Anterior talofibular ligament (ATFL) sprain is one of the most prevalent sports-related injuries, so proper evaluation of ligament sprains is critical for treatment options. However, existing tests suffer from a lack of standardized quantitative evaluation criteria, interindividual variability, incompatible materials, or risks of infection. Although advanced medical diagnostic methods already have been using noninvasive, portable, and wearable diagnostic electronics, these devices have insufficient adhesion to accurately respond to internal body injuries. Therefore, we propose a high-adhesive hydrogel-based strain sensor made from gelatin, cellulose nanofiber (CNF), and cross-linked poly(acrylic acid) grafted with N-hydrosuccinimide ester. The adhesive strain sensor, with excellent conformability and stretchability, firmly adheres to the skin, making it suitable for accurately evaluating the severity of anterior talofibular ligament sprain. Its strong adhesive (up to 192 kPa) can adapt to the surface characterization of ankles. The high-adhesive hydrogel-based strain sensor has a high tensile strength (680%) and achieves a high gauge factor (GF) of 8.29. Simultaneously, it also presents a 40 μm ultralow detection limit. Additionally, after a deep learning model was integrated to improve sensing accuracy, the system achieved a diagnostic accuracy of 95%, significantly surpassing the magnetic resonance imaging (MRI) gold standard of 81.1%.

Tactile sensing plays a crucial role as an approach of human-machine interaction in the digital era. In this work, inspired by animal fur, we propose a triboelectric tactile sensing array in order to achieve sensitive awareness of gentle palm interactions. Which is of great significance for enhancing the functionality of companion and pet robots. The tactile sensor consists of a cluster of single-electrode structured triboelectric sensing units based on PTFE film coated conductive yarns. The dynamic signals of the tactile sensing array can accurately perceive 3Ddepth interaction information of touch such as direction, speed and gesture. It allows for more comprehensive and accurate pattern recognition in human-machine interaction scenarios. Furthermore, a deep learning model is used to assist the recognition of complex signals from the tactile sensing array under various interactions. Among a dataset of 17 palmar interactions, a recognition accuracy up to 96.8 % was achieved. As a demonstration, we finally construct an intelligent human-machine interface based on this tactile sensing array. It is integrated onto a shape-shifting robot working as a piece of fur of an intelligent pet robot. With the help of this fur, the robot can accurately recognize the interaction from the tester and act accordingly.

Hand palpation is a widely used method in clinical practice to estimate the hardness of soft tissues, but the results of this method are quite subjective. Therefore, it is necessary to develop some objective tools to assess the hardness of soft tissues. We present a wearable ultrasound sensor for measuring the hardness of soft tissues. This wearable sensor utilizes an ultrasound transducer to indent the tissue, applying pressure to the transducer by inflating an airbag located above the transducer. By analyzing the ultrasound signals, we can distinguish the deformation differences of soft tissues under the same pressure, which allows us to assess the hardness of the soft tissues. The experimental results indicate that our wearable ultrasound sensor can distinguish the hardness differences of the biceps during relaxation and exertion. In the preliminary experiments conducted on a human arm model, we identified the differences in hardness before and after embedding a foreign object into the arm model, which demonstrates the potential of this sensor in detecting internal soft tissue lesions.

Manual palpation serves as a conventional clinical method for assessing soft tissue stiffness; however, its results are susceptible to subjective factors and exhibit limited reliability. To achieve objective evaluation of pathological tissue stiffness, this study utilizes ultrasonic transducers to measure the time-of-flight (ToF) difference in ultrasound signals in silicone samples and ex vivo animal tissues under specific pressure gradients. A correlation model between the ToF difference and tissue stiffness was established, thereby enabling the detection of tissue stiffness. Based on this methodology, a wearable sensing system incorporating ultrasonic transducers was developed. The system applies fixed gradient pressure to human tissues via a pneumatic control unit and detects the corresponding ToF difference, allowing real-time monitoring of stiffness variations in the biceps brachii and thigh during relaxation and contraction, in the forearm during gripping and release actions, as well as in simulated lesions. This study provides a quantitative technological framework for wearable tissue stiffness monitoring, and its objective measurement characteristics offer support for clinical diagnostic decision-making.

Power supply is playing an increasingly important role in the rapidly developing era of the Internet of Things. Achieving a sustainable and clean power supply for electronic devices is an urgent and challenging task. In this study, we present a heat-triggered triboelectric nanogenerator (TENG) and develop a self-powered fire alarm system to achieve an early warning without an external power supply. A TENG comprises a gear system that can utilize the elastic potential energy of a spring. A wax block was used as a heat trigger. When melted at high temperatures, the TENG will be triggered to work and generate considerable electric energy. Within a single operation cycle of approximately 6 s, a 22 μ F capacitor can be charged up to 3.7 V. Such electrical energy is sufficient to drive a wireless transmission module through an automatic switching circuit. Overall, this study provides a feasible approach for a self-powered wireless warning system in power-shortage areas.

A unique oscillating wind-driven triboelectric nanogenerator(OWTENG)based on the sphere's vortex-induced vibration(VIV)behavior is proposed in this study,which can harvest wind energy across a multitude of horizontal directions.With the Euler-Lagrange method,the coupled governing equations of the OWTENG are estab-lished and subsequently validated by experimental tests.The vibrational properties and output performance of the OWTENG for varying wind speeds are analyzed,demonstrat-ing its effectiveness in capturing wind energy across a broad range of wind speeds(from 2.20 m/s to 8.84 m/s),and the OWTENG achieves its peak output power of 106.3 μW at a wind speed of 5.72 m/s.Furthermore,the OWTENG maintains a steady output power across various wind directions within the speed range of 2.20 m/s to 7.63 m/s.Nevertheless,when the wind speed exceeds 7.63 m/s,the vibrational characteristics of the sphere shift based on the wind direction,leading to fluctuations in the OWTENG's output power.This research presents an innovative approach for designing vibrational triboelectric nanogenerators,offering valuable insights into harvesting wind energy from diverse directions and speeds.

We present a skin-like tactile sensor array incorporating single-electrode structured triboelectric nanogenerators. The triboelectric sensor exhibits remarkable sensitivity and adaptability to dynamic stimuli, demonstrating advantages in potential difference during both contact and separation phases. The silicone rubber, shaped with a 320- grit sandpaper, yielded the highest output voltage, reaching nearly 15 V under a 1.5 Hz frequency contact. In contrast to traditional tactile sensors that measure parameters such as pressure, shape/profile, and texture, we leverage the sensitivity of the triboelectric nanogenerator sensor to dynamic stimuli. By establishing dual criteria based on array data and triboelectric sensor units, we achieve the recognition of various gestures, directions, and speeds.

Tactile sensors play a critical role in robotic intelligence and human-machine interaction. In this manuscript, we propose a hybrid tactile sensor by integrating a triboelectric sensing unit and a capacitive sensing unit based on porous PDMS. The triboelectric sensing unit is sensitive to the surface material and texture of the grasped objects, while the capacitive sensing unit responds to the object’s hardness. By combining signals from the two sensing units, tactile object recognition can be achieved among not only different objects but also the same object in different states. In addition, both the triboelectric layer and the capacitor dielectric layer were fabricated through the same manufacturing process. Furthermore, deep learning was employed to assist the tactile sensor in accurate object recognition. As a demonstration, the identification of 12 samples was implemented using this hybrid tactile sensor, and an recognition accuracy of 98.46% was achieved. Overall, the proposed hybrid tactile sensor has shown great potential in robotic perception and tactile intelligence.

Lung cancer, the second most common type of cancer worldwide, is primarily treated through surgery. During the operations, preserving pulmonary arteries and veins is a crucial problem. In recent years, 3D visualization techniques like virtual reality and 3D printing have been increasingly used in clinical practice for lung cancer surgery planning. Under the success of these techniques, automatic segmentation of pulmonary arteries and veins plays a key role. Particularly, the state-of-art approaches rely on two techniques, i.e. the deep neural networks (DNNs) or the traditional machine learning (ML) method, and both techniques have respective shortages. Basically, the ML-based methods generally demonstrate a limited performance, while the DNN-based methods lack sufficient annotation for accurate segmentation. In response to such a dilemma, this paper proposes a fusion method to combine the DNN-based and ML-based methods to segment pulmonary arteries and veins for lung cancer surgery planning. Particularly, the anatomy prior mask corresponding to pulmonary arteries and veins are identified using the marching cubes algorithm and Attention U-Net. Subsequently, an enhanced attention U-Net, is used to integrate the original CT scans with the anatomy prior mask to generate the refined segmentation results. Following this, an anatomy structure enhancement module is used to refine the segmentation further by refining disconnected vessel segments and correcting misclassified vessels based on anatomy prior masks. We experimented the proposed approach on a private dataset of 95 CT scans collected from patients after surgery, and then annotated by lung cancer experts. The results demonstrate that our approach outperforms the existing methods with an improvement of 5.1

A novel energy harvester based on vortex-induced vibration of the sphere and piezoelectric effect is proposed to efficiently gather wind energy from all horizontal directions. An elastically-supported foam sphere is vertically arranged and considerably oscillates in the cross-flow direction when the wind speed is in the lock-in region. A piezoelectric beam is attached to the supported sphere by a spring and deforms periodically around the original buckling state, thereby converting kinetic energy into electrical energy. Experimental studies are performed to assess the power output of the harvester when exposed to wind flows with varying directions and speeds. The omnidirectional wind flow is divided to 12 orientations with the interval angle of 30°, and the wind speeds range from 1.17 m s −1 to 7.87 m s −1 . The testing findings indicate that the harvester has excellent consistency in both lock-in region and average power for various wind azimuths. When the wind direction is changed from 0° to 360°, the peak average power changes from 179.8 μ W to 247.5 μ W, and the wind speed region where the sphere undergoes vibration changes from 2.58 m s −1 ∼ 7.05 m s −1 to 2.58 m s −1 ∼ 7.87 m s −1 . Following that, the effects of the length of the supporting spring and the diameter of the sphere on the output average power and lock-in region are investigated experimentally. Finally, a demonstration of powering a wireless sensing node is performed to show applications of the designed energy harvester in windy conditions.

Hypertrophic obstructive cardiomyopathy (HOCM) is a leading cause of sudden cardiac death in young people. Septal myectomy surgery has been recognized as the gold standard for non-pharmacological therapy of HOCM, in which aortic and mitral valves are critical regions for surgical planning. Currently, manual segmentation of aortic and mitral valves is widely performed in clinical practice to construct 3D models used for HOCM surgical planning. Such a process, however, is time-consuming and costly. In this paper, we integrate anatomical prior knowledge into deep learning for automatic segmentation of aortic and mitral valves. In particular, a two-stage method is proposed: we first obtain the region of interest (RoI) from a CT image, where heart segmentation is then performed. The spatial relationship between heart substructures is utilized to identify a valve region that contains the aortic and mitral valves. Unlike typical two-stage methods, we feed the refined segmentation of the left ventricle, left atrium, and aorta as additional input for the valve segmentation. By incorporating this anatomical prior knowledge, deep neural networks (DNNs) can leverage the surrounding anatomical structures to improve valve segmentation. We collected a dataset of 27 CT images from patients with a medical history of septal myectomy surgery. Experimental results show that our method achieves an average Dice score of 71.2% and an improvement of 4.2% over existing methods. Our dataset and code will be released to the public Dataset.

An anterior cruciate ligament (ACL) tear is a common musculoskeletal injury with a high incidence. Traditional diagnosis employs magnetic response imaging (MRI), physical testing, or other clinical examination, which relies on complex and expensive medical instruments, or individual doctoral experience. Herein, we propose a wearable displacement sensing system based on a grating-structured triboelectric stretch sensor to diagnose the ACL injuries. The stretch sensor exhibits a high resolution (0.2 mm) and outstanding robustness (over 1,000,000 continuous operation cycles). This system is employed in clinical trial to diagnose ACL injuries. It measures the displacement difference between the affected leg and the healthy leg during Lachman test. And when such a difference is greater than 3 mm, the ACL is considered to be at risk for injury or tear. Compared with the gold standard of arthroscopy, the consistency rate of this wearable diagnostic system reached about 85.7%, which is higher than that of the Kneelax3 arthrometer (78.6%) with a large volume. This shows that the wearable system possesses the feasibility to supplement and improve existing arthrometers for facile diagnosing ACL injuries. It may take a promising step for wearable healthcare.

Recently, hydrogels have attracted great attention because of their unique properties, including stretchability, self-adhesion, transparency, and biocompatibility. They can transmit electrical signals for potential applications in flexible electronics, human-machine interfaces, sensors, actuators, et al. MXene, a newly emerged two-dimensional (2D) nanomaterial, is an ideal candidate for wearable sensors, benefitting from its surface's negatively charged hydrophilic nature, biocompatibility, high specific surface area, facile functionalization, and high metallic conductivity. However, stability has been a limiting factor for MXene-based applications, and fabricating MXene into hydrogels has been proven to significantly improve their stability. The unique and complex gel structure and gelation mechanism of MXene hydrogels require intensive research and engineering at nanoscale. Although the application of MXene-based composites in sensors has been widely studied, the preparation methods and applications of MXene-based hydrogels in wearable electronics is relatively rare. Thus, in order to facilitate the effective evolution of MXene hydrogel sensors, the design strategies, preparation methods, and applications of MXene hydrogels for flexible and wearable electronics are comprehensively discussed and summarized in this work.

Harvesting biomechanical energy for electricity as well as physiological monitoring is a major development trend for wearable devices. In this article, we report a wearable triboelectric nanogenerator (TENG) with a ground-coupled electrode. It has a considerable output performance for harvesting human biomechanical energy and can also be used as a human motion sensor. The reference electrode of this device achieves a lower potential by coupling with the ground to form a coupling capacitor. Such a design can significantly improve the TENG's outputs. A maximum output voltage up to 946 V and a short-circuit current of 36.3 μA are achieved. The quantity of the charge that transfers during one step of an adult walking reaches 419.6 nC, while it is only 100.8 nC for the separate single-electrode-structured device. In addition, using the human body as a natural conductor to connect the reference electrode allows the device to drive the shoelaces with integrated LEDs. Finally, the wearable TENG is able to perform motion monitoring and sensing, such as human gait recognition, step count and movement speed calculation. These show great application prospects of the presented TENG device in wearable electronics.

In recent years, a proliferation of wearable applications has been observed, fueled by the rapid development of sensor and integrated circuit manufacturing technology. This surge extends beyond a fleeting trend, signifying a substantial shift in our interaction with technology and our approach to data collection in daily life. Accompanying this shift, a key research focus has emerged on the integration of artificial intelligence and machine learning methods, aiming to augment and broaden the wearable systems’ applications. Enabled by these methods, machine learning-assisted wearable intelligent sensing systems are not merely passive data collectors. Active monitoring and tracking of human activities and vital signs are conducted, unlocking considerable potential in human-computer interactions, digital health, and clinical diagnosis areas. We have organized and summarized the recent advancements in wearable sensor devices, machine learning algorithms, and their collaborative roles in wearable sensing applications. The evolution of wearable devices is traced from simple fitness trackers to sophisticated devices capable of monitoring a wide spectrum of biological and physical parameters. Various types of wearable devices and the diverse sensors they incorporate are then classified. These sensors, empowered with advanced technologies, are designed to monitor an extensive array of human activities and vital signs, including heart rate, blood pressure, body temperature, and physical activities. Furthermore, a thorough analysis is provided on the different categories of wearable devices, encompassing but not limited to smartwatches, fitness bands, smart clothing, and implantable devices. Each category’s unique features and applications have been evolved, driven by both technological advancements and user needs. We turn our attention to the crucial function of machine learning within the framework of wearable sensing systems. Renowned for their capabilities to adapt from data and foresee results, machine learning algorithms are utilized to sift through data collected by wearable technology, unlocking valuable insights in the process. This portion of the review provides an in-depth examination of different machine learning paradigms: Supervised, unsupervised, reinforcement, and deep learning, and elucidates their tailored applications in wearable sensing systems for identifying activities, monitoring health, and detecting anomalies. Additionally, the challenges faced by machine learning-assisted wearable sensing systems are addressed. These challenges span data privacy and security, energy efficiency, and the need for robust and reliable algorithms. Emphasis is placed on areas requiring improvement and further research, including enhancing the accuracy and reliability of sensors and developing energy-efficient algorithms. In conclusion, potential solutions and future directions are proposed for the development of machine learning-assisted wearable sensing systems, with an emphasis on the need for continued innovation and research in this field.

Aiming to improve the energy harvesting efficiency under low wind speed, we propose a dual auxiliary beam galloping triboelectric nanogenerator (GTENG) in this work. The structural design of a single main beam and a pair of auxiliary beams enables the device to work under a higher vibration frequency when triggered by wind. A stable and improved working frequency of about 4.6 Hz was observed at various wind speeds. The device started to vibrate at a wind speed of 1.7 m/s and generated an output voltage of about 100 V. The outputs of this GTENG approach to saturation at a wind speed of around 5 m/s. The output voltage and short-circuit current reached 260 V and 20 μA, respectively. A maximum power of about 1 mW was obtained under a wind speed of 5.7 m/s with a load of 33 MΩ. Moreover, the effectivity and long-term stability of the device were demonstrated under low wind speeds. A digital watch is powered for 45 s after charging a 47 μF capacitor for 120 s at a wind speed of 3.1 m/s.

In this work, MXene films incorporating cellulose nanofibers (CNFs) with a spider-web-like structure were fabricated using a facile vacuum-assisted filtration method. The CNFs significantly improved the flexibility and stability of the MXene membranes. The resulting composites functioned well as electrodes and friction layers in triboelectric nanogenerators (TENGs) when paired with either polytetrafluoroethylene (PTFE) as an electropositive material or nylon as an electronegative material. A membrane containing 20 wt % CNFs in conjunction with PTFE was extremely effective during the prolonged operation of a TENGs, generating an output voltage in excess of 1120 V at a frequency of 3.5 Hz. The surface charge density of this device was as high as 100 μC m-2. When paired with nylon, the MXene/CNF film produced a surface charge density of over 60 μC m-2. The microstructures on the rough surface of these membranes, together with the presence of -F and other polar terminations on the MXene, are responsible for the high performance of the nanocomposite. This work demonstrates that MXenes are not necessarily equivalent to PTFE within the triboelectric series and suggests that the MXene-based friction layer could greatly enhance the performance of TENGs.

We propose a novel vortex-induced vibration (VIV) based triboelectric nanogenerator (TENG) that makes it possible for efficient energy harvesting from wind at low speed. A theoretical model for VIV-TENG is constructed to investigate the vibration response and output voltage varying with wind speed, which is compared with and validated by experiments. Results show that there is a lock-in region where the vibration amplitude and output voltage are large. When the wind speed is 2.78 m/s, the average power is achieved to be 392.72 mu W and the average power density is 96.79 mW/m2, which are much higher than those reported in previous studies. In order to broaden the lock-in region, we further design a new model of tandem vortex-induced vibration triboelectric nanogenerators. The resonance wind speed region is remarkably widened due to interactions between two cylinders subjected to VIV. The demonstration of continuously powering wireless sensors is performed to show the practicability of the designed VIV-TENG. Overall, the present study enables applications of TENG for realizing self-powered wireless sensing under low wind speed environments.

Tactile sensors can enable a robotic manipulator to identify the object in contact. However, due to the dynamics and diversity of target objects, as well as the complexity of real environment, accurate recognition of objects by existing tactile sensors has been very challenging. This paper proposes a hybrid tactile sensor that integrates a triboelectric active sensing unit with an electromagnetic inductance transducer. The triboelectric signal relates strongly to the specific charge condition of the surface material of a target object, while the inductive signal manifests the electromagnetic characteristics at a certain depth inside the object. With the help of machine learning, the triboelectric signals and inductive signals can be used for object identification. We demonstrate a robotic gripper with random operation settings can recognize eight different fruits with an accuracy as high as 98.75%. Furthermore, the hybrid sensor can recognize objects packaged in different ways. The recognition ac-curacy of four different fruits in three different packages can reach 95.93%. This study demonstrates the po-tential of hybrid tactile sensor to improve the artificial intelligence of robots, in particular their ability to distinguish objects in complex settings and sorting them effectively.

We present an optimized flutter-driven triboelectric nanogenerator (TENG) for wind energy harvesting. The vibration and power generation characteristics of this TENG are investigated in detail, and a low cut-in wind speed of 3.4 m/s is achieved. It is found that the air speed, the thickness and length of the membrane, and the distance between the electrode plates mainly determine the PTFE membrane’s vibration behavior and the performance of TENG. With the optimized value of the thickness and length of the membrane and the distance of the electrode plates, the peak open-circuit voltage and output power of TENG reach 297 V and 0.46 mW at a wind speed of 10 m/s. The energy generated by TENG can directly light up dozens of LEDs and keep a digital watch running continuously by charging a capacitor of 100 μF at a wind speed of 8 m/s.