High-performance ultraviolet photodetectors based on CdS/CdS:SnS2 superlattice nanowires

CdS heterostructure nanomaterials are attractive for their potential applications in integrated optoelectronic devices. Herein, the high-quality CdS/CdS:SnS2 superlattice nanowires were synthesized through a micro-environmental controlled co-evaporation technique, which shows periodic emission properties and that their structures are periodic and alternating. For the first time, we demonstrate the fabrication of high-performance ultraviolet photodetectors using unique CdS/CdS:SnS2 superlattice nanowires. The optoelectronic properties of the photodetectors were studied and compared to those devices based on pure CdS nanowires. The as-fabricated photodetectors (under 365 nm) based on CdS/CdS:SnS2 superlattice nanowires showed a high photocurrent to dark current ratio of 10(5), a large photoresponsivity of 2.5 × 10(3) A W(-1), a fast response time of 10 ms and an excellent external quantum efficiency of 8.6 × 10(5) at room temperature, which shows better performance than pure CdS nanowires photodetectors. The results indicate that CdS/CdS:SnS2 superlattice nanowires are very promising potential candidates in nanoscale electronic and optoelectronic devices.