Sensors

15 pages, 1467 KiB

Open AccessArticle

Which Visual Modality Is Important When Judging the Naturalness of the Agent (Artificial Versus Human Intelligence) Providing Recommendations in the Symbolic Consumption Context?

by Kyungmi Chung, Jin Young Park, Kiwan Park and Yaeri Kim

Sensors 2020, 20(17), 5016; https://doi.org/10.3390/s20175016 - 3 Sep 2020

Cited by 3 | Viewed by 3307

Abstract

This study aimed to explore how the type and visual modality of a recommendation agent’s identity affect male university students’ (1) self-reported responses to agent-recommended symbolic brand in evaluating the naturalness of virtual agents, human, or artificial intelligence (AI) and (2) early event-related [...] Read more.

This study aimed to explore how the type and visual modality of a recommendation agent’s identity affect male university students’ (1) self-reported responses to agent-recommended symbolic brand in evaluating the naturalness of virtual agents, human, or artificial intelligence (AI) and (2) early event-related potential (ERP) responses between text- and face-specific scalp locations. Twenty-seven participants (M = 25.26, SD = 5.35) whose consumption was more motivated by symbolic needs (vs. functional) were instructed to perform a visual task to evaluate the naturalness of the target stimuli. As hypothesized, the subjective evaluation showed that they had lower attitudes and perceived higher unnaturalness when the symbolic brand was recommended by AI (vs. human). Based on this self-report, two epochs were segmented for the ERP analysis: human-natural and AI-unnatural. As revealed by P100 amplitude modulation on visual modality of two agents, their evaluation relied more on face image rather than text. Furthermore, this tendency was consistently observed in that of N170 amplitude when the agent identity was defined as human. However, when the agent identity was defined as AI, reversed N170 modulation was observed, indicating that participants referred more to textual information than graphical information to assess the naturalness of the agent. Full article

(This article belongs to the Special Issue Sensors for Behavioral Science—Social, Affective, and Cognitive Science Perspectives)

► Show Figures

Figure 1

15 pages, 4206 KiB

Open AccessArticle

Vibroarthrographic Signal Spectral Features in 5-Class Knee Joint Classification

by Adam Łysiak, Anna Froń, Dawid Bączkowicz and Mirosław Szmajda

Sensors 2020, 20(17), 5015; https://doi.org/10.3390/s20175015 - 3 Sep 2020

Cited by 11 | Viewed by 3298

Abstract

Vibroarthrography (VAG) is a non-invasive and potentially widely available method supporting the joint diagnosis process. This research was conducted using VAG signals classified to five different condition classes: three stages of chondromalacia patellae, osteoarthritis, and control group (healthy knee joint). Ten new spectral [...] Read more.

Vibroarthrography (VAG) is a non-invasive and potentially widely available method supporting the joint diagnosis process. This research was conducted using VAG signals classified to five different condition classes: three stages of chondromalacia patellae, osteoarthritis, and control group (healthy knee joint). Ten new spectral features were proposed, distinguishing not only neighboring classes, but every class combination. Additionally, Frequency Range Maps were proposed as the frequency feature extraction visualization method. The results were compared to state-of-the-art frequency features using the Bhattacharyya coefficient and the set of ten different classification algorithms. All methods evaluating proposed features indicated the superiority of the new features compared to the state-of-the-art. In terms of Bhattacharyya coefficient, newly proposed features proved to be over 25% better, and the classification accuracy was on average 9% better. Full article

(This article belongs to the Special Issue Advances in Sensors and Sensing for Technical Condition Assessment and NDT)

► Show Figures

Figure 1

20 pages, 7966 KiB

Open AccessArticle

Remotely Sensed Data Fusion for Spatiotemporal Geostatistical Analysis of Forest Fire Hazard

by Stavros Sakellariou, Pedro Cabral, Mário Caetano, Filiberto Pla, Marco Painho, Olga Christopoulou, Athanassios Sfougaris, Nicolas Dalezios and Christos Vasilakos

Sensors 2020, 20(17), 5014; https://doi.org/10.3390/s20175014 - 3 Sep 2020

Cited by 22 | Viewed by 4340

Abstract

Forest fires are a natural phenomenon which might have severe implications on natural and anthropogenic ecosystems. Future projections predict that, under a climate change environment, the fire season would be lengthier with higher levels of droughts, leading to higher fire severity. The main [...] Read more.

Forest fires are a natural phenomenon which might have severe implications on natural and anthropogenic ecosystems. Future projections predict that, under a climate change environment, the fire season would be lengthier with higher levels of droughts, leading to higher fire severity. The main aim of this paper is to perform a spatiotemporal analysis and explore the variability of fire hazard in a small Greek island, Skiathos (a prototype case of fragile environment) where the land uses mixture is very high. First, a comparative assessment of two robust modeling techniques was examined, namely, the Analytical Hierarchy Process (AHP) knowledge-based and the fuzzy logic AHP to estimate the fire hazard in a timeframe of 20 years (1996–2016). The former technique was proven more representative after the comparative assessment with the real fire perimeters recorded on the island (1984–2016). Next, we explored the spatiotemporal dynamics of fire hazard, highlighting the risk changes in space and time through the individual and collective contribution of the most significant factors (topography, vegetation features, anthropogenic influence). The fire hazard changes were not dramatic, however, some changes have been observed in the southwestern and northern part of the island. The geostatistical analysis revealed a significant clustering process of high-risk values in the southwestern and northern part of the study area, whereas some clusters of low-risk values have been located in the northern territory. The degree of spatial autocorrelation tends to be greater for 1996 rather than for 2016, indicating the potential higher transmission of fires at the most susceptible regions in the past. The knowledge of long-term fire hazard dynamics, based on multiple types of remotely sensed data, may provide the fire and land managers with valuable fire prevention and land use planning tools. Full article

(This article belongs to the Special Issue Remote Sensing and Geoinformatics in Wildfire Management)

► Show Figures

Figure 1

17 pages, 8593 KiB

Open AccessArticle

A Novel Inspection Technique for Electronic Components Using Thermography (NITECT)

by Haochen Liu, Lawrence Tinsley, Wayne Lam, Sri Addepalli, Xiaochen Liu, Andrew Starr and Yifan Zhao

Sensors 2020, 20(17), 5013; https://doi.org/10.3390/s20175013 - 3 Sep 2020

Cited by 4 | Viewed by 3670

Abstract

Unverified or counterfeited electronic components pose a big threat globally because they could lead to malfunction of safety-critical systems and reduced reliability of high-hazard assets. The current inspection techniques are either expensive or slow, which becomes the bottleneck of large volume inspection. As [...] Read more.

Unverified or counterfeited electronic components pose a big threat globally because they could lead to malfunction of safety-critical systems and reduced reliability of high-hazard assets. The current inspection techniques are either expensive or slow, which becomes the bottleneck of large volume inspection. As a complement of the existing inspection capabilities, a pulsed thermography-based screening technique is proposed in this paper using a digital twin methodology. A FEM-based simulation unit is initially developed to simulate the internal structure of electronic components with deviations of multiple physical properties, informed by X-ray data, along with its thermal behaviour under exposure to instantaneous heat. A dedicated physical inspection unit is then integrated to verify the simulation unit and further improve the simulation by taking account of various uncertainties caused by equipment and samples. Principle component analysis is used for feature extraction, and then a set of machine learning-based classifiers are employed for quantitative classification. Evaluation results of 17 chips from different sources successfully demonstrate the effectiveness of the proposed technique. Full article

(This article belongs to the Special Issue Sensors for Fault Diagnosis and Prognostics)

► Show Figures

Figure 1

30 pages, 11574 KiB

Open AccessArticle

Cryptographic Keys Generating and Renewing System for IoT Network Nodes—A Concept

by Janusz Furtak

Sensors 2020, 20(17), 5012; https://doi.org/10.3390/s20175012 - 3 Sep 2020

Cited by 5 | Viewed by 3322

Abstract

Designers and users of the Internet of Things (IoT) are devoting more and more attention to the issues of security and privacy as well as the integration of data coming from various areas. A critical element of cooperation is building mutual trust and [...] Read more.

Designers and users of the Internet of Things (IoT) are devoting more and more attention to the issues of security and privacy as well as the integration of data coming from various areas. A critical element of cooperation is building mutual trust and secure data exchange. Because IoT devices usually have small memory resources, limited computing power, and limited energy resources, it is often impossible to effectively use a well-known solution based on the Certification Authority. This article describes the concept of the system for a cryptographic Key Generating and Renewing system (KGR). The concept of the solution is based on the use of the hardware Trusted Platform Module (TPM) v2.0 to support the procedures of creating trust structures, generating keys, protecting stored data, and securing data exchange between system nodes. The main tasks of the system are the secure distribution of a new symmetric key and renewal of an expired key for data exchange parties. The KGR system is especially designed for clusters of the IoT nodes but can also be used by other systems. A service based on the Message Queuing Telemetry Transport (MQTT) protocol will be used to exchange data between nodes of the KGR system. Full article

(This article belongs to the Section Internet of Things)

► Show Figures

Figure 1

16 pages, 1807 KiB

Open AccessLetter

Mechanical Fault Diagnostic in PMSM from Only One Current Measurement: A Tacholess Order Tracking Approach

by Abdallah Allouche, Erik Etien, Laurent Rambault, Thierry Doget, Sebastien Cauet and Anas Sakout

Sensors 2020, 20(17), 5011; https://doi.org/10.3390/s20175011 - 3 Sep 2020

Cited by 7 | Viewed by 2768

Abstract

This article presents a mechanical fault diagnosis methodology in synchronous machines using only a single current measurement in variable speed conditions. The proposed methodology uses order tracking in order to sample the analysis signal as a function of the rotor angle. The spectrum [...] Read more.

This article presents a mechanical fault diagnosis methodology in synchronous machines using only a single current measurement in variable speed conditions. The proposed methodology uses order tracking in order to sample the analysis signal as a function of the rotor angle. The spectrum of the signal is then independent of speed and it could be employed in frequency analysis. Order tracking is usually applied using rotor position measurement. In this work, the proposed method uses one current measurement to estimate the position as well as the analysis signal (rotation speed). Furthermore, a statistical approach is used to create a complete diagnosis protocol. At variable speed and with only one current measurement the diagnosis is challenging. However, order tracking will allow simpler analysis. The method is proved in simulations and experimental set-up. Full article

(This article belongs to the Special Issue Sensing Technology and Data Interpretation in Machine Diagnosis and Systems Condition Monitoring)

► Show Figures

Figure 1

19 pages, 5785 KiB

Open AccessArticle

A Non-Linear Temperature Compensation Model for Improving the Measurement Accuracy of an Inductive Proximity Sensor and Its Application-Specific Integrated Circuit Implementation

by Li Wang, Hui-Bin Tao, Hang Dong, Zhi-Biao Shao and Fei Wang

Sensors 2020, 20(17), 5010; https://doi.org/10.3390/s20175010 - 3 Sep 2020

Cited by 7 | Viewed by 3462

Abstract

The non-linear characteristic of a non-contacting Inductive Proximity Sensor (IPS) with the temperature affects the computation accuracy when measuring the target distance in real time. The linear model based method for distance estimation shows a large deviation at a low temperature. Accordingly, this [...] Read more.

The non-linear characteristic of a non-contacting Inductive Proximity Sensor (IPS) with the temperature affects the computation accuracy when measuring the target distance in real time. The linear model based method for distance estimation shows a large deviation at a low temperature. Accordingly, this paper presents a non-linear measurement model, which computes the target distance accurately in real time within a wide temperature range from

- 55 ° C

to

125 ° C

. By revisiting the temperature effect on the IPS system, this paper considers the non-linear characteristic of the IPS measurement system due to the change of temperature. The proposed model adopts a non-linear polynomial algorithm rather than the simple linear Look-Up Table (LUT) method, which provides more accurate distance estimation compared to the previous work. The introduced model is fabricated in a 0.18

μ

m Complementary Metal Oxide Semiconductor (CMOS) process and packaged in a CQFN40. For the most commonly used sensing distance of 4 mm, the computed distance deviation of the Application-Specific Integrated Circuit (ASIC) chips falls within the range of

[- 0.2, 0.2]

mm. According to the test results of the ASIC chips, this non-linear temperature compensation model successfully achieves real-time and high-accuracy computation within a wide temperature range with low hardware resource consumption. Full article

(This article belongs to the Special Issue Sensors and Methods for Dynamic Measurement)

► Show Figures

Figure 1

16 pages, 6541 KiB

Open AccessArticle

Autonomous Energy Harvester Based on Textile-Based Enzymatic Biofuel Cell for On-Demand Usage

by Seonho Seok, Cong Wang, Elie Lefeuvre and Jungyul Park

Sensors 2020, 20(17), 5009; https://doi.org/10.3390/s20175009 - 3 Sep 2020

Cited by 7 | Viewed by 2991

Abstract

This paper presents an autonomous energy harvester based on a textile-based enzymatic biofuel cell, enabling an efficient power management and on-demand usage. The proposed biofuel cell works by an enzymatic reaction with glucose in sweat absorbed by the specially designed textile for sustainable [...] Read more.

This paper presents an autonomous energy harvester based on a textile-based enzymatic biofuel cell, enabling an efficient power management and on-demand usage. The proposed biofuel cell works by an enzymatic reaction with glucose in sweat absorbed by the specially designed textile for sustainable and efficient energy harvesting. The output power of the textile-based biofuel cell has been optimized by changing electrode size and stacking electrodes and corresponding fluidic channels suitable for following power management circuit. The output power level of single electrode is estimated less than 0.5 μW and thus a two-staged power management circuit using intermediate supercapacitor has been presented. As a solution to produce a higher power level, multiple stacks of biofuel cell electrodes have been proposed and thus the textile-based biofuel cell employing serially connected 5 stacks produces a maximal power of 13 μW with an output voltage of 0.88 V when load resistance is 40 kΩ. A buck-boost converter employing a crystal oscillator directly triggered by DC output voltage of the biofuel cell makes it possible to obtain output voltage of the DC–DC converter is 6.75 V. The efficiency of the DC–DC converter is estimated as approximately 50% when the output power of the biofuel cell is tens microwatts. In addition, LT-spice modeling and simulation has been presented to estimate power consumption of each element of the proposed DC–DC converter circuit and the predicted output voltage has good agreement with measurement result. Full article

(This article belongs to the Section Biosensors)

► Show Figures

Figure 1

17 pages, 2478 KiB

Open AccessArticle

A Fast Estimation Method for Direction of Arrival Using Tripole Vector Antenna

by Bodong Zhang, Xuan Zou, Tingyi Zhang, Yunong Tang and Hao Zeng

Sensors 2020, 20(17), 5008; https://doi.org/10.3390/s20175008 - 3 Sep 2020

Cited by 3 | Viewed by 2732

Abstract

The tripole vector antenna comprises three orthogonal dipole antennas, so it could completely capture all the electric field of the incident electromagnetic (EM) wave. Then, the electric field information could be used to estimate the direction of arrival (DOA) of the EM wave [...] Read more.

The tripole vector antenna comprises three orthogonal dipole antennas, so it could completely capture all the electric field of the incident electromagnetic (EM) wave. Then, the electric field information could be used to estimate the direction of arrival (DOA) of the EM wave if two conditions are satisfied. One is that there exists only one single EM wave in space. The other is that the EM wave is elliptically or circularly polarized. The new estimation method obtains two snapshot vectors through the output of a tripole antenna and computes their cross-product vector. Furthermore, the direction of the cross-product vector is used to estimate the DOA of the EM wave directly. We analyze the statistical characteristics of the DOA estimation error to prove that the new scheme is an asymptotic unbiased estimation. Furthermore, unlike the existing Multiple Signal Classification (MUSIC)-based algorithms, the proposed approach only need one tripole vector antenna instead of an antenna array. Meanwhile, the new method also outperforms existing MUSIC-based algorithms in the term of computational complexity. Finally, the performance and advantages of the proposed method are verified by numerical simulations. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

15 pages, 10706 KiB

Open AccessArticle

A Deep-Learning Method for Radar Micro-Doppler Spectrogram Restoration

by Yuan He, Xinyu Li, Runlong Li, Jianping Wang and Xiaojun Jing

Sensors 2020, 20(17), 5007; https://doi.org/10.3390/s20175007 - 3 Sep 2020

Cited by 7 | Viewed by 4514

Abstract

Radio frequency interference, which makes it difficult to produce high-quality radar spectrograms, is a major issue for micro-Doppler-based human activity recognition (HAR). In this paper, we propose a deep-learning-based method to detect and cut out the interference in spectrograms. Then, we restore the [...] Read more.

Radio frequency interference, which makes it difficult to produce high-quality radar spectrograms, is a major issue for micro-Doppler-based human activity recognition (HAR). In this paper, we propose a deep-learning-based method to detect and cut out the interference in spectrograms. Then, we restore the spectrograms in the cut-out region. First, a fully convolutional neural network (FCN) is employed to detect and remove the interference. Then, a coarse-to-fine generative adversarial network (GAN) is proposed to restore the part of the spectrogram that is affected by the interferences. The simulated motion capture (MOCAP) spectrograms and the measured radar spectrograms with interference are used to verify the proposed method. Experimental results from both qualitative and quantitative perspectives show that the proposed method can mitigate the interference and restore high-quality radar spectrograms. Furthermore, the comparison experiments also demonstrate the efficiency of the proposed approach. Full article

(This article belongs to the Special Issue Advanced Radar Techniques, Applications and Developments)

► Show Figures

Figure 1

Figure 1
The pipeline of the proposed deep-learning method for interference restoration. (a) The FCN is trained with the spectrograms where there is interference with the supervision of the label. Then when a spectrogram with interference is fed into the trained FCN, the FCN can locate the interference accurately. (b) The GAN is trained with clear radar spectrograms. In this way, the GAN can learn the data distribution of clear spectrograms for further interference mitigation. (c) Finally, when a spectrogram with interference is fed into the FCN and fed into the GAN subsequently, a clear spectrogram can be restored. Full article ">Figure 2
The structure of generator in the proposed GAN for interference mitigation. The descriptions with a form of “A × B/C, D” represent that there are D convolution kernels with a size of A × B. In addition, the convolution stride is C. DR refers to dilated rate of dilated convolution. Full article ">Figure 3
The structure of discriminator in the proposed GAN for interference mitigation. The descriptions with a form of “A × B/C, D” represent that there are D convolution kernels with a size of A × B. In addition, the convolution stride is C. DR refers to dilated rate of dilated convolution. Full article ">Figure 4
Contextual attention layer. Several patches (3 × 3) are extracted from the background and reshaped to the size of the missing part feature maps after two downsampling blocks. Full article ">Figure 5
Results of the proposed method and three state-of-the-art methods on the simulated data. (a) The clean radar spectrograms. (b) The spectrograms with diverse interference. (c) The locations of interference detected by the proposed FCN model. The red boxes represent the ground truth of the locations of interferences. (d) The radar spectrograms restored with our method. (e) The spectrograms restored with Zeroing. (f) The spectrograms restored with FCNs. (g) The spectrograms with restored ResNet. Full article ">Figure 6
Results of the proposed hybrid FCN and GAN method and three state-of-the-art methods on the measured data. (a) The clean radar spectrograms. (b) The spectrograms with diverse types of interferences. (c) The locations of interferences detected by the proposed FCN model. The red boxes represent the ground truth of the locations of interferences. (d) The radar spectrograms restored with our method. (e) The spectrograms restored with Zeroing. (f) The spectrograms restored with FCNs. (g) The spectrograms with restored ResNet. Full article ">

20 pages, 2139 KiB

Open AccessReview

Benefits of Home-Based Solutions for Diagnosis and Treatment of Acute Coronary Syndromes on Health Care Costs: A Systematic Review

by Pau Redón, Atif Shahzad, Talha Iqbal and William Wijns

Sensors 2020, 20(17), 5006; https://doi.org/10.3390/s20175006 - 3 Sep 2020

Cited by 2 | Viewed by 3416

Abstract

Diagnosing and treating acute coronary syndromes consumes a significant fraction of the healthcare budget worldwide. The pressure on resources is expected to increase with the continuing rise of cardiovascular disease, other chronic diseases and extended life expectancy, while expenditure is constrained. The objective [...] Read more.

Diagnosing and treating acute coronary syndromes consumes a significant fraction of the healthcare budget worldwide. The pressure on resources is expected to increase with the continuing rise of cardiovascular disease, other chronic diseases and extended life expectancy, while expenditure is constrained. The objective of this review is to assess if home-based solutions for measuring chemical cardiac biomarkers can mitigate or reduce the continued rise in the costs of ACS treatment. A systematic review was performed considering published literature in several relevant public databases (i.e., PUBMED, Cochrane, Embase and Scopus) focusing on current biomarker practices in high-risk patients, their cost-effectiveness and the clinical evidence and feasibility of implementation. Out of 26,000 references screened, 86 met the inclusion criteria after independent full-text review. Current clinical evidence highlights that home-based solutions implemented in primary and secondary prevention reduce health care costs by earlier diagnosis, improved patient outcomes and quality of life, as well as by avoidance of unnecessary use of resources. Economical evidence suggests their potential to reduce health care costs if the incremental cost-effectiveness ratio or the willingness-to-pay does not surpass £20,000/QALY or €50,000 limit per 20,000 patients, respectively. The cost-effectiveness of these solutions increases when applied to high-risk patients. Full article

(This article belongs to the Special Issue Internet of Medical Things in Healthcare Applications)

► Show Figures

Figure 1

17 pages, 2934 KiB

Open AccessLetter

A Framework for Human-Robot-Human Physical Interaction Based on N-Player Game Theory

by Rui Zou, Yubin Liu, Jie Zhao and Hegao Cai

Sensors 2020, 20(17), 5005; https://doi.org/10.3390/s20175005 - 3 Sep 2020

Cited by 7 | Viewed by 3489

Abstract

In order to analyze the complex interactive behaviors between the robot and two humans, this paper presents an adaptive optimal control framework for human-robot-human physical interaction. N-player linear quadratic differential game theory is used to describe the system under study. N-player differential game [...] Read more.

In order to analyze the complex interactive behaviors between the robot and two humans, this paper presents an adaptive optimal control framework for human-robot-human physical interaction. N-player linear quadratic differential game theory is used to describe the system under study. N-player differential game theory can not be used directly in actual scenerie, since the robot cannot know humans’ control objectives in advance. In order to let the robot know humans’ control objectives, the paper presents an online estimation method to identify unknown humans’ control objectives based on the recursive least squares algorithm. The Nash equilibrium solution of human-robot-human interaction is obtained by solving the coupled Riccati equation. Adaptive optimal control can be achieved during the human-robot-human physical interaction. The effectiveness of the proposed method is demonstrated by rigorous theoretical analysis and simulations. The simulation results show that the proposed controller can achieve adaptive optimal control during the interaction between the robot and two humans. Compared with the LQR controller, the proposed controller has more superior performance. Full article

(This article belongs to the Special Issue Human-Robot Collaborations in Industrial Automation)

► Show Figures

Figure 1

18 pages, 13822 KiB

Open AccessArticle

Bathymetric Monitoring of Alluvial River Bottom Changes for Purposes of Stability of Water Power Plant Structure with a New Methodology for River Bottom Hazard Mapping (Wloclawek, Poland)

by Dariusz Popielarczyk, Marian Marschalko, Tomasz Templin, Dominik Niemiec, Isik Yilmaz and Barbara Matuszková

Sensors 2020, 20(17), 5004; https://doi.org/10.3390/s20175004 - 3 Sep 2020

Cited by 5 | Viewed by 2568

Abstract

The aim of this research was to produce a new methodology for a special river bottom hazard mapping for the stability purposes of the biggest Polish water power plant: Włocławek. During the operation period of the water power plant, an engineering-geological issue in [...] Read more.

The aim of this research was to produce a new methodology for a special river bottom hazard mapping for the stability purposes of the biggest Polish water power plant: Włocławek. During the operation period of the water power plant, an engineering-geological issue in the form of pothole formation on the Wisła River bed in the gravel-sand alluvium was observed. This was caused by increased fluvial erosion resulting from a reduced water level behind the power plant, along with frequent changes in the water flow rates and water levels caused by the varying technological and economic operation needs of the power plant. Data for the research were obtained by way of a 4-year geodetic/bathymetric monitoring of the river bed implemented using integrated GNSS (Global Navigation Satellite System), RTS (Robotized Total Station) and SBES (Single Beam Echo Sounder) methods. The result is a customized river bottom hazard map which takes into account a high, medium, and low risk levels of the potholes for the water power plant structure. This map was used to redevelop the river bed by filling. The findings show that high hazard is related to 5% of potholes (capacity of 4308 m³), medium with 38% of potholes (capacity of 36,455 m³), and low hazard with 57% of potholes (capacity of 54,396 m³). Since the construction of the dam, changes due to erosion identified by the monitoring have concerned approximately 405,252 m³ of the bottom, which corresponds to 130 Olympic-size pools. This implies enormous changes, while a possible solution could be the construction of additional cascades on the Wisła River. Full article

(This article belongs to the Special Issue Telemetry and Monitoring for Land and Water Ecosystems)

► Show Figures

Figure 1

29 pages, 2796 KiB

Open AccessArticle

Empirical and Comparative Validation for a Building Energy Model Calibration Methodology

by Vicente Gutiérrez González, Germán Ramos Ruiz and Carlos Fernández Bandera

Sensors 2020, 20(17), 5003; https://doi.org/10.3390/s20175003 - 3 Sep 2020

Cited by 32 | Viewed by 3928

Abstract

The digital world is spreading to all sectors of the economy, and Industry 4.0, with the digital twin, is a reality in the building sector. Energy reduction and decarbonization in buildings are urgently required. Models are the base for prediction and preparedness for [...] Read more.

The digital world is spreading to all sectors of the economy, and Industry 4.0, with the digital twin, is a reality in the building sector. Energy reduction and decarbonization in buildings are urgently required. Models are the base for prediction and preparedness for uncertainty. Building energy models have been a growing field for a long time. This paper proposes a novel calibration methodology for a building energy model based on two pillars: simplicity, because there is an important reduction in the number of parameters (four) to be adjusted, and cost-effectiveness, because the methodology minimizes the number of sensors provided to perform the process by 47.5%. The new methodology was validated empirically and comparatively based on a previous work carried out in Annex 58 of the International Energy Agency (IEA). The use of a tested and structured experiment adds value to the results obtained. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

19 pages, 3679 KiB

Open AccessArticle

Application of Machine Learning for the in-Field Correction of a PM_2.5 Low-Cost Sensor Network

by Wen-Cheng Vincent Wang, Shih-Chun Candice Lung and Chun-Hu Liu

Sensors 2020, 20(17), 5002; https://doi.org/10.3390/s20175002 - 3 Sep 2020

Cited by 22 | Viewed by 4251

Abstract

Many low-cost sensors (LCSs) are distributed for air monitoring without any rigorous calibrations. This work applies machine learning with PM_2.5 from Taiwan monitoring stations to conduct in-field corrections on a network of 39 PM_2.5 LCSs from July 2017 to December 2018. [...] Read more.

Many low-cost sensors (LCSs) are distributed for air monitoring without any rigorous calibrations. This work applies machine learning with PM_2.5 from Taiwan monitoring stations to conduct in-field corrections on a network of 39 PM_2.5 LCSs from July 2017 to December 2018. Three candidate models were evaluated: Multiple linear regression (MLR), support vector regression (SVR), and random forest regression (RFR). The model-corrected PM_2.5 levels were compared with those of GRIMM-calibrated PM_2.5. RFR was superior to MLR and SVR in its correction accuracy and computing efficiency. Compared to SVR, the root mean square errors (RMSEs) of RFR were 35% and 85% lower for the training and validation sets, respectively, and the computational speed was 35 times faster. An RFR with 300 decision trees was chosen as the optimal setting considering both the correction performance and the modeling time. An RFR with a nighttime pattern was established as the optimal correction model, and the RMSEs were 5.9 ± 2.0 μg/m³, reduced from 18.4 ± 6.5 μg/m³ before correction. This is the first work to correct LCSs at locations without monitoring stations, validated using laboratory-calibrated data. Similar models could be established in other countries to greatly enhance the usefulness of their PM_2.5 sensor networks. Full article

(This article belongs to the Collection Sensors for Air Quality Monitoring)

► Show Figures

Figure 1

Figure 1
Distribution of AS-LUNG-O sets, Environmental Protection Administration (EPA) stations, and sensors of the citizen air quality network (CAQN) in (a) the whole of Taiwan island, (b) the Taipei metropolitan area from July 2017 to December 2018, and (c) an AS-LUNG-O set at street level. Full article ">Figure 1 Cont.
Distribution of AS-LUNG-O sets, Environmental Protection Administration (EPA) stations, and sensors of the citizen air quality network (CAQN) in (a) the whole of Taiwan island, (b) the Taipei metropolitan area from July 2017 to December 2018, and (c) an AS-LUNG-O set at street level. Full article ">Figure 2
Flow chart of the data correction process. Full article ">Figure 3
Model performance of (a) multiple linear regression (MLR) with a training set, (b) MLR with a validation set, (c) support vector regression (SVR) with a training set, (d) SVR with a validation set, (e) random forest regression (RFR) with a training set, and (f) RFR with a validation set. The RMSE, R2, and n are listed in the graphs. Full article ">Figure 4
Model efficiency of the random forest regression model. Full article ">Figure 5
Correction results of RFR with a nighttime pattern for different seasons with (a) RMSE and (b) MAE as performance indicators. Full article ">Figure 6
Time-series of raw PM2.5, model-corrected PM2.5, and GRIMM-calibrated PM2.5. Full article ">

22 pages, 2480 KiB

Open AccessArticle

Robust Wavelength Selection Using Filter-Wrapper Method and Input Scaling on Near Infrared Spectral Data

by Divo Dharma Silalahi, Habshah Midi, Jayanthi Arasan, Mohd Shafie Mustafa and Jean-Pierre Caliman

Sensors 2020, 20(17), 5001; https://doi.org/10.3390/s20175001 - 3 Sep 2020

Cited by 12 | Viewed by 2384

Abstract

The extraction of relevant wavelengths from a large dataset of Near Infrared Spectroscopy (NIRS) is a significant challenge in vibrational spectroscopy research. Nonetheless, this process allows the improvement in the chemical interpretability by emphasizing the chemical entities related to the chemical parameters of [...] Read more.

The extraction of relevant wavelengths from a large dataset of Near Infrared Spectroscopy (NIRS) is a significant challenge in vibrational spectroscopy research. Nonetheless, this process allows the improvement in the chemical interpretability by emphasizing the chemical entities related to the chemical parameters of samples. With the complexity in the dataset, it may be possible that irrelevant wavelengths are still included in the multivariate calibration. This yields the computational process to become unnecessary complex and decreases the accuracy and robustness of the model. In multivariate analysis, Partial Least Square Regression (PLSR) is a method commonly used to build a predictive model from NIR spectral data. However, in the PLSR method and common commercial chemometrics software, there is no standard wavelength selection procedure applied to screen the irrelevant wavelengths. In this study, a new robust wavelength selection procedure called the modified VIP-MCUVE (mod-VIP-MCUVE) using Filter-Wrapper method and input scaling strategy is introduced. The proposed method combines the modified Variable Importance in Projection (VIP) and modified Monte Carlo Uninformative Variable Elimination (MCUVE) to calculate the scale matrix of the input variable. The modified VIP uses the orthogonal components of Partial Least Square (PLS) in investigating the informative variable in the model by applying the amount of variation both in

X

and

y

{SSX, SSY}

, simultaneously. The modified MCUVE uses a robust reliability coefficient and a robust tolerance interval in the selection procedure. To evaluate the superiority of the proposed method, the classical VIP, MCUVE, and autoscaling procedure in classical PLSR were also included in the evaluation. Using artificial data with Monte Carlo simulation and NIR spectral data of oil palm (Elaeis guineensis Jacq.) fruit mesocarp, the study shows that the proposed method offers advantages to improve model interpretability, to be computationally extensive, and to produce better model accuracy. Full article

(This article belongs to the Section Chemical Sensors)

► Show Figures

Figure 1

19 pages, 3986 KiB

Open AccessArticle

A Soft Sensor Approach Based on an Echo State Network Optimized by Improved Genetic Algorithm

by Ruoyu Huang, Zetao Li and Bin Cao

Sensors 2020, 20(17), 5000; https://doi.org/10.3390/s20175000 - 3 Sep 2020

Cited by 7 | Viewed by 3038

Abstract

In the process of fault diagnosis and the health and safety operation evaluation of modern industrial processes, it is crucial to measure important state variables, which cannot be directly detected due to limitations of economy, technology, environment and space. Therefore, this paper proposes [...] Read more.

In the process of fault diagnosis and the health and safety operation evaluation of modern industrial processes, it is crucial to measure important state variables, which cannot be directly detected due to limitations of economy, technology, environment and space. Therefore, this paper proposes a data-driven soft sensor approach based on an echo state network (ESN) optimized by an improved genetic algorithm (IGA). Firstly, with an ESN, a data-driven model (DDM) between secondary variables and dominant variables is established. Secondly, in order to improve the prediction performance, the IGA is utilized to optimize the parameters of the ESN. Then, the immigration strategy is introduced and the crossover and mutation operators are changed adaptively to improve the convergence speed of the algorithm and address the problem that the algorithm falls into the local optimum. Finally, a soft sensor model of an ESN optimized by an IGA is established (IGA-ESN), and the advantages and performance of the proposed method are verified by estimating the alumina concentration in an aluminum reduction cell. The experimental results illustrated that the proposed method is efficient, and the error was significantly reduced compared with the traditional algorithm. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

18 pages, 2576 KiB

Open AccessArticle

Assessing the Validity and Reliability of A Low-Cost Microcontroller-Based Load Cell Amplifier for Measuring Lower Limb and Upper Limb Muscular Force

by Julie Gaudet and Grant Handrigan

Sensors 2020, 20(17), 4999; https://doi.org/10.3390/s20174999 - 3 Sep 2020

Cited by 3 | Viewed by 5186

Abstract

Lower and upper limb maximum muscular force development is an important indicator of physical capacity. Manual muscle testing, load cell coupled with a signal conditioner, and handheld dynamometry are three widely used techniques for measuring isometric muscle strength. Recently, there is a proliferation [...] Read more.

Lower and upper limb maximum muscular force development is an important indicator of physical capacity. Manual muscle testing, load cell coupled with a signal conditioner, and handheld dynamometry are three widely used techniques for measuring isometric muscle strength. Recently, there is a proliferation of low-cost tools that have potential to be used to measure muscle strength. This study examined both the criterion validity, inter-day reliability and intra-day reliability of a microcontroller-based load cell amplifier for quantifying muscle strength. To do so, a low-cost microcontroller-based load cell amplifier for measuring lower and upper limb maximal voluntary isometric muscular force was compared to a commercial grade signal conditioner and to a handheld dynamometer. The results showed that the microcontroller-based load cell amplifier correlated nearly perfectly (Pearson's R-values between 0.947 to 0.992) with the commercial signal conditioner and the handheld dynamometer, and showed good to excellent association when calculating ICC scores, with values of 0.9582 [95% C.I.: 0.9297–0.9752] for inter-day reliability and of 0.9269 [95% C.I.: 0.8909–0.9533] for session one, intra-day reliability. Such results may have implications for how the evaluation of muscle strength measurement is conducted in the future, particularly for offering a commercial-like grade quality, low cost, portable and flexible option. Full article

(This article belongs to the Special Issue Sensor-Based Systems for Kinematics and Kinetics)

► Show Figures

Figure 1

14 pages, 2902 KiB

Open AccessArticle

FRET-Based Ca²⁺ Biosensor Single Cell Imaging Interrogated by High-Frequency Ultrasound

by Sangpil Yoon, Yijia Pan, Kirk Shung and Yingxiao Wang

Sensors 2020, 20(17), 4998; https://doi.org/10.3390/s20174998 - 3 Sep 2020

Cited by 8 | Viewed by 5001

Abstract

Fluorescence resonance energy transfer (FRET)-based biosensors have advanced live cell imaging by dynamically visualizing molecular events with high temporal resolution. FRET-based biosensors with spectrally distinct fluorophore pairs provide clear contrast between cells during dual FRET live cell imaging. Here, we have developed a [...] Read more.

Fluorescence resonance energy transfer (FRET)-based biosensors have advanced live cell imaging by dynamically visualizing molecular events with high temporal resolution. FRET-based biosensors with spectrally distinct fluorophore pairs provide clear contrast between cells during dual FRET live cell imaging. Here, we have developed a new FRET-based Ca²⁺ biosensor using EGFP and FusionRed fluorophores (FRET-GFPRed). Using different filter settings, the developed biosensor can be differentiated from a typical FRET-based Ca²⁺ biosensor with ECFP and YPet (YC3.6 FRET Ca²⁺ biosensor, FRET-CFPYPet). A high-frequency ultrasound (HFU) with a carrier frequency of 150 MHz can target a subcellular region due to its tight focus smaller than 10 µm. Therefore, HFU offers a new single cell stimulations approach for FRET live cell imaging with precise spatial resolution and repeated stimulation for longitudinal studies. Furthermore, the single cell level intracellular delivery of a desired FRET-based biosensor into target cells using HFU enables us to perform dual FRET imaging of a cell pair. We show that a cell pair is defined by sequential intracellular delivery of the developed FRET-GFPRed and FRET-CFPYPet into two target cells using HFU. We demonstrate that a FRET-GFPRed exhibits consistent 10–15% FRET response under typical ionomycin stimulation as well as under a new stimulation strategy with HFU. Full article

(This article belongs to the Special Issue Ultrasonic Systems for Biomedical Sensing)

► Show Figures

Figure 1

Figure 1
(A) A schematic diagram of a new fluorescence resonance energy transfer (FRET) biosensor (FRET-GFPRed). A new FRET-based Ca2+ biosensor with EGFP and FusionRed fluorophores (FRET-GFPRed) was designed to show FRET between EGFP and FusionRed. FRET increases when calcium ions bind to Calmodulin (CaM) and M13, which provides dynamic readouts of intracellular calcium concentration. (B) Time course of the FRET/EGFP signal ratio. FRET-GFPRed expressing HeLa cells were stimulated with ionomycin. Images on the left represents before (t = 0 s), 5 s, and 100 s after ionomycin stimulation (1 μM) in HeLa cells at t = 0 s. A green arrowhead surrounded by red lines indicates HeLa cells transfected with FRET-GFPRed. A scale bar represents 10 μm. Error bars indicate ± one SEM. Full article ">Figure 2
High-frequency ultrasonic transducer and integrated system for the intracellular delivery and stimulation. (A) Schematic diagram of 150 MHz ultrasonic transducer shows SMA connector (yellow part) and needle type transducer (grey part) for cell manipulation. Lithium niobate (LibNO3) was used to generate 150 MHz ultrasound pulse for the intracellular delivery and stimulation. Aperture (1 mm) was focused to have 1 mm focal depth with fnumber of 1. Electrical pulse (EP) was applied through positive connection to trigger LibNO3. (B) Experimentally measured echo and its spectrum of the developed transducer shows that actual center frequency was 150 MHz. (C) An integrated system for the intracellular delivery and stimulation of target cells was comprised of a fluorescence microscope and a high frequency ultrasound transducer shown in A. A function generator and a power amplifier generated EP to excite the high frequency ultrasound transducer. Peak-to-peak voltage (Vpp) and pulse duration (tw) are controlled by a function generator. Number of pulses (NP) is the total number of ultrasound pulses to target cells. The location of the transducer was controlled by a 3D translation/rotation stage. (D) Target cell in red is being treated by ultrasound pulse generated by high frequency ultrasonic transducer. For the intracellular delivery of macromolecules, small molecules can be presented in the same Petri dish. Yellow cone between the transducer tip and the target cell indicates ultrasound pulse (not to scale). Full article ">Figure 3
The characterization of a new FRET-based Ca2+ biosensor with EGFP and FusionRed fluorophores (FRET-GFPRed) by chemical stimulation. A FRET-based Ca2+ biosensor with ECFP and YPet (FRET-CFPYPet) was used for comparison. (A) Normalized fluorescence spectra showing excitation (EX, thin lines) and emission (EM, thick lines) of EGFP (green lines) and FRET-FusionRed (red lines) are presented. The filter setting GR has excitation filters for EGFP (blue region, 450–490 nm) and FRET-FusionRed (purple region, 400–440 nm) as well as emission filters for EGFP (green region, 500–550 nm) and FRET-FusionRed (red region, 600–700 nm). Time course of the typical FRET-GFPRed signal after ionomycin stimulation is presented in <a href="#sensors-20-04998-f001" class="html-fig">Figure 1</a>B. (B) Normalized fluorescence spectra showing excitation (EX, thin lines) and emission (EM, thick lines) of ECFP (cyan lines) and YPet (green lines) are presented. The filter setting CY has an excitation filter (purple region, 400–450 nm) for ECFP/YPet and emission filters for ECFP (blue region, 460–500 nm) and YPet (green region, 520–550 nm). (C,D) Two Hela cells in one field of view are shown. The upper and lower cells were transfected with FRET-GFPRed and FRET-CFPYPet, respectively. The representative (C) FRET/EGFP and (D) FRET/ECFP ratio images of two cells before and after ionomycin stimulation are presented. (C) Time course of the FRET/EGFP signal ratio of FRET-GFPRed (left) and FRET-CFPYPet (right) indicating a spectral bleed-through of FRET-CFPYPet signal using the filter setting GR. (D) Time course of the FRET/ECFP signal ratio of FRET-CFPYPet (left) and FRET-GFPRed (right) indicating no spectral bleed-through of the FRET-GFPRed signal using the filter setting CY. (C,D) Normalized FRET/EGFP and FRET/ECFP signal ratios were obtained by live cell imaging using the filter settings GR and CY, respectively. Green arrowheads surrounded by red lines indicate HeLa cells transfected with FRET-GFPRed. Blue arrowheads surrounded by yellow lines represent cells transfected with FRET-CFPYPet. Black arrows represent ionomycin (chemical) stimulation. Scale bars represent 10 μm. Error bars represent +/− one SD. Full article ">Figure 4
A new FRET-based Ca2+ biosensor with EGFP and FusionRed fluorophores (FRET-GFPRed) stimulated with high frequency ultrasound. A FRET-based Ca2+ biosensor with ECFP and YPet (FRET-CFPYPet) was used for comparison. FRET-GFPRed and FRET-CFPYPet were transfected by Llipofectamine 3000. (A) Representative FRET/EGFP ratio images of FRET-GFPRed expressed HeLa cells before (t = 0 s), 5 s, and 100 s after ultrasound stimulation. FRET images were taken using the filter setting GR as shown in <a href="#sensors-20-04998-f003" class="html-fig">Figure 3</a>A. A high frequency ultrasound was used to target only the upper cell for ultrasound stimulation. The lower cell was not stimulated. (B) Time course of the FRET/EGFP signal ratio of FRET-GFPRed after ultrasound stimulation is presented. (C) Representative FRET/ECFP signal ratio images of FRET-CFPYPet before (t = 0 s), 5 s, and 100 s after ultrasound stimulation of HeLa cells are presented. FRET images were taken using the filter setting CY as shown in <a href="#sensors-20-04998-f003" class="html-fig">Figure 3</a>B. (D) Time course of the FRET/ECFP signal ratio of FRET-CFPYPet after ultrasound stimulation is presented. Green arrowheads surrounded by red lines indicate HeLa cells transfected with FRET-GFPRed. Blue arrowheads surrounded by yellow lines indicate cells transfected with FRET-CFPYPet. Black arrows indicate ultrasound stimulation. The duration of the ultrasound stimulation was from 5 to 10 μs. Scale bars represent 10 μm. Error bars represent +/− one SEM. Full article ">Figure 5
Repeated and transient HFU stimulation. (A) Images and (B) time course of FRET/ECFP signal ratio of Ca2+ biosensor show transient and repeated stimulations by high-frequency ultrasound at the target single cell (dashed line in A). (B) High-frequency ultrasound (arrowheads) was applied three times for 5 μs. (C) A DIC image of cells in (A). Full article ">Figure 6
A new FRET-based Ca2+ biosensor with EGFP and FusionRed fluorophores (FRET-GFPRed) stimulated with high frequency ultrasound. A FRET-based Ca2+ biosensor with ECFP and YPet (FRET-CFPYPet) was used for comparison. FRET-GFPRed and FRET-CFPYPet were transfected by high frequency ultrasound, which can specifically transfect single cells with the desired FRET biosensor. (A) Representative FRET/EGFP signal ratio images of FRET-GFPRed in a HeLa cell before (t = 0 s), 5 s, and 100 s after ultrasound stimulation are presented. FRET images were taken using the filter setting GR as shown in <a href="#sensors-20-04998-f003" class="html-fig">Figure 3</a>A. (B) Time course of the FRET/EGFP signal ratio of FRET-GFPRed after ultrasound stimulation is presented. (C) Representative FRET/ECFP signal ratio images of FRET-CFPYPet before (t = 0 s), 5 s, and 100 s after ultrasound stimulation is shown. FRET images were taken using the filter setting CY as shown in <a href="#sensors-20-04998-f003" class="html-fig">Figure 3</a>B. (D) Time course of the FRET/ECFP ratio of the FRET-CFPYPet after ultrasound stimulation (t = 0) is presented. Green arrowheads surrounded by red lines indicate HeLa cells transfected with FRET-GFPRed. Blue arrowheads surrounded by yellow lines represent cells transfected with FRET-CFPYPet. Black arrows indicate ultrasound stimulation. The duration of the stimulation was from 5 to 10 μs. Scale bars represent 10 μm. Error bars represent +/− one SEM. Full article ">

15 pages, 10188 KiB

Open AccessArticle

An Improved Design and Implementation of a Range-Controlled Communication System for Mobile Phones

by Mingyang Gong, Haichun Zhang and Zhenglin Liu

Sensors 2020, 20(17), 4997; https://doi.org/10.3390/s20174997 - 3 Sep 2020

Viewed by 2973

Abstract

The Short-range-controlled communication system (RCC) based on a subscriber identity module (SIM) card is a replacement for the standard near-field communication (NFC) system to support near-field payment applications. The RCC uses both the low-frequency (LF) and high-frequency (HF) wireless communication system. The RCC [...] Read more.

The Short-range-controlled communication system (RCC) based on a subscriber identity module (SIM) card is a replacement for the standard near-field communication (NFC) system to support near-field payment applications. The RCC uses both the low-frequency (LF) and high-frequency (HF) wireless communication system. The RCC communication distance is controlled under 10 cm. However, current RCCs suffer from compatibility issues, and the LF communication distance is lower than 0.5 cm in some phones with completely metallic shells. In this paper, we propose an improved LF communication system design, including an LF transmitter circuit, LF receiver chip, and LF-HF communication protocol. The LF receiver chip has a rail-to-rail amplifier and a self-correcting clock recovery differential Manchester decoder, which do not have the limitations of accurate gain and high system clock. The LF receiver chip is fabricated in a 0.18

μ

m CMOS technology platform, with a die size of 1.05 mm × 0.9 mm and current consumption of 41

μ

A. The experiments show that the improved RCC has better compatibility, and the communication distance reaches to 4.2 cm in phones with completely metallic shells. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Figure 1

21 pages, 11170 KiB

Open AccessArticle

Fast Wearable Sensor–Based Foot–Ground Contact Phase Classification Using a Convolutional Neural Network with Sliding-Window Label Overlapping

by Haneul Jeon, Sang Lae Kim, Soyeon Kim and Donghun Lee

Sensors 2020, 20(17), 4996; https://doi.org/10.3390/s20174996 - 3 Sep 2020

Cited by 13 | Viewed by 4539

Abstract

Classification of foot–ground contact phases, as well as the swing phase is essential in biomechanics domains where lower-limb motion analysis is required; this analysis is used for lower-limb rehabilitation, walking gait analysis and improvement, and exoskeleton motion capture. In this study, sliding-window label [...] Read more.

Classification of foot–ground contact phases, as well as the swing phase is essential in biomechanics domains where lower-limb motion analysis is required; this analysis is used for lower-limb rehabilitation, walking gait analysis and improvement, and exoskeleton motion capture. In this study, sliding-window label overlapping of time-series wearable motion data in training dataset acquisition is proposed to accurately detect foot–ground contact phases, which are composed of 3 sub-phases as well as the swing phase, at a frequency of 100 Hz with a convolutional neural network (CNN) architecture. We not only succeeded in developing a real-time CNN model for learning and obtaining a test accuracy of 99.8% or higher, but also confirmed that its validation accuracy was close to 85%. Full article

(This article belongs to the Collection Sensors for Gait, Human Movement Analysis, and Health Monitoring)

► Show Figures

Figure 1

25 pages, 1016 KiB

Open AccessArticle

The Probability of Ischaemic Stroke Prediction with a Multi-Neural-Network Model

by Yan Liu, Bo Yin and Yanping Cong

Sensors 2020, 20(17), 4995; https://doi.org/10.3390/s20174995 - 3 Sep 2020

Cited by 13 | Viewed by 3979

Abstract

As is known, cerebral stroke has become one of the main diseases endangering people’s health; ischaemic strokes accounts for approximately 85% of cerebral strokes. According to research, early prediction and prevention can effectively reduce the incidence rate of the disease. However, it is [...] Read more.

As is known, cerebral stroke has become one of the main diseases endangering people’s health; ischaemic strokes accounts for approximately 85% of cerebral strokes. According to research, early prediction and prevention can effectively reduce the incidence rate of the disease. However, it is difficult to predict the ischaemic stroke because the data related to the disease are multi-modal. To achieve high accuracy of prediction and combine the stroke risk predictors obtained by previous researchers, a method for predicting the probability of stroke occurrence based on a multi-model fusion convolutional neural network structure is proposed. In such a way, the accuracy of ischaemic stroke prediction is improved by processing multi-modal data through multiple end-to-end neural networks. In this method, the feature extraction of structured data (age, gender, history of hypertension, etc.) and streaming data (heart rate, blood pressure, etc.) based on a convolutional neural network is first realized. A neural network model for feature fusion is then constructed to realize the feature fusion of structured data and streaming data. Finally, a predictive model for predicting the probability of stroke is obtained by training. As shown in the experimental results, the accuracy of ischaemic stroke prediction reached 98.53%. Such a high prediction accuracy will be helpful for preventing the occurrence of stroke. Full article

(This article belongs to the Collection Deep Learning in Biomedical Informatics and Healthcare)

► Show Figures

Figure 1

14 pages, 4096 KiB

Open AccessArticle

Simulation/Experiment Confrontation, an Efficient Approach for Sensitive SAW Sensors Design

by Bilel Achour, Ghada Attia, Chouki Zerrouki, Najla Fourati, Kosai Raoof and Nourdin Yaakoubi

Sensors 2020, 20(17), 4994; https://doi.org/10.3390/s20174994 - 3 Sep 2020

Cited by 11 | Viewed by 4096

Abstract

Sensitivity is one of the most important parameters to put in the foreground in all sensing applications. Its increase is therefore an ongoing challenge, particularly for surface acoustic wave (SAW) sensors. Herein, finite element method (FEM) simulation using COMSOL Multiphysics software is first [...] Read more.

Sensitivity is one of the most important parameters to put in the foreground in all sensing applications. Its increase is therefore an ongoing challenge, particularly for surface acoustic wave (SAW) sensors. Herein, finite element method (FEM) simulation using COMSOL Multiphysics software is first used to simulate the physical and electrical properties of SAW delay line. Results indicate that 2D configuration permits to accurately obtain all pertinent parameters, as in 3D simulation, with very substantial time saving. A good agreement between calculation and experiment, in terms of transfer functions (S21 spectra), was also shown to evaluate the dependence of the SAW sensors sensitivity on the operating frequency; 2D simulations have been conducted on 104 MHz and 208 MHz delay lines, coated with a polyisobutylene (PIB) as sensitive layer to dichloromethane (DCM). A fourfold increase in sensitivity was obtained by doubling frequency. Both sensors were then realized and tested as chem-sensors to detect zinc ions in liquid media. 9-{[4-({[4-(9anthrylmethoxy)phenyl]sulfanyl} methyl)]methyl] anthracene (TDP-AN) was selected as the sensing layer. Results show a comparable response curves for both designed sensors, in terms of limit of detection and dissociation constants K_d values. On the other hand, experimental sensitivity values were of the order of [7.0 ± 2.8] × 10⁸ [°/M] and [16.0 ± 7.6] × 10⁸ [°/M] for 104 MHz and 208 MHz sensors, respectively, confirming that the sensitivity increases with frequency. Full article

(This article belongs to the Special Issue Sensors for Environmental and Life Science Applications)

► Show Figures

Figure 1

17 pages, 5363 KiB

Open AccessArticle

Prototype System for Measuring and Analyzing Movements of the Upper Limb for the Detection of Occupational Hazards

by Dolores Parras-Burgos, Alfonso Gea-Martínez, Lucas Roca-Nieto, Daniel G. Fernández-Pacheco and Francisco J. F. Cañavate

Sensors 2020, 20(17), 4993; https://doi.org/10.3390/s20174993 - 3 Sep 2020

Cited by 3 | Viewed by 3324

Abstract

In the work environment, there are usually different pathologies that are related to Repetitive Efforts and Movements (REM) that tend to predominantly affect the upper limbs. To determine whether a worker is at risk of suffering some type of pathology, observation techniques are [...] Read more.

In the work environment, there are usually different pathologies that are related to Repetitive Efforts and Movements (REM) that tend to predominantly affect the upper limbs. To determine whether a worker is at risk of suffering some type of pathology, observation techniques are usually used by qualified technical personnel. In order to define from quantitative data if there is a risk of suffering a pathology due to movements and repetitive efforts in the upper limb, a prototype of a movement measurement system has been designed and manufactured. This system interferes minimally with the activity studied, maintaining a reduced cost of manufacture and use. The system allows the study of the movements made by the subject in the work environment by determining the origin of the Musculoskeletal Disorder (MSD) from the movements of the elbow and wrist, collecting data on the position and accelerations of the arm, forearm and hand, and taking into account the risk factors established for suffering from an MSD: high repetition of movements, the use of a high force in a repetitive manner, or the adoption of forced positions. The data obtained with this system can be analyzed by qualified personnel from tables, graphs, and 3D animations at the time of execution, or stored for later analysis. Full article

(This article belongs to the Section Wearables)

► Show Figures

Figure 1

16 pages, 4807 KiB

Open AccessArticle

Classification Accuracy Improvement for Small-Size Citrus Pests and Diseases Using Bridge Connections in Deep Neural Networks

by Shuli Xing and Malrey Lee

Sensors 2020, 20(17), 4992; https://doi.org/10.3390/s20174992 - 3 Sep 2020

Cited by 13 | Viewed by 2991

Abstract

Due to the rich vitamin content in citrus fruit, citrus is an important crop around the world. However, the yield of these citrus crops is often reduced due to the damage of various pests and diseases. In order to mitigate these problems, several [...] Read more.

Due to the rich vitamin content in citrus fruit, citrus is an important crop around the world. However, the yield of these citrus crops is often reduced due to the damage of various pests and diseases. In order to mitigate these problems, several convolutional neural networks were applied to detect them. It is of note that the performance of these selected models degraded as the size of the target object in the image decreased. To adapt to scale changes, a new feature reuse method named bridge connection was developed. With the help of bridge connections, the accuracy of baseline networks was improved at little additional computation cost. The proposed BridgeNet-19 achieved the highest classification accuracy (95.47%), followed by the pre-trained VGG-19 (95.01%) and VGG-19 with bridge connections (94.73%). The use of bridge connections also strengthens the flexibility of sensors for image acquisition. It is unnecessary to pay more attention to adjusting the distance between a camera and pests and diseases. Full article

(This article belongs to the Special Issue Smart Agriculture Sensors)

► Show Figures

Figure 1

19 pages, 5137 KiB

Open AccessArticle

A Fully-Printed CRLH Dual-Band Dipole Antenna Fed by a Compact CRLH Dual-Band Balun

by Muhammad Kamran Khattak, Changhyeong Lee, Heejun Park and Sungtek Kahng

Sensors 2020, 20(17), 4991; https://doi.org/10.3390/s20174991 - 3 Sep 2020

Cited by 7 | Viewed by 4423

Abstract

In this paper, a new design method is proposed for a planar and compact dual-band dipole antenna. The dipole antenna has arms as a hybrid CRLH (Composite right- and left-handed) transmission-line comprising distributed and lumped elements for the dual-band function. The two arms [...] Read more.

In this paper, a new design method is proposed for a planar and compact dual-band dipole antenna. The dipole antenna has arms as a hybrid CRLH (Composite right- and left-handed) transmission-line comprising distributed and lumped elements for the dual-band function. The two arms are fed by the outputs of a compact and printed CRLH dual-band balun which consists of a CRLH hybrid coupler and an additional CRLH phase-shifter. Its operational frequencies are 2.4 and 5.2 GHz as popular mobile applications. Verifying the method, the circuit approach, EM (Electromagnetics) simulation and measurement are conducted and their results turn out to agree well with each other. Additionally, the CRLH property is shown with the dispersion diagram and the effective size-reduction is mentioned. Full article

(This article belongs to the Special Issue Antennas and Propagation)

► Show Figures

Figure 1

14 pages, 2952 KiB

Open AccessArticle

Using Optical Tracking System Data to Measure Team Synergic Behavior: Synchronization of Player-Ball-Goal Angles in a Football Match

by Daniel Carrilho, Micael Santos Couceiro, João Brito, Pedro Figueiredo, Rui J. Lopes and Duarte Araújo

Sensors 2020, 20(17), 4990; https://doi.org/10.3390/s20174990 - 3 Sep 2020

Cited by 11 | Viewed by 5215

Abstract

The ecological dynamics approach to interpersonal relationships provides theoretical support to the use of kinematic data, obtained with sensor-based systems, in which players of a team are linked mainly by information from the performance environment. Our goal was to capture the properties of [...] Read more.

The ecological dynamics approach to interpersonal relationships provides theoretical support to the use of kinematic data, obtained with sensor-based systems, in which players of a team are linked mainly by information from the performance environment. Our goal was to capture the properties of synergic behavior in football, using spatiotemporal data from one match of the 2018 FIFA WORLD CUP RUSSIA, to explore the application of player-ball-goal angles in cluster phase analysis. Linear mixed effects models were used to test the statistical significance of different effects, such as: team, half(-time), role and pitch zones. Results showed that the cluster phase values (synchronization) for the home team, had a

3.812 \times 10^{- 2} \pm 0.536 \times 10^{- 2}

increase with respect to the away team (

X^{2} (41) = 259.8, p < 0.001

) and that changing the role from with ball to without ball increased synchronization by

16.715 \times 10^{- 2} \pm 0.283 \times 10^{- 2}

(

X^{2} (41) = 12227.0, p < 0.001

). The interaction between effects was also significant. The player-team relative phase, the player-ball-goal angles relative frequency and the team configurations, showed that variations of synchronization might indicate critical performance changes (ball possession changes, goals scored, etc.). This study captured the ongoing player-environment link and the properties of team synergic behavior, supporting the use of sensor-based data computations in the development of relevant indicators for tactical analysis in sports. Full article

(This article belongs to the Special Issue Sensor-Based Systems for Kinematics and Kinetics)

► Show Figures

Figure 1

Figure 1
The player-ball-goal angle (PBGA). The numbers next to each player represent the PBGA in radian, measured with the vertex on the ball (white filled circle), considering the ball-player and ball-goal vectors, illustrated by the white arrows. Full article ">Figure 2
Mean and standard deviation (SD) of the synchronization (<math display="inline"><semantics> <mrow> <mover> <mi>r</mi> <mo>´</mo> </mover> </mrow> </semantics></math>) values per team, in the two halves of the match and by ball possession. Values in red represent the away team, values in blue represent the home team. Full article ">Figure 3
The impact of each zone in mean synchronization (<math display="inline"><semantics> <mrow> <mover> <mi>r</mi> <mo>´</mo> </mover> </mrow> </semantics></math>) values, for the without ball role. The lateral and longitudinal coordinates are represented, in meters, next to each zone code: O—offensive; MO—mid-offensive; MD—mid-defensive; D—defensive; R—right; CR—center-right; CL—center-left; L—left. The red-blue gradient scale indicates the synchronization values range from 0 to 1. Ball position data were inverted on the second half to make the results uniform for one unique direction per team. Full article ">Figure 4
Variations of the order parameter (synchronization) and the player relative phase to the team, during a specific time frame (30 s) in the first half of the match. (a) Relative phase of each player to the team, in both teams; (b) synchrony (<math display="inline"><semantics> <mrow> <mover> <mi>r</mi> <mo>´</mo> </mover> </mrow> </semantics></math>) measures for each team. The grey background indicates that the game is not running. Red or blue background indicate when the away team (red) or the home team (blue) have possession of the ball and white background indicates that the team does not have possession of the ball; (c) exemplar key events of the match that are expressed by three apexes of the synchronization values, circled in red on the left-side (away team). Full article ">Figure 5
Relative frequency of the player-ball-goal angle (PBGA) illustrated by the polar graphs of four players of the away team in the roles with ball (red) and without ball (blue). The PBGA ranges from 0 to π. The relative frequency ranges from 0 to 7000 and can be seen on the bottom horizontal axis of each polar graph. To the right side of each player’s polar graph there is a heatmap of the respective player’s positions. Full article ">Figure 6
Convex hull of each subgroup in each configuration, in three selected frames of the match. Left: configurations of the team with ball (red). Right: configurations of the team without ball (blue). Subgroups are represented by the codes: FS/C—Front Support/Cover; LS/C—Lateral Support/Cover; BS/C—Back Support/Cover. On the top-left corner of each frame there is the synchronization (<math display="inline"><semantics> <mrow> <mover> <mi>r</mi> <mo>´</mo> </mover> </mrow> </semantics></math>) value and on the bottom-left corner, the team configuration code (TCC). The ball is represented by a white filled circle. Full article ">

17 pages, 2494 KiB

Open AccessArticle

No Free Lunch—Characterizing the Performance of 6TiSCH When Using Different Physical Layers

by Mina Rady, Quentin Lampin, Dominique Barthel and Thomas Watteyne

Sensors 2020, 20(17), 4989; https://doi.org/10.3390/s20174989 - 3 Sep 2020

Cited by 10 | Viewed by 2617

Abstract

Low-power wireless applications require different trade-off points between latency, reliability, data rate and power consumption. Given such a set of constraints, which physical layer should I be using? We study this question in the context of 6TiSCH, a state-of-the-art recently standardized protocol stack [...] Read more.

Low-power wireless applications require different trade-off points between latency, reliability, data rate and power consumption. Given such a set of constraints, which physical layer should I be using? We study this question in the context of 6TiSCH, a state-of-the-art recently standardized protocol stack developed for harsh industrial applications. Specifically, we augment OpenWSN, the reference 6TiSCH open-source implementation, to support one of three physical layers from the IEEE802.15.4g standard: FSK 868 MHz which offers long range, OFDM 868 MHz which offers high data rate, and O-QPSK 2.4 GHz which offers more balanced performance. We run the resulting firmware on the 42-mote OpenTestbed deployed in an office environment, once for each physical layer. Performance results show that, indeed, no physical layer outperforms the other for all metrics. This article argues for combining the physical layers, rather than choosing one, in a generalized 6TiSCH architecture in which technology-agile radio chips (of which there are now many) are driven by a protocol stack which chooses the most appropriate physical layer on a frame-by-frame basis. Full article

(This article belongs to the Section Sensor Networks)

► Show Figures

Figure 1

14 pages, 4950 KiB

Open AccessLetter

Characteristics Research of a High Sensitivity Piezoelectric MOSFET Acceleration Sensor

by Chunpeng Ai, Xiaofeng Zhao and Dianzhong Wen

Sensors 2020, 20(17), 4988; https://doi.org/10.3390/s20174988 - 3 Sep 2020

Cited by 10 | Viewed by 4648

Abstract

In order to improve the output sensitivity of the piezoelectric acceleration sensor, this paper proposed a high sensitivity acceleration sensor based on a piezoelectric metal oxide semiconductor field effect transistor (MOSFET). It is constituted by a piezoelectric beam and an N-channel depletion MOSFET. [...] Read more.

In order to improve the output sensitivity of the piezoelectric acceleration sensor, this paper proposed a high sensitivity acceleration sensor based on a piezoelectric metal oxide semiconductor field effect transistor (MOSFET). It is constituted by a piezoelectric beam and an N-channel depletion MOSFET. A silicon cantilever beam with Pt/ZnO/Pt/Ti multilayer structure is used as a piezoelectric beam. Based on the piezoelectric effect, the piezoelectric beam generates charges when it is subjected to acceleration. Due to the large input impedance of the MOSFET, the charge generated by the piezoelectric beam can be used as a gate control signal to achieve the purpose of converting the output charge of the piezoelectric beam into current. The test results show that when the external excitation acceleration increases from 0.2 g to 1.5 g with an increment of 0.1 g, the peak-to-peak value of the output voltage of the proposed sensors increases from 0.327 V to 2.774 V at a frequency of 1075 Hz. The voltage sensitivity of the piezoelectric beam is 0.85 V/g and that of the proposed acceleration sensor was 2.05 V/g, which is 2.41 times higher than the piezoelectric beam. The proposed sensor can effectively improve the voltage output sensitivity and can be used in the field of structural health monitoring. Full article

(This article belongs to the Special Issue MEMS and NEMS Sensors)

► Show Figures

Figure 1

14 pages, 2847 KiB

Open AccessArticle

Predicting Advanced Balance Ability and Mobility with an Instrumented Timed Up and Go Test

by Ronny Bergquist, Corinna Nerz, Kristin Taraldsen, Sabato Mellone, Espen A.F. Ihlen, Beatrix Vereijken, Jorunn L. Helbostad, Clemens Becker and A. Stefanie Mikolaizak

Sensors 2020, 20(17), 4987; https://doi.org/10.3390/s20174987 - 3 Sep 2020

Cited by 17 | Viewed by 4074

Abstract

Extensive test batteries are often needed to obtain a comprehensive picture of a person’s functional status. Many test batteries are not suitable for active and healthy adults due to ceiling effects, or require a lot of space, time, and training. The Community Balance [...] Read more.

Extensive test batteries are often needed to obtain a comprehensive picture of a person’s functional status. Many test batteries are not suitable for active and healthy adults due to ceiling effects, or require a lot of space, time, and training. The Community Balance and Mobility Scale (CBMS) is considered a gold standard for this population, but the test is complex, as well as time- and resource intensive. There is a strong need for a faster, yet sensitive and robust test of physical function in seniors. We sought to investigate whether an instrumented Timed Up and Go (iTUG) could predict the CBMS score in 60 outpatients and healthy community-dwelling seniors, where features of the iTUG were predictive, and how the prediction of CBMS with the iTUG compared to standard clinical tests. A partial least squares regression analysis was used to identify latent components explaining variation in CBMS total score. The model with iTUG features was able to predict the CBMS total score with an accuracy of 85.2% (84.9–85.5%), while standard clinical tests predicted 82.5% (82.2–82.8%) of the score. These findings suggest that a fast and easily administered iTUG could be used to predict CBMS score, providing a valuable tool for research and clinical care. Full article

(This article belongs to the Collection Sensors for Gait, Human Movement Analysis, and Health Monitoring)

► Show Figures

Figure 1

Figure 1
Illustration of the test set-up for the Timed Up and Go (iTUG). Full article ">Figure 2
The 3-axis acceleration (upper) and angular velocity (lower) sensor signals recorded during five repetitions of an iTUG for one subject. The task is segmented into five phases separated by the green vertical lines. Full article ">Figure 3
Schematic presentation of the data that were used as predictor and response variables in the two separate partial least squares regression (PLSR) models presented in this paper. Model 1 included descriptive data and iTUG features, while model two included descriptive data and standard clinical tests. The Community Balance and Mobility Scale (CBMS) scores was used as a response variable. Full article ">Figure 4
(A) Predicted vs. measured CBMS total score from iTUG for outpatients and community-dwellers; (B) Mean root mean square error of prediction (RMSEP) +/− one standard deviation across 10 components. Full article ">Figure 5
The VIP scores (light grey) and R2 (dark grey) of the iTUG features selected in the PLSR model. The horizontal lines represent the lower (0.83), middle (1) and upper (1.21) cut-off values used for interpreting the VIP of individual predictor variables. Full article ">Figure 6
(A) Predicted vs. measured CBMS total score from standard clinical tests for outpatients and community-dwellers; (B) mean RMSEP +/− one standard deviation across eight components. Full article ">Figure 7
The VIP scores (light grey) and R2 (dark grey) of the clinical variables selected in the PLSR model. The horizontal lines represent the lower (0.83), middle (1), and upper (1.21) cut-off values used for interpreting the VIP of individual predictor variables. Full article ">

Journal Menu

Journal Browser

Sensors, Volume 20, Issue 17 (September-1 2020) – 360 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI