Search

The general silhouette‐based gait recognition methods usually rely on binary human silhouette, which is easily affected by external factors, making it unsuitable for situations while wearing heavy clothes or carrying objects, etc. In this study, a new skeleton‐based gait recognition model is proposed. The model first extracts the spatial and temporal features of gait using the space and time relationship between body joints, and second, it eliminates redundant features by decomposing the feature map, to achieve a better recognition accuracy in the presence of external factors. Through abundant experiments on two common datasets, CASIA‐B and OUMVLP‐Pose, the proposed model has been proved to have higher recognition accuracy and remarkable robustness.

Touch orientation detection is important for piezoelectric touch panels to stabilize force-voltage responsivities. Current touch orientation estimation techniques utilize machine learning algorithms for orientation classification. However, environmental noise could weaken the data quality which will result in a lowered detection accuracy. To address this issue, in this article, we present a noise robustness technique, in which different levels of noise data are injected into the training data. The performance of “dirty” data trained model exhibits a good performance (average mean absolute error (MAE) of 7.8 degrees) among signal-to-noise ratio (SNR) from 3 dB to 40 dB, indicating that an improved user experience can be obtained.

Gait analysis is important in the field of healthcare, due to its close relationship to chronic diseases. With the development of the Internet of Health Things (IoHT), long-term gait monitoring and corresponding analysis can be performed remotely, reducing a patient’s time and traffic cost, while providing doctors more valuable gait information. In this paper, we present a piezoelectric insole gait monitoring system and its use in an IoHT architecture. Through the experimental results, the high detection sensitivity of 3.56 kPa and responsivity of 2.35 mV/kPa are achieved. Besides, computer software is developed with Qt platform, and the cloud database is also established. The plantar pressure distribution and center of pressure (CoP) trajectory are provided for further analysis, thereby satisfying the need for analyzing various diseases. Furthermore, the assembled system can continuously work for 16 hours, indicating its successful utilization when long-term gait monitoring is required. The presented work provides a feasible means for real-time, long-term, and accurate gait monitoring, prompting the development of gait analysis in the IoHT.

In order to quantify the manipulation process of acupuncture, in this article, a piezoelectric glove based wearable stress sensing system is presented. Served as the sensitive element with small volume and high tensile resistance, PVDF greatly meet the need of quantitative analysis. Through piezoelectric force sensing glove, the system is capable of detecting both perpendicular stress as well as shear stress. Besides, key parameters including peak stress at needle are detected and extracted, potentially allowing for a higher learning efficiency hence advancing the development of acupuncture.

Acupuncture is one of the most significant therapies of Chinese traditional medical science, and it is now globally utilized for diverse treatments, e.g., pain management. Traditionally, there is no quantification means for storing masters’ skills and examining trainee’s learning effect, hence, strongly limiting the development of acupuncture. For addressing this issue, in this article, a piezoelectric glove based wearable stress sensing technique is presented. Experimental results showcase that through the piezoelectric force-sensing glove, key parameters (e.g., peak stress at needle) during performing acupuncture are detected and extracted. Simultaneously, the Leap Motion, as a finger gesture monitoring technique, is utilized to capture the finger motion during acupuncture, allowing trainees to have a straightforward understanding of how gesture details are correlated to the stress information. The technique presented in this article potentially improves the learning efficiency of trainees and therefore advancing the progress of acupuncture.

In the era of information technology, secure access to electronic systems is important for protecting private information. Among various authentication techniques, keystroke authentication methods based on users’ touch behavior have received increasing attention. In this paper, we present a technique for obtaining high multi-user authentication accuracy by utilizing users’ touch time and force information, which were obtained from a piezoelectric touch panel. After feeding a support vector machine (SVM) with users’ touch features, a high classification accuracy of 97% is achieved, validating the feasibility of the proposed technique for achieving highly secure multi-user authentication, hence advancing the development of security techniques.

Artificial smart skins capable of interacting with people and sensing environmental stimuli have become a research topic in humanoid robotic applications. However, previously reported architectures suffer difficulties in achieving concurrent multi-dimensional sensing in a simple structure with low system costs. To address this issue, in this paper, an artificial smart skin constructed with polyimide/copper/polyvinylidene fluoride (PVDF) is presented for detecting 2D-position, proximity, dynamic force, and humidity via a smart combination of piezoelectric- and capacitive-effects. Moreover, a follow-up filtering algorithm is proposed to separate and calculate the stimuli associated with capacitance changes (position, proximity, and humidity). The proposed system achieves overall force and capacitive sensitivities of 0.05 N and 7 fF, a humidity responsivity at 0.22%/RH% over a relative humidity range of 10%–90% RH, a spatial resolution of 0.29 mm. This simple-structured device supports multiple functions with low system costs, thus advancing the field of robotics smart skins.

Concurrent high force detection accuracy and extended battery lifetime are strongly expected in wearable gait monitoring systems, which are essential for many Internet of Health Things (IoHT) applications. In this article, a piezoelectric insole device and a rectifying circuitry-based technique is presented to achieve these two ultimate goals. Here, walking induced positive and negative charges are separated for plantar stress detecting and energy harvesting, respectively; these two functions are realized concurrently. Experimental results demonstrate that first, the high detection sensitivity of 55 mN and responsivity of 231 mV/N are achieved, satisfying the need for diagnosing various diseases; second, the electric energy of 9.6 pJ (2.67×10-15 W·hr) is stored during a walking cycle, showing the potential of extending the battery lifetime. The developed technique enhances the development of gait monitoring in IoHT.

Rapid development of high-throughput technologies has permitted the identification of an increasing number of disease-associated genes (DAGs), which are important for understanding disease initiation and developing precision therapeutics. However, DAGs often contain large amounts of redundant or false positive information, leading to difficulties in quantifying and prioritizing potential relationships between these DAGs and human diseases. In this study, a network-oriented gene entropy approach (NOGEA) is proposed for accurately inferring master genes that contribute to specific diseases by quantitatively calculating their perturbation abilities on directed disease-specific gene networks. In addition, we confirmed that the master genes identified by NOGEA have a high reliability for predicting disease-specific initiation events and progression risk. Master genes may also be used to extract the underlying information of different diseases, thus revealing mechanisms of disease comorbidity. More importantly, approved therapeutic targets are topologically localized in a small neighborhood of master genes in the interactome network, which provides a new way for predicting drug-disease associations. Through this method, 11 old drugs were newly identified and predicted to be effective for treating pancreatic cancer and then validated by in vitro experiments. Collectively, the NOGEA was useful for identifying master genes that control disease initiation and co-occurrence, thus providing a valuable strategy for drug efficacy screening and repositioning. NOGEA codes are publicly available at https://github.com/guozihuaa/NOGEA.

Insole gait analysis has become a research focus in recent years, owing to its potential in sports training improvement, and real-time and long-term health status monitoring. Among diverse insole gait analysis technologies, piezoelectric-based techniques receive growing attention, due to its nature in converting force to electric signals, potentially ensuring long-battery lifetime compared to its counterparts. However, current literature mainly reports on the detection of 6–8 main plantar positions for motion interpretation, lacking in providing enough information for illness diagnosis, e.g., hip disease. In this article, smart integration of piezoelectric films with different polarization orientations and 32 electrodes is presented, for supporting both normal and shear stress detection with high three-dimensional resolution. High sensitivities at 56 and 173 mN are demonstrated experimentally for normal and shear stress sensing. With our comprehensive algorithms, accurate and real-time gait monitoring for the majority parts of the plantar area is obtained. This paper can support diverse utilization purposes, advancing the area of insole gait analysis.

With the rapid development of electric and information technologies, security issues are becoming a pressing concern. This has given rise to a plethora of authentication techniques, leading among them being the fixed-password method due to its simple mechanism and independence from any specific hardware in the modern smartphone. While fixed-password methods enjoy benefits such as ease of use, potential security issues associated with password leakage cannot be ignored. This article, investigates an alternative strategy based on user keystrokes. Here, the user touch times and force features are extracted from a piezoelectric force touch panel which is an integral part of the hardware. Three broadly adopted machine learning classifiers are used for the collected data set, finally achieving an Equal Error Rate (EER) of 0.720%.

Piezoelectric force touch panels have attracted a significant amount of interest in the field of human–machine interactivity, owing to its merits, such as a low power consumption, high force detection sensitivity and simple panel structure. However, the unstable force–voltage responsivity introduced by various touch orientations limits its successful use in interactive displays. To address this issue, in this article, we present a machine learning-based technique, in which a user finger touch-induced capacitive pattern is used to train an artificial neural network (ANN). Using this technique, a high touch angle classification accuracy (95.7%) and a high average force detection accuracy (90%) of distinct touch orientations are achieved. The presented technique prompts the development of piezoelectric force touch panels.

Touch and vision perception are two important functions humans use to interact with the real world. To mimic human‐like abilities, tactile‐ and visual‐sensing‐based intelligent humanoids have emerged and are going through a fast phase of development. Studies have demonstrated that the combination of tactile and visual information not only enables humanoids to better learn the environment, but also allows them to have pseudocognitive ability. Being a new and rapidly developing field of research, a significant growth in articles reporting different aspects of sensing and related machine learning is being witnessed. To help readers comprehensively understand the fundamentals and insights, and the current state of the art, this review is compiled to explain the working mechanisms of tactile and visual sensing, introduce the application of intelligent humanoids in diverse scenarios, discuss current challenges, and predict future trends.

With its unique advantages, unmanned aerial vehicle (UAV) radar remote sensing has become an advanced means of forest resource investigation. At present, mainstream radar sensors are difficult to meet the needs of micro-UAV payloads due to the limitations of mass, size, and power. As a contribution to the topic, this paper designed a highly integrated, lightweight and reliable Micro-UAV-Radar system and developed a novel method for retrieving vegetation height information on this basis. In order to test the availability of the system, a feasibility study has been carried out in Mangrove Reserve, China on April 2021. The results of this flight campaign show that the vegetation height retrieved by our method can reach high accuracy. Compared with other approaches, the method proposed in this study, which has lower data acquisition cost and simpler data processing, provides a novel way for retrieving vegetation height information.

Piezoelectric-based force sensing in interactive displays has attracted worldwide attention due to its intrinsic ability to efficiently convert mechanical energy into electric energy. However, because of faulty detection accuracy, commercialized products using piezoelectric materials have not succeeded in fully entering the market. To explain this situation, we investigate the matter through a finite element analysis (FEA) method. The results of our investigation reveal that unstable forcevoltage responsivity and propagated stress resulting from adjacent force-touch locations are two key issues that substantially degrade detection accuracy.

Piezoelectric force touch panels receive increased attentions in recent years. However, user-induced nonstable force-voltage responsivity limits their successful use in interactive displays. In this work, touch-induced capacitive information is used for estimating contact area and touch angle, which are further employed to interpret user performed force amplitude. A promising result of improving the stability of force-voltage responsivity by 85% is achieved, enhancing user experience and advancing the development of piezoelectric force sensing in interactive displays.

Interactive displays have been intensively used in consumer electronics, e.g., smartphones, for providing highly efficient human-machine interactions, assisting users to process a vast amount of data. However, current mainstream interactive displays only support 2-dimensional detection, strongly limiting user experience. In this paper, we present a piezoelectric and capacitive based technique for obtaining 4-dimensional detection, including (x-y axis) location, force, and hovers sensing in interactive displays. Here, electrodes are located on and underneath the piezoelectric layer, constructing a sandwich structured device. A self-capacitance-based technique is employed for location and hover sensing. Force-induced charges are collected and interpreted as force amplitude. As capacitive and piezoelectric sensing techniques occupy different frequency bands, these two signals can be smoothly separated, and 4-dimensional touch sensing is achieved concurrently, with the recognition of static force and hovers touch events.

Distributed scatterer (DS) decorrelation poses a challenge to multibaseline SAR interferometry. To overcome this challenge, the SqueeSAR retrieves an optimal phase time-series using a maximum likelihood estimation (MLE) method, which has been commonly used due to its remarkable effect. Unfortunately, however, the MLE's performance is compromised for various reasons, such as inaccurate statistically homogeneous pixels (SHPs) and the bias in the estimator used. In this paper, we present an approach aiming to improve the MLE's performance. The proposed approach includes the employment of the Kullback-Leibler divergence to realize more accurate SHP selection and the use of the second kind statistical estimator to mitigate the coherence bias. The performance of the conventional MLE is significantly improved by the proposed approach, making it close to its optimal performance. The experimental results on both simulated and real TerraSAR-X data demonstrate the improvements of the proposed approach with respect to the conventional MLE.

Amyotrophic lateral sclerosis (ALS) patients suffer from difficulties in grasping objects in daily life. To help them regain impaired functionalities at a certain level, in this article, we present a visual feedback based eye tracking and EEG based human-machine interactive technique, in which the eye tracking system is used for locating user desired objects, and the EEG system is for instructing corresponding operations. A robotic arm is employed here to grasp the targeted objects. Experimental results demonstrate that the developed system can accurately reflect user’s intentions in terms of grasping surrounding objects.

Mechanical hands with closed-loop control and haptic functions are in great need in recent years due to their merits such as satisfactory user experience, accurate grasp operation and etc. However, the high financial cost of these smart systems strongly limits the broad use of these smart systems. To address this issue, we designed a low-cost mechanical hand with haptic functionality. Experimental results demonstrate an average delay time of 11.8ms of the feedback system, and 1 kg of the maximum loading, and an average finger rotational speed of 155°/s, which illustrate that this design has met the common standard of similar products. The total cost of this system is only $1.7k, indicating its affordability for broad users.

With the development of intelligent transportation and autonomous vehicles, conventional tires are upgrading to smart tires, which are expected to provide new functions. In this article, we proposed a piezoelectric device integrated smart tire, through which, driving induced mechanical energy is converted to the electric energy, which is further used to interpret vehicle’s load and speed, and energy harvesting. In our experiments, the average detection error of speed and load reached 4.15 % and 5.23 %, respectively, and 22.6 pJ can be collected averagely during one wheel rotation. The presented work demonstrates a possible developing trend of smart tires, potentially advancing the filed.

Emotion recognition based on touch event's temporal and force information receives global interests. However, current consumer touch panels cannot provide user's accurate force data. Moreover, conventional studies extracting various features from complex missions, can't achieve real-time emotion detection. To address these two issues, in this paper, a piezoelectric based keystroke dynamic technique for quick emotion detection is presented. The high sensitivity of force detection is achieved for the nature of piezoelectric materials. Meanwhile, we simplify the mission to merely password entry and extract features from only time and pressure dimension, reducing the time spent for feature extraction and processing. The discrete model (PAD 3dimensional-model) for emotion classification is employed. International Affective Digitized Sounds (IADS) is applied to elicit emotions and a Chinese version of abbreviated PAD emotion scale is used to evaluate the degree of emotion induction. With Random Forest Classifier, a 4-emotion-classification (happiness, sadness, fear, disgust) with an average accuracy of 78.31% is achieved. The proposed technique improves the reliability and practicability of emotion recognition in realistic applications.

Teleoperation is a crucial function for human-robotic cooperation. This paper presents an advanced multisensory controlling robotic arm based on flexible electromyography (EMG) sensor and inertial measurement unit (IMU). Signals from these sensors map to different degrees of freedom (DOFs) of the human body. And a support vector machine based algorithm process the data sampling from the sensors to classify three standard muscle modes. This method could reach a high accuracy of 98.7%. Through an adaptive control method based control algorithm, one user could use the robotic arm to finish one task with an average time of 37.6s. This article demonstrates a feasible means for robotic teleoperation by integrating multiple wearable sensing techniques and the comprehensive control algorithm, potentially advancing the development of human-robotic cooperation.

Amyotrophic lateral sclerosis (ALS) patients will gradually lost their motion abilities and have huge inconvenience in their daily life. In this article, an artificial throat and eye tracking based smart assistive system is presented. Through detecting eye motions and throat vibrations, the established neural network can interpret the patient’s intention, and perform simple operations by a robotic arm (e.g. grasp an object). The experimental results demonstrate that the developed system can help patients obtain basic self-care ability.

The introduction of force touch detection improves the efficiency of human-machine interactivities. In previous studies, characteristics of dynamic force touch events and corresponding classification methods have been developed. However, static force touch is still untouched due to two challenges: user-induced pulses and hysteresis. In this article, an artificial neural network (ANN) is designed and optimized for addressing these two undesired phenomena. Experimental results demonstrate after attenuating these two issues, an average detection error of 0.15 N is successfully achieved.

Current commercial force touch panels can merely detect a single force touch’s location and amplitude. However, in many applications, multiple force touch events can occur at the same time among different locations of the touch panel. To satisfy this need, in this article, a piezoelectric and machine learning-based technique is proposed. Here, the piezoelectric film-based touch panel is used to detect different force levels, while the machine learning algorithm is developed to interpret the locations and strengths of user applied multiple force touch events. High detection accuracy of 92.3% for location determination and 88.2% for force level recognition is achieved.

Patients physically handicapped can't take care of themselves. Helping them easily control the objects around them will reduce their psychological burden and social pressure. In this article, a semi-autonomous grasping system based on eye movement and EEG is presented to achieve this goal. Patient just needs to gaze the target object and keep focused, the manipulator will automatically move to its position and grasp it. Experimental results verify the reliability of the system. This system promotes the development of human-computer interaction system based on multi-sensor fusion.

The severity of fundus arteriosclerosis can be determined and divided into four grades according to fundus images. Automatically grading of the fundus arteriosclerosis is helpful in clinical practices, so this paper proposes a convolutional neural network (CNN) method based on hierarchical attention maps to solve the automatic grading problem. First, we use the retinal vessel segmentation model to separate the important vascular region and the non-vascular background region from the fundus image and obtain two attention maps. The two maps are regarded as inputs to construct a two-stream CNN (TSNet), to focus on feature information through mutual reference between the two regions. In addition, we use convex hull attention maps in the one-stream CNN (OSNet) to learn valuable areas where the retinal vessels are concentrated. Then, we design an integrated OTNet model which is composed of TSNet that learns image feature information and OSNet that learns discriminative areas. After obtaining the representation learning parts of the two networks, we can train the classification layer to achieve better results. Our proposed TSNet reaches the AUC value of 0.796 and the ACC value of 0.592 on the testing set, and the integrated model OTNet reaches the AUC value of 0.806 and the ACC value of 0.606, which are better than the results of other comparable models. As far as we know, this is the first attempt to use deep learning to classify the severity of atherosclerosis in fundus images. The prediction results of our proposed method can be accepted by doctors, which shows that our method has a certain application value.

Since the 21st century, Internet technologies have entered a stage of rapid developments, and gradually been widely used in various industries. Internet Medical is gradually matured and used widely. Aiming at the problems such as the easy leakage of privacy in mobile medical equipment and untrustworthy data, we make use of a role-separated mechanism to generate trusted anonymous certificates and propose a lightweight identity authentication scheme to protect the medical data security. It reduces the calculation cost. Meanwhile, in view of problems of transparency and visibility of blockchain information, we adapt searchable encryption algorithm to realize ciphertext processing in the whole life cycle. Experiments show that our scheme can reduce the cost of computation on basis of ensuring traffic. In process of dynamic updating of ciphertext keywords, except the keyword identifier, less information is leaked to the server, which protect users’ privacy.

Online news is an important means to spread information. However, in order to attract readers, some online news writers often add pictures that have nothing to do with the news content, which seriously affects the credibility of the media. To solve these problems, we propose a news detection mechanism based on the knowledge graph, which can enrich the news image description and use the news image description to calculate the sentence similarity with the news text, so as to identify whether a news belongs to the news image mismatch. The realization of the image-text matching detection mechanism of news enhance the credibility of social media and purify the network environment.

Online news is an important means of disseminating information, but some online news texts do not match the content described in the original news pictures, and even cause ambiguity among readers. This phenomenon has seriously affected the authority of news and the credibility of news media. In response to this problem, this paper proposes a rich news image description mechanism based on the Concept-Net knowledge graph. The model we provide consists of two parts, namely, extracting the image content and rendering with natural language to generate an accurate description of the news image.

Background and objective Arteriosclerosis can reflect the severity of hypertension, which is one of the main diseases threatening human life safety. But Arteriosclerosis retinopathy detection involves costly and time-consuming manual assessment. To meet the urgent needs of automation, this paper developed a novel arteriosclerosis retinopathy grading method based on convolutional neural network. Methods Firstly, we propose a good scheme for extracting features facing the fundus blood vessel background using image merging for contour enhancement. In this step, the original image is dealt with adaptive threshold processing to generate the new contour channel, which merge with the original three-channel image. Then, we employ the pre-trained convolutional neural network with transfer learning to speed up training and contour image channel parameter with Kaiming initialization. Moreover, ArcLoss is applied to increase inter-class differences and intra-class similarity aiming to the high similarity of images of different classes in the dataset. Results The accuracy of arteriosclerosis retinopathy grading achieved by the proposed method is up to 65.354%, which is nearly 4% higher than those of the exiting methods. The Kappa of our method is 0.508 in arteriosclerosis retinopathy grading. Conclusions An experimental study on multiple metrics demonstrates the superiority of our method, which will be a useful to the toolbox for arteriosclerosis retinopathy grading.

The research of haptic interaction has a very long history. Nowadays, how to integrate haptic interaction and user experience is still a main challenge in the field of human–machine interaction. In the complete interaction process, the status of interaction model from industry usually presented as “visual as base, haptic and acoustic as assist.” In this chapter, the definition, importance, and the future development direction of user experience are discussed from the perspective of users. The methodologies for the evaluation of user experience in different haptic interfaces are deeply explored. From the outside to the inside, from behavioral observation to physiological measurement, with the help of various case studies, which kind of qualitative and quantitative research can assist to evaluate user’s haptic perception, emotional experience, and immersion is discussed.

Since the first touch panel was developed in 1963, different touch-sensing technologies have been developed to satisfy diverse application scenarios. In general, touch detection technologies can be divided into two types: contact- and non-contact-based architectures. Between them, the former (including resistive-, capacitive-, acoustic-, and optical-based techniques) is mainly used for small- and medium-sized interactive displays, while the latter (including cameras, motion sensors, and electric capacitive tomography-based techniques) is more popular for large area displays. In this chapter, both mainstream contact and non-contact touch-sensing technologies are described.

Touch is one of the most fundamental activities used by human beings to learn about the surrounding world. Traditionally, through touches, an infant starts to recognize different objectives, customers can distinguish clothing materials, a mother can estimate the temperature of a bottle of milk for her baby, and the blind can read braille. In recent years, with the rapid development of electronic and information technologies, smart-end terminals have emerged, e.g., laptops, smartphones, and smart watches, based on which, touch becomes not only a means to learn the world but also an effective method to exchange information with the world. Today, people input and retrieve information using various touch-supported systems. In pursuit of the development of the Internet of things (IoT) and smart cities, touch-based interactivity is expected to be employed ubiquitously. Looking at our living and working environments, many touches are performed every day for surfing the Internet and engaging in online banking and messaging. It is not difficult to conclude that touch-based human–machine interactivities (HMIs) are omnipresent. In this chapter, the history and current status of touch sensing are discussed.

Touch interactivity is of significance in the area of HMI. The development of touch sensing changes the conventional manner of controlling electronic and mechanical systems, thus boosting HMI efficiency. In recent years, excellent system performance (e.g., high location detection accuracy) and user experiences in touch-sensing systems have been provided by a variety of commercialized products, e.g., smartphones, which are now deemed to be necessary components in people’s daily lives. With the rapid progress of electronic and information technologies, touch interactivities are thought to play a wider role.

The amount of volumetric brain image increases rapidly, which requires a vast amount of resources for storage and transmission, so it’s urgent to explore an efficient volumetric compression method. Recent years have witnessed the progress of deep learning-based approaches for two-dimensional (2D) natural image compression, but the field of learned volumetric image compression still remains unexplored. In this paper, we propose the first end-to-end learning framework for volumetric image compression by extending the advanced techniques of 2D image compression to volumetric images. Specifically, a convolutional autoencoder is used to compress 3D image cubes, and the non-local attention models are embedded in the convolutional autoencoder to jointly capture local and global correlations. Both hyperprior and autoregressive models are used to perform the conditional probability estimation in entropy coding. To reduce model complexity, we introduce a convolutional long short-term memory network for the autoregressive model based on channel-wise prediction. Experimental results on volumetric mouse brain images show that the proposed method outperforms JPEG2000-3D, HEVC and state-of-the-art 2D methods.

Piezoelectric touch sensing in interactive displays gains increasing attentions due to its high force detection sensitivity and intrinsic mechanical-to-electrical conversion ability. However, the instable force-voltage responsivity induced by preload effect of piezoelectric materials reduces the force detection accuracy of secondary force touches, which is important to touch and haptic applications such as peek-pop. To address this issue, in this article, we present a preload effect elimination technique, in which the relationship between the piezoelectric coefficients and static preload is studied first, and then the detected secondary force touch is calibrated by using the previously applied static force information. Experimental results demonstrate that the force detection accuracy is boosted by 15.17% after applying the developed technique to secondary force touches with different preload values, potentially allowing the system to precisely interpret secondary force touch amplitude and hence enhancing the development of touch sensing in interactive displays.

Acupuncture is one of the most significant therapies of Chinese traditional medical science, and it is now globally utilized for treatment, e.g., pain management. Traditionally, there is no quantification means for storing masters’ skills and examining trainee’s learning effect, hence, strongly limiting the development of acupuncture. To address this issue, in this article, a piezoelectric glove based wearable stress sensing system is presented. Experimental results showcase that through the piezoelectric force sensing glove, key parameters (e.g., peak stress at needle) during performing acupuncture are detected and extracted, potentially improving the learning efficiency of trainees and therefore advancing the progress of acupuncture.

Benefiting from the development of the Internet of Healthcare Things (IoHT) in recent years, locomotion mode recognition using wearable sensors plays an important role in the field of in-home rehabilitation. In this paper, a smart sensing system utilizing flexible electromyography (EMG) sensors and ground reaction force (GRF) sensors for locomotion mode recognition is presented, together with its use under the IoHT architecture. EMG and GRF information from ten healthy subjects in five common locomotion modes in daily life were collected, analyzed, and then transmitted to remote end terminals (e.g., personal computers). The data analysis process was implemented with machine learning techniques (Support Vector Machine), through which the locomotion modes were determined with a high accuracy of 96.38%. This article demonstrates a feasible means for accurate locomotion mode recognition by combining wearable sensing techniques and the machine learning algorithm, potentially advancing the development for IoHT based in-home rehabilitation.

In the field of rehabilitation, the electromyography (EMG) signal plays an important role in interpreting patients’ intentions and physical conditions. Nevertheless, utilizing merely the EMG signal suffers from difficulty in recognizing slight body movements, and the detection accuracy is strongly influenced by environmental factors. To address the above issues, multisensory integration-based EMG pattern recognition (PR) techniques have been developed in recent years, and fruitful results have been demonstrated in diverse rehabilitation scenarios, such as achieving high locomotion detection and prosthesis control accuracy. Owing to the importance and rapid development of the EMG centered multisensory fusion technologies in rehabilitation, this paper reviews both theories and applications in this emerging field. The principle of EMG signal generation and the current pattern recognition process are explained in detail, including signal preprocessing, feature extraction, classification algorithms, etc. Mechanisms of collaborations between two important multisensory fusion strategies (kinetic and kinematics) and EMG information are thoroughly explained; corresponding applications are studied, and the pros and cons are discussed. Finally, the main challenges in EMG centered multisensory pattern recognition are discussed, and a future research direction of this area is prospected.

Rapid development of high-throughput technologies has permitted the identification of an increasing number of disease-associated genes (DAGs), which are important for understanding disease initiation and developing precision therapeutics. However, DAGs often contain large amounts of redundant or false positive information, leading to difficulties in quantifying and prioritizing potential relationships between these DAGs and human diseases. In this study, a network-oriented gene entropy approach (NOGEA) is proposed for accurately inferring master genes that contribute to specific diseases by quantitatively calculating their perturbation abilities on directed disease-specific gene networks. In addition, we confirmed that the master genes identified by NOGEA have a high reliability for predicting disease-specific initiation events and progression risk. Master genes may also be used to extract the underlying information of different diseases, thus revealing mechanisms of disease comorbidity. More importantly, approved therapeutic targets are topologically localized in a small neighborhood of master genes on the interactome network, which provides a new way for predicting new drug-disease associations. Through this method, 11 old drugs were newly identified and predicted to be effective for treating pancreatic cancer and then validated by in vitro experiments. Collectively, the NOGEA was useful for identifying master genes that control disease initiation and co-occurrence, thus providing a valuable strategy for drug efficacy screening and repositioning. NOGEA codes are publicly available at https://github.com/guozihuaa/NOGEA.

Piezoelectric force touch panels are attractive as human-machine interfaces and 3-dimensional touch sensing in internet of things (IoT) applications. The piezoelectric material has the intrinsic ability to convert mechanical to electrical signals. But the force responsivity issue induced by different touch orientations can be unstable. This paper presents a piezoelectric touch panel that is sensitive to both capacitive and force stimulation. A touch orientation classification technique is developed to calibrate the detected force amplitude by training a machine learning model with finger induced capacitive information. A high and stable force voltage responsivity of 87.5% is achieved experimentally, demonstrating its potential significance in force touch based human-machine interactivity.

Artificial smart skins capable of interacting with people and sensing environmental stimuli have become a research topic in humanoid robotic applications. However, previously reported architectures suffer difficulties in achieving multi-dimensional sensing in a simple structure with a low system cost. To address this issue, in this paper, an artificial smart skin constructed with polyimide/copper/polyvinylidene fluoride (PVDF) is presented for detecting 2D-position, proximity, dynamic force, and humidity via a smart combination of piezoelectric- and capacitive-effects. The proposed system achieves overall force and capacitive sensitivities of 0.051 N and 8.7 fF; the humidity measurements show a responsivity at 0.20%/RH% over a relative humidity range of 10%–90% RH. And a follow-up filtering algorithm is proposed to separate the stimuli associated with capacitance changes (position, proximity, and humidity). This simple-structured device supports multiple functions with its low system cost, thus advancing the field of robotics smart skins.

In Internet of Things (IoT) applications, secure access to smart systems, e.g., smartphones, is important for protecting private information. Among various authentication techniques, keystroke authentication methods based on touch behavior of the user have received increasing attention. This is due to the unique benefits, such as no additional hardware component and the ease of use in most smart systems. In this paper, we present a technique for obtaining high user authentication accuracy by utilizing a user’s touch time and force information, which are obtained from a piezoelectric touch panel. After combining artificial neural networks with the user’s touch features, an equal error rate (EER) of 1.09% is achieved, validating the feasibility of the proposed technique for achieving highly secure user authentication, hence advancing the development of security techniques potentially deployable in the field of IoT.

Concurrent high force detection accuracy and extended battery lifetime are strongly expected in wearable gait monitoring systems, which are important for many Internet of Health Things (IoHT) applications. In this article, a piezoelectric insole device and rectifying circuitry based technique is presented to achieve these two ultimate goals. Here, walking induced positive and negative charges are separated for plantar stress detection and energy harvesting respectively, realizing the two functions concurrently. Experimental results demonstrate that first, the high detection sensitivity of 55 mN and responsivity of 231 mV/N are achieved, satisfying the need for diagnosing various diseases; second, energy of 1.6 pJ is stored during a walking event, consequently extending the battery lifetime. The developed technique enhances the development of gait monitoring in IoHT.

In traditional touch panels, electrodes are settled on or above the surface of displays, weakening the optical transmittance, hence resulting in high power consumption of the display for providing customers satisfied visual experience, giving rise to reduced battery’s lifetime which brings users inconvenience. To address this issue, in this article, we propose a new sensor architecture, in which electrodes are settled only at the edge of the touch panel, ensuring a very high optical transmittance. The touch event detection relies on electrical capacitance tomography (ECT) technique, through which 2-dimensional location recognition is achieved, indicating the presented technique provides a feasible means to boost the optical transmittance of the touch panel layer.

Force touch based interactivity has been widely integrated into displays equipped in most of smart electronic systems such as smartphones and tablets. This paper reports on application of artificial neural networks to analyze data generated from piezoelectric based touch panels for providing customized force sensing operation. Based on the experimental results, high force sensing accuracy (93.3%) is achieved when three force levels are used. Two-dimensional sensing, also achieved with the proposed technique, with high detection accuracy (95.2%). The technique presented here not only achieves high accuracy, but also allows users to define the range of force levels through behavioral means thus enhancing interactivity experience.