Machine-Learning-Assisted Recognition on Bioinspired Soft Sensor Arrays

Advances in Machine Learning for Wearable Sensors

Recent years have witnessed tremendous advances in machine learning techniques for wearable sensors and bioelectronics, which play an essential role in real-time sensing data analysis to provide clinical-grade information for personalized healthcare. To this end, supervised learning and unsupervised learning algorithms have emerged as powerful tools, allowing for the detection of complex patterns and relationships in large, high-dimensional data sets. In this Review, we aim to delineate the latest advancements in machine learning for wearable sensors, focusing on key developments in algorithmic techniques, applications, and the challenges intrinsic to this evolving landscape. Additionally, we highlight the potential of machine-learning approaches to enhance the accuracy, reliability, and interpretability of wearable sensor data and discuss the opportunities and limitations of this emerging field. Ultimately, our work aims to provide a roadmap for future research endeavors in this exciting and rapidly evolving area.

Wearable, Biocompatible, and Dual-Emission Ocular Multisensor Patch for Continuous Profiling of Fluoroquinolone Antibiotics in Tears

The abuse or misuse of antibiotics in clinical and agricultural settings severely endangers human health and ecosystems, which has raised profound concerns for public health worldwide. Trace detection and reliable discrimination of commonly used fluoroquinolone (FQ) antibiotics and their analogues have consequently become urgent to guide the rational use of antibiotic medicines and deliver efficient treatments for associated diseases. Herein, we report a wearable eye patch integrated with a quadruplex nanosensor chip for noninvasive detection and discrimination of primary FQ antibiotics in tears during routine eyedrop treatment. A set of dual-mode fluorescent nanoprobes of red- or green-emitting CdTe quantum dots integrated with lanthanide ions and a sensitizer, adenosine monophosphate, were constructed to provide an enhanced fluorescence up to 45-fold and nanomolar sensitivity toward major FQs owing to the aggregation-regulated antenna effect. The aggregation-driven, CdTe-Ln(III)-based microfluidic sensor chip is highly specific to FQ antibiotics against other non-FQ counterparts or biomolecular interfering species and is able to accurately discriminate nine types of FQ or non-FQ eyedrop suspensions using linear discriminant analysis. The prototyped wearable sensing detector has proven to be biocompatible and nontoxic to human tissues, which integrates the entire optical imaging modules into a miniaturized, smartphone-based platform for field use and reduces the overall assay time to ∼5 min. The practicability of the wearable eye patch was demonstrated through accurate quantification of antibiotics in a bactericidal event and the continuous profiling of FQ residues in tears after using a typical prescription antibiotic eyedrop. This technology provides a useful supplement to the toolbox for on-site and real-time examination and regulation of inappropriate daily drug use that might potentially lead to long-term antibiotic abuse and has great implications in advancing personal healthcare techniques for the regulation of daily medication therapy.

Neuromorphic Computing-Assisted Triboelectric Capacitive-Coupled Tactile Sensor Array for Wireless Mixed Reality Interaction

Flexible tactile sensors show promise for artificial intelligence applications due to their biological adaptability and rapid signal perception. Triboelectric sensors enable active dynamic tactile sensing, while integrating static pressure sensing and real-time multichannel signal transmission is key for further development. Here, we propose an integrated structure combining a capacitive sensor for static spatiotemporal mapping and a triboelectric sensor for dynamic tactile recognition. A liquid metal-based flexible dual-mode triboelectric-capacitive-coupled tactile sensor (TCTS) array of 4 × 4 pixels achieves a spatial resolution of 7 mm, exhibiting a pressure detection limit of 0.8 Pa and a fast response of 6 ms. Furthermore, neuromorphic computing using the MXene-based synaptic transistor achieves 100% recognition accuracy of handwritten numbers/letters within 90 epochs based on dynamic triboelectric signals collected by the TCTS array, and cross-spatial information communication from the perceived multichannel tactile data is realized in the mixed reality space. The results illuminate considerable application possibilities of dual-mode tactile sensing technology in human-machine interfaces and advanced robotics.

Morse Code Recognition Based on a Flexible Tactile Sensor with Carbon Nanotube/Polyurethane Sponge Material by the Long Short-Term Memory Model

Morse code recognition plays a very important role in the application of human-machine interaction. In this paper, based on the carbon nanotube (CNT) and polyurethane sponge (PUS) composite material, a flexible tactile CNT/PUS sensor with great piezoresistive characteristic is developed for detecting Morse code precisely. Thirty-six types of Morse code, including 26 letters (A-Z) and 10 numbers (0-9), are applied to the sensor. Each Morse code was repeated 60 times, and 2160 (36 × 60) groups of voltage time-sequential signals were collected to construct the dataset. Then, smoothing and normalization methods are used to preprocess and optimize the raw data. Based on that, the long short-term memory (LSTM) model with excellent feature extraction and self-adaptive ability is constructed to precisely recognize different types of Morse code detected by the sensor. The recognition accuracies of the 10-number Morse code, the 26-letter Morse code, and the whole 36-type Morse code are 99.17%, 95.37%, and 93.98%, respectively. Meanwhile, the Gated Recurrent Unit (GRU), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Random Forest (RF) models are built to distinguish the 36-type Morse code (letters of A-Z and numbers of 0-9) based on the same dataset and achieve the accuracies of 91.37%, 88.88%, 87.04%, and 90.97%, respectively, which are all lower than the accuracy of 93.98% based on the LSTM model. All the experimental results show that the CNT/PUS sensor can detect the Morse code's tactile feature precisely, and the LSTM model has a very efficient property in recognizing Morse code detected by the CNT/PUS sensor.

Displacement-pressure biparametrically regulated softness sensory system for intraocular pressure monitoring

High intraocular pressure (IOP) is one of the high-risk pathogenic factors of glaucoma. Existing methods of IOP measurement are based on the direct interaction with the cornea. Commercial ophthalmic tonometers based on snapshot measurements are expensive, bulky, and their operation requires trained personnel. Theranostic contact lenses are easy to use, but they may block vision and cause infection. Here, we report a sensory system for IOP assessment that uses a soft indentor with two asymmetrically deployed iontronic flexible pressure sensors to interact with the eyelid-eyeball in an eye-closed situation. Inspired by human fingertip assessment of softness, the sensory system extracts displacement-pressure information for soft evaluation, achieving high accuracy IOP monitoring (>96%). We further design and custom-make a portable and wearable ophthalmic tonometer based on the sensory system and demonstrate its high efficacy in IOP screening. This sensory system paves a way towards cost-effective, robust, and reliable IOP monitoring.

AI-Driven Sensing Technology: Review

Machine learning and deep learning technologies are rapidly advancing the capabilities of sensing technologies, bringing about significant improvements in accuracy, sensitivity, and adaptability. These advancements are making a notable impact across a broad spectrum of fields, including industrial automation, robotics, biomedical engineering, and civil infrastructure monitoring. The core of this transformative shift lies in the integration of artificial intelligence (AI) with sensor technology, focusing on the development of efficient algorithms that drive both device performance enhancements and novel applications in various biomedical and engineering fields. This review delves into the fusion of ML/DL algorithms with sensor technologies, shedding light on their profound impact on sensor design, calibration and compensation, object recognition, and behavior prediction. Through a series of exemplary applications, the review showcases the potential of AI algorithms to significantly upgrade sensor functionalities and widen their application range. Moreover, it addresses the challenges encountered in exploiting these technologies for sensing applications and offers insights into future trends and potential advancements.

Paper-based fluorescence sensor array with functionalized carbon quantum dots for bacterial discrimination using a machine learning algorithm

The rapid discrimination of bacteria is currently an emerging trend in the fields of food safety, medical detection, and environmental observation. Traditional methods often require lengthy culturing processes, specialized analytical equipment, and bacterial recognition receptors. In response to this need, we have developed a paper-based fluorescence sensor array platform for identifying different bacteria. The sensor array is based on three unique carbon quantum dots (CQDs) as sensing units, each modified with a different antibiotic (polymyxin B, ampicillin, and gentamicin). These antibiotic-modified CQDs can aggregate on the bacterial surface, triggering aggregation-induced fluorescence quenching. The sensor array exhibits varying fluorescent responses to different bacterial species. To achieve low-cost and portable detection, CQDs were formulated into fluorescent ink and used with an inkjet printer to manufacture paper-based sensor arrays. A smartphone was used to collect the responses generated by the bacteria and platform. Diverse machine learning algorithms were utilized to discriminate bacterial types. Our findings showcase the platform's remarkable capability to differentiate among five bacterial strains, within a detection range spanning from 1.0 × 103 CFU/mL to 1.0 × 107 CFU/mL. Its practicality is further validated through the accurate identification of blind bacterial samples. With its cost-effectiveness, ease of fabrication, and high degree of integration, this platform holds significant promise for on-site detection of diverse bacteria.

On-Demand MXene-Coupled Pyroelectricity for Advanced Breathing Sensors and IR Data Receivers

MXene-inspired two-dimensional (2D) materials like Ti3C2Tx are widely known for their versatile properties, including surface plasmon, higher electrical conductivity, exceptional in-plane tensile strength, EMI shielding, and IR thermal properties. The MXene nanosheets coupled poly(vinylidene fluoride) (PVDF) nanofibers with d33 ∼-26 pm V-1 are able to capture the smaller thermal fluctuation due to a superior pyroelectric coefficient of ∼130 nC m-2 K-1 with an improved (∼7 times with respect to neat PVDF nanofibers) pyroelectric current figure of merit (FOMi). The significant enhancement of the pyroelectric response is attributed to the confinement effect of 2D MXene (Ti3C2Tx) nanosheets within PVDF nanofibers, as evidenced from polarized Fourier transform infrared (FTIR) spectroscopy and scanning probe microscopy (SPM). In subsequent studies, the practical applications of self-powered pyroelectric sensors of MXene-PVDF have been demonstrated. The fabricated flexible, hydrophobic pyroelectric sensor could be utilized as an excellent pyroelectric breathing sensor, a proximity sensor, and an IR data transmission receiver. Further, supervised machine learning algorithms are proposed to distinguish different types of breathing signals with ∼98% accuracy for healthcare monitoring purposes.

Machine learning-coupled tactile recognition with high spatiotemporal resolution based on cross-striped nanocarbon piezoresistive sensor array

Flexible pressure sensor arrays have been playing important roles in various applications of human-machine interface, including robotic tactile sensing, electronic skin, prosthetics, and human-machine interaction. However, it remains challenging to simultaneously achieve high spatial and temporal resolution in developing pressure sensor arrays for tactile sensing with robust function to achieve precise signal recognition. This work presents the development of a flexible high spatiotemporal piezoresistive sensor array (PRSA) by coupling with machine learning algorithms to enhance tactile recognition. The sensor employs cross-striped nanocarbon-polymer composite as an active layer, though screen printing manufacture processes. A miniaturized signal readout circuit and transmission board is developed to achieve high-speed acquisition of distributed pressure signals from the PRSA. Test results indicate that the developed PRSA platform simultaneously possesses the characteristics of high spatial resolution up to 1.5 mm, fast temporal resolution of about 5 ms, and long-term durability with a variation of less than 2%. The PRSA platform also exhibits excellent performance in real-time visualization of multi-point touch, mapping embossed shapes, and tracking motion trajectory. To test the performance of PRSA in recognizing different shapes, we acquired pressure images by pressing the finger-type device coated with PRSA film on different embossed shapes and implementing the T-distributed Stochastic Neighbor Embedding model to visualize the distinction between images of different shapes. Then we adopted a one-layer neural network to quantify the discernibility between images of different shapes. The analysis results show that the PRSA could capture the embossed shapes clearly by one contact with high discernibility up to 98.9%. Collectively, the PRSA as a promising platform demonstrates its promising potential for robotic tactile sensing.

A Versatile Strategy for Concurrent Passive Daytime Radiative Cooling and Sustainable Energy Harvesting

Developing versatile systems that can concurrently achieve energy saving and energy generation is critical to accelerate carbon neutrality. However, challenges on designing highly effective, large scale, and multifunctional photonic film hinder the concurrent combination of passive daytime radiative cooling (PDRC) and utilization of sustainable clean energies. Herein, a versatile scalable photonic film (Ecoflex@h-BN) with washable property and excellent mechanical stability is developed by combining the excellent scattering efficiency of the hexagonal boron nitride (h-BN) nanoplates with the high infrared emissivity and ideal triboelectric negative property of the Ecoflex matrix. Strikingly, sufficiently high solar reflectance (0.92) and ideal emissivity (0.97) endow the Ecoflex@h-BN film with subambient cooling effect of ≈9.5 °C at midday during the continuous outdoor measurements. In addition, the PDRC Ecoflex@h-BN film-based triboelectric nanogenerator (PDRC-TENG) exhibits a maximum peak power density of 0.5 W m-2 . By reasonable structure design, the PDRC-TENG accomplishes effective wind energy harvesting and can successfully drive the electronic device. Meanwhile, an on-skin PDRC-TENG is fabricated to harvest human motion energy and monitor moving states. This research provides a novel design of a multifunctional PDRC photonic film, and offers a versatile strategy to realize concurrent PDRC and sustainable energies harvesting.

Synergizing Machine Learning Algorithm with Triboelectric Nanogenerators for Advanced Self-Powered Sensing Systems

The advancement of the Internet of Things (IoT) has increased the demand for large-scale intelligent sensing systems. The periodic replacement of power sources for ubiquitous sensing systems leads to significant resource waste and environmental pollution. Human staffing costs associated with replacement also increase the economic burden. The triboelectric nanogenerators (TENGs) provide both an energy harvesting scheme and the possibility of self-powered sensing. Based on contact electrification from different materials, TENGs provide a rich material selection to collect complex and diverse data. As the data collected by TENGs become increasingly numerous and complex, different approaches to machine learning (ML) and deep learning (DL) algorithms have been proposed to efficiently process output signals. In this paper, the latest advances in ML algorithms assisting solid-solid TENG and liquid-solid TENG sensors are reviewed based on the sample size and complexity of the data. The pros and cons of various algorithms are analyzed and application scenarios of various TENG sensing systems are presented. The prospects of synergizing hardware (TENG sensors) with software (ML algorithms) in a complex environment and their main challenges for future developments are discussed.

A multimodal magnetoelastic artificial skin for underwater haptic sensing

Future exploitation of marine resources in a sustainable and eco-friendly way requires autonomous underwater robotics with human-like perception. However, the development of such intelligent robots is now impeded by the lack of adequate underwater haptic sensing technology. Inspired by the populational coding strategy of the human tactile system, we harness the giant magnetoelasticity in soft polymer systems as an innovative platform technology to construct a multimodal underwater robotic skin for marine object recognition with intrinsic waterproofness and a simple configuration. The bioinspired magnetoelastic artificial skin enables multiplexed tactile modality in each single taxel and obtains an impressive classification rate of 95% in identifying seven types of marine creatures and marine litter. By introducing another degree of freedom in underwater haptic sensing, this work represents a milestone toward sustainable marine resource exploitation.

Implantable Triboelectric Nanogenerators for Self-Powered Cardiovascular Healthcare

Triboelectric nanogenerators (TENGs) have gained significant traction in recent years in the bioengineering community. With the potential for expansive applications for biomedical use, many individuals and research groups have furthered their studies on the topic, in order to gain an understanding of how TENGs can contribute to healthcare. More specifically, there have been a number of recent studies focusing on implantable triboelectric nanogenerators (I-TENGs) toward self-powered cardiac systems healthcare. In this review, the progression of implantable TENGs for self-powered cardiovascular healthcare, including self-powered cardiac monitoring devices, self-powered therapeutic devices, and power sources for cardiac pacemakers, will be systematically reviewed. Long-term expectations of these implantable TENG devices through their biocompatibility and other utilization strategies will also be discussed.

Electrochromic Sensor Augmented with Machine Learning for Enzyme-Free Analysis of Antioxidants

Our study presents an electrochromic sensor that operates without the need for enzymes or multiple oxidant reagents. This sensor is augmented with machine learning algorithms, enabling the identification, classification, and prediction of six different antioxidants with high accuracy. We utilized polyaniline (PANI), Prussian blue (PB), and copper-Prussian blue analogues (Cu-PBA) at their respective oxidation states as electrochromic materials (ECMs). By designing three readout channels with these materials, we were able to achieve visual detection of antioxidants without relying on traditional "lock and key" specific interactions. Our sensing approach is based on the direct electrochemical reactions between oxidized electrochromic materials (ECMsox) as electron acceptors and various antioxidants, which act as electron donors. This interaction generates unique fingerprint patterns by switching the ECMsox to reduced electrochromic materials (ECMsred), causing their colors to change. Through the application of density functional theory (DFT), we demonstrated the molecular-level basis for the distinct multicolor patterns. Additionally, machine learning algorithms were employed to correlate the optical patterns with RGB data, enabling complex data analysis and the prediction of unknown samples. To demonstrate the practical applications of our design, we successfully used the EC sensor to diagnose antioxidants in serum samples, indicating its potential for the on-site monitoring of antioxidant-related diseases. This advancement holds promise for various applications, including the real-time monitoring of antioxidant levels in biological samples, the early diagnosis of antioxidant-related diseases, and personalized medicine. Furthermore, the success of our electrochromic sensor design highlights the potential for exploring similar strategies in the development of sensors for diverse analytes, showcasing the versatility and adaptability of this approach.

Self-powered TENG probe for scanning surface charge distribution

Triboelectric nanogenerators are remarkable devices that show great potential in harvesting energy from mechanical work and are generally used for sensing purposes. Here we report a novel method for the fabrication of ZnO microspheres and the formation of TENG based on ZnO/PDMS composite. The zinc oxide microspheres with needle decorated structure via thermal oxidation of metallic zinc was grown at 500 °C. The TENG was then fabricated using ZnO/PDMS composite with Au sputtered electrode. While PDMS is a good triboelectric material, its output power density is low. Embedding ZnO micro/nanostructures in PDMS increases the output power of PDMS-based TENG manifolds. ZnO with a high dielectric constant exhibits semiconductor properties as well as piezoelectric properties. This combines with the triboelectric properties of PDMS and gives a significant boost to the TENG performance. This composite structure is used for the fabrication of high output power density TENG using contact separation mode, where the power density of 27Wm-2was achieved. Consequently, a novel device application to detect surface charge density through the fabricated TENG is reported and the subsequent reconstruction of surface charge topology based on the detected surface charge density on large surfaces is presented. This technique may be used for the study of surface charge morphology, electrostatics, triboelectric constants, and various other material properties for characterization and application purposes.

Machine Learning-Enabled Intelligent Gesture Recognition and Communication System Using Printed Strain Sensors

Gesture contains abundant and complicated information in daily life; as a consequence, gesture recognition attracts a wide range of application prospects and academic values as an important way of achieving human-machine interactions (HMIs). Here, we report an intelligent system consisting of a smart glove made by printed CNT-graphene/PDMS strain sensors. The smart glove shows excellent fitness, comfort, and lightness for human hands. Inspired by machine learning strategies, several objects and gestures can be well classified and implemented by a customized artificial neural network. Several data sets of different sign language gestures and object-grabbing gestures were established, and the result shows that the intelligent system can achieve an average accuracy of 97% and up to 99.4% for a number of gesture groups. Moreover, a robot hand is connected to this system, which is able to react to the motion of human hands with certain gestures where simple sign communication is achieved. These features provide a feasible practical application scheme for gesture recognition in HMIs.

Tracing the Flu Symptom Progression via a Smart Face Mask

Respiration and body temperature are largely influenced by the highly contagious influenza virus, which poses persistent global public health challenges. Here, we present a wireless all-in-one sensory face mask (WISE mask) made of ultrasensitive fibrous temperature sensors. The WISE mask shows exceptional thermosensitivity, excellent breathability, and wearing comfort. It offers highly sensitive body temperature monitoring and respiratory detection capabilities. Capitalizing on the advances in the Internet of Things and artificial intelligence, the WISE mask is further demonstrated by customized flexible circuitry, deep learning algorithms, and a user-friendly interface to continuously recognize the abnormalities of both the respiration and body temperature. The WISE mask represents a compelling approach to tracing flu symptom progression in a cost-effective and convenient manner, serving as a powerful solution for personalized health monitoring and point-of-care systems in the face of ongoing influenza-related public health concerns.

All Resistive Pressure-Temperature Bimodal Sensing E-Skin for Object Classification

Electronic skin (E-skin) with multimodal sensing ability demonstrates huge prospects in object classification by intelligent robots. However, realizing the object classification capability of E-skin faces severe challenges in multiple types of output signals. Herein, a hierarchical pressure-temperature bimodal sensing E-skin based on all resistive output signals is developed for accurate object classification, which consists of laser-induced graphene/silicone rubber (LIG/SR) pressure sensing layer and NiO temperature sensing layer. The highly conductive LIG is employed as pressure-sensitive material as well as the interdigital electrode. Benefiting from high conductivity of LIG, pressure perception exhibits an excellent sensitivity of -34.15 kPa-1 . Meanwhile, a high temperature coefficient of resistance of -3.84%°C-1 is obtained in the range of 24-40 °C. More importantly, based on only electrical resistance as the output signal, the bimodal sensing E-skin with negligible crosstalk can simultaneously achieve pressure and temperature perception. Furthermore, a smart glove based on this E-skin enables classifying various objects with different shapes, sizes, and surface temperatures, which achieves over 92% accuracy under assistance of deep learning. Consequently, the hierarchical pressure-temperature bimodal sensing E-skin demonstrates potential application in human-machine interfaces, intelligent robots, and smart prosthetics.

Ti3C2Tx MXene-Based Multifunctional Tactile Sensors for Precisely Detecting and Distinguishing Temperature and Pressure Stimuli

Although skin-like sensors that can simultaneously detect various physical stimuli are of fair importance in cutting-edge human-machine interaction, robotic, and healthcare applications, they still face challenges in facile, scalable, and cost-effective production using conventional active materials. The emerging two-dimensional transition metal carbide, Ti3C2Tx MXene, integrated with favorable thermoelectric properties, metallic-like conductivity, and a hydrophilic surface, is promising for solving these problems. Herein, skin-like multifunctional sensors are designed to precisely detect and distinguish temperature and pressure stimuli without cross-talk by decorating elastic and porous substrates with MXene sheets. Because the combination of the thermoelectric and conductive MXene with the thermally insulating, elastic, and porous substrate integrates efficient Seebeck and piezoresistive effects, the resultant sensor exhibits not only an ultralow detection limit (0.05 K), high signal-to-noise ratio, and excellent cycling stability for temperature detection but also high sensitivity, fast response time, and outstanding durability for pressure detection. Based on the impressive dual-mode sensing properties and independent temperature and pressure detections, a multimode input terminal and an electronic skin are created, exhibiting great potential in robotic and human-machine interaction applications. This work provides a scalable fabrication of multifunctional tactile sensors for precisely detecting and distinguishing temperature and pressure stimuli.

Soft Robotics Enables Neuroprosthetic Hand Design

Development and implementation of neuroprosthetic hands is a multidisciplinary field at the interface between humans and artificial robotic systems, which aims at replacing the sensorimotor function of the upper-limb amputees as their own. Although prosthetic hand devices with myoelectric control can be dated back to more than 70 years ago, their applications with anthropomorphic robotic mechanisms and sensory feedback functions are still at a relatively preliminary and laboratory stage. Nevertheless, a recent series of proof-of-concept studies suggest that soft robotics technology may be promising and useful in alleviating the design complexity of the dexterous mechanism and integration difficulty of multifunctional artificial skins, in particular, in the context of personalized applications. Here, we review the evolution of neuroprosthetic hands with the emerging and cutting-edge soft robotics, covering the soft and anthropomorphic prosthetic hand design and relating bidirectional neural interactions with myoelectric control and sensory feedback. We further discuss future opportunities on revolutionized mechanisms, high-performance soft sensors, and compliant neural-interaction interfaces for the next generation of neuroprosthetic hands.

Ultrastretchable High-Conductivity MXene-Based Organohydrogels for Human Health Monitoring and Machine-Learning-Assisted Recognition

Conductive hydrogels as promising candidates of wearable electronics have attracted considerable interest in health monitoring, multifunctional electronic skins, and human-machine interfaces. However, to simultaneously achieve excellent electrical properties, superior stretchability, and a low detection threshold of conductive hydrogels remains an extreme challenge. Herein, an ultrastretchable high-conductivity MXene-based organohydrogel (M-OH) is developed for human health monitoring and machine-learning-assisted object recognition, which is fabricated based on a Ti₃C₂T_x MXene/lithium salt (LS)/poly(acrylamide) (PAM)/poly(vinyl alcohol) (PVA) hydrogel through a facile immersion strategy in a glycerol/water binary solvent. The fabricated M-OH demonstrates remarkable stretchability (2000%) and high conductivity (4.5 S/m) due to the strong interaction between MXene and the dual-network PVA/PAM hydrogel matrix and the incorporation between MXene and LS, respectively. Meanwhile, M-OH as a wearable sensor enables human health monitoring with high sensitivity and a low detection limit (12 Pa). Furthermore, based on pressure mapping image recognition technology, an 8 × 8 pixelated M-OH-based sensing array can accurately identify different objects with a high accuracy of 97.54% under the assistance of a deep learning neural network (DNN). This work demonstrates excellent comprehensive performances of the ultrastretchable high-conductive M-OH in health monitoring and object recognition, which would further explore extensive potential application prospects in personal healthcare, human-machine interfaces, and artificial intelligence.

Machine Learning-Enhanced Flexible Mechanical Sensing

To realize a hyperconnected smart society with high productivity, advances in flexible sensing technology are highly needed. Nowadays, flexible sensing technology has witnessed improvements in both the hardware performances of sensor devices and the data processing capabilities of the device's software. Significant research efforts have been devoted to improving materials, sensing mechanism, and configurations of flexible sensing systems in a quest to fulfill the requirements of future technology. Meanwhile, advanced data analysis methods are being developed to extract useful information from increasingly complicated data collected by a single sensor or network of sensors. Machine learning (ML) as an important branch of artificial intelligence can efficiently handle such complex data, which can be multi-dimensional and multi-faceted, thus providing a powerful tool for easy interpretation of sensing data. In this review, the fundamental working mechanisms and common types of flexible mechanical sensors are firstly presented. Then how ML-assisted data interpretation improves the applications of flexible mechanical sensors and other closely-related sensors in various areas is elaborated, which includes health monitoring, human-machine interfaces, object/surface recognition, pressure prediction, and human posture/motion identification. Finally, the advantages, challenges, and future perspectives associated with the fusion of flexible mechanical sensing technology and ML algorithms are discussed. These will give significant insights to enable the advancement of next-generation artificial flexible mechanical sensing.

In Situ Polymerized MXene/Polypyrrole/Hydroxyethyl Cellulose-Based Flexible Strain Sensor Enabled by Machine Learning for Handwriting Recognition

Flexible strain sensors have significant progress in the fields of human-computer interaction, medical monitoring, and handwriting recognition, but they also face many challenges such as the capture of weak signals, comprehensive acquisition of the information, and accurate recognition. Flexible strain sensors can sense externally applied deformations, accurately measure human motion and physiological signals, and record signal characteristics of handwritten text. Herein, we prepare a sandwich-structured flexible strain sensor based on an MXene/polypyrrole/hydroxyethyl cellulose (MXene/PPy/HEC) conductive material and a PDMS flexible substrate. The sensor features a wide linear strain detection range (0-94%), high sensitivity (gauge factor 357.5), reliable repeatability (>1300 cycles), ultrafast response-recovery time (300 ms), and other excellent sensing properties. The MXene/PPy/HEC sensor can detect human physiological activities, exhibiting excellent performance in measuring external strain changes and real-time motion detection. In addition, the signals of English words, Arabic numerals, and Chinese characters handwritten by volunteers measured by the MXene/PPy/HEC sensor have unique characteristics. Through machine learning technology, different handwritten characters are successfully identified, and the recognition accuracy is higher than 96%. The results show that the MXene/PPy/HEC sensor has a significant impact in the fields of human motion detection, medical and health monitoring, and handwriting recognition.

Intelligent wearable devices based on nanomaterials and nanostructures for healthcare

Emerging classes of flexible electronic sensors as alternatives to conventional rigid sensors offer a powerful set of capabilities for detecting and quantifying physiological and physical signals from human skin in personal healthcare. Unfortunately, the practical applications and commercialization of flexible sensors are generally limited by certain unsatisfactory aspects of their performance, such as biocompatibility, low sensing range, power supply, or single sensory function. This review intends to provide up-to-date literature on wearable devices for smart healthcare. A systematic review is provided, from sensors based on nanomaterials and nanostructures, algorithms, to multifunctional integrated devices with stretchability, self-powered performance, and biocompatibility. Typical electromechanical sensors are investigated with a specific focus on the strategies for constructing high-performance sensors based on nanomaterials and nanostructures. Then, the review emphasizes the importance of tailoring the fabrication techniques in order to improve stretchability, biocompatibility, and self-powered performance. The construction of wearable devices with high integration, high performance, and multi-functionalization for multiparameter healthcare is discussed in depth. Integrating wearable devices with appropriate machine learning algorithms is summarized. After interpretation of the algorithms, intelligent predictions are produced to give instructions or predictions for smart implementations. It is desired that this review will offer guidance for future excellence in flexible wearable sensing technologies and provide insight into commercial wearable sensors.

Improving Relative Permittivity and Suppressing Dielectric Loss of Triboelectric Layers for High-Performance Wearable Electricity Generation

High relative permittivity and low dielectric loss are two desired parameters of a triboelectric layer to enhance its mechanical-to-electrical energy conversion efficiency in a triboelectric nanogenerator (TENG). However, the elevated permittivity of the triboelectric layer is always accompanied by increasing dielectric loss, limiting further improvement or even reducing the electrical output. Herein, we report a method for improving the relative permittivity and suppressing the dielectric loss of the triboelectric layer via nanoscale design at the particle-polymer interface. When incorporated with 2 wt % Ag@C, the triboelectric-layer-enhanced TENG (TLE-TENG) presents a 2.6-fold increment in relative permittivity and a 302% current enhancement. An instantaneous peak power density of 1.22 W m^-2, an excellent pressure sensitivity of 90.95 V kPa^-1, and an optimized sheet resistance (∼0.14 Ω/sq) are attributes of this greatly enhanced device. Such improvements bode well for the implementation of these enhancing strategies to help position TLE-TENGs as pervasive and sustainable power sources and active self-powered sensors in the era of the Internet of Things.

Airline Point-of-Care System on Seat Belt for Hybrid Physiological Signal Monitoring

With a focus on disease prevention and health promotion, a reactive and disease-centric healthcare system is revolutionized to a point-of-care model by the application of wearable devices. The convenience and low cost made it possible for long-term monitoring of health problems in long-distance traveling such as flights. While most of the existing health monitoring systems on aircrafts are limited for pilots, point-of-care systems provide choices for passengers to enjoy healthcare at the same level. Here in this paper, an airline point-of-care system containing hybrid electrocardiogram (ECG), breathing, and motion signals detection is proposed. At the same time, we propose the diagnosis of sleep apnea-hypopnea syndrome (SAHS) on flights as an application of this system to satisfy the inevitable demands for sleeping on long-haul flights. The hardware design includes ECG electrodes, flexible piezoelectric belts, and a control box, which enables the system to detect the original data of ECG, breathing, and motion signals. By processing these data with interval extraction-based feature selection method, the signals would be characterized and then provided for the long short-term memory recurrent neural network (LSTM-RNN) to classify the SAHS. Compared with other machine learning methods, our model shows high accuracy up to 84-85% with the lowest overfit problem, which proves its potential application in other related fields.

Advances in Photoplethysmography for Personalized Cardiovascular Monitoring

Photoplethysmography (PPG) is garnering substantial interest due to low cost, noninvasiveness, and its potential for diagnosing cardiovascular diseases, such as cardiomyopathy, heart failure, and arrhythmia. The signals obtained through PPG can yield information based on simple analyses, such as heart rate. In contrast, when accompanied by the complex analysis of sophisticated signals, valuable information, such as blood pressure, sympathetic nervous system activity, and heart rate variability, can be obtained. For a complex analysis, a better understanding of the sources of noise, which create limitations in the application of PPG, is needed to get reliable information to assess cardiovascular health. Therefore, this Special Issue handles literature about noises and how they affect the waveform of the PPG caused by individual variations (e.g., skin tone, obesity, age, and gender), physiology (e.g., respiration, venous pulsation, body site of measurement, and body temperature), and external factors (e.g., motion artifact, ambient light, and applied pressure to the skin). It also covers the issues that still need to be considered in each situation.

Self-Powered Smart Gloves Based on Triboelectric Nanogenerators

The hands are used in all facets of daily life, from simple tasks such as grasping and holding to complex tasks such as communication and using technology. Finding a way to not only monitor hand movements and gestures but also to integrate that data with technology is thus a worthwhile task. Gesture recognition is particularly important for those who rely on sign language to communicate, but the limitations of current vision-based and sensor-based methods, including lack of portability, bulkiness, low sensitivity, highly expensive, and need for external power sources, among many others, make them impractical for daily use. To resolve these issues, smart gloves can be created using a triboelectric nanogenerator (TENG), a self-powered technology that functions based on the triboelectric effect and electrostatic induction and is also cheap to manufacture, small in size, lightweight, and highly flexible in terms of materials and design. In this review, an overview of the existing self-powered smart gloves will be provided based on TENGs, both for gesture recognition and human-machine interface, concluding with a discussion on the future outlook of these devices.

Recent Progress in Flexible Pressure Sensor Arrays

Flexible pressure sensors that can maintain their pressure sensing ability with arbitrary deformation play an essential role in a wide range of applications, such as aerospace, prosthetics, robotics, healthcare, human–machine interfaces, and electronic skin. Flexible pressure sensors with diverse conversion principles and structural designs have been extensively studied. At present, with the development of 5G and the Internet of Things, there is a huge demand for flexible pressure sensor arrays with high resolution and sensitivity. Herein, we present a brief description of the present flexible pressure sensor arrays with different transduction mechanisms from design to fabrication. Next, we discuss the latest progress of flexible pressure sensor arrays for applications in human–machine interfaces, healthcare, and aerospace. These arrays can monitor the spatial pressure and map the trajectory with high resolution and rapid response beyond human perception. Finally, the outlook of the future and the existing problems of pressure sensor arrays are presented.