On-body Sensors Research Articles

Deep Learning and Recurrent Signature Based Classification for Sensor-Based HAR: Addressing Explainability and Complexity in 5G Networks

When it comes to clinical applications, sensor-based human activity recognition (HAR) is invaluable, and numerous machine learning algorithms have effectively used to obtain excellent presentation. Using a variety of on-body sensors, these systems attempt to ascertain the subject's status relative to their immediate surroundings. There was a time when feature extraction was done by hand, but now more and more people are using Artificial Neural Networks (ANNs). A number of innovative approaches to HAR have surfaced since the advent of deep learning. Problems arise, however, for sensor-based HAR classification algorithms in today's communication networks. Among these, you can find solutions to problems like deal with complicated and large-scale data signals, extract characteristics from complicated datasets, and meet explainability standards. For complicated 5G networks, these difficulties become even more apparent. In particular, explainability is now critical for the broad use of sensor-based HAR in 5G networks and beyond. The research suggests a classification approach based on path signatures, recurrent signature (ReS), to address these issues. This cutting-edge model employs deep-learning (DL) approaches to circumvent the tedious feature selection challenge. Furthermore, the study investigates how to improve the ReS model's classification accuracy by using graph-based optimisation methods. To test how well the suggested framework worked, to dug deep into the publicly available dataset, which included a separate set of tasks. The paper's empirical results on AReM datasets achieved an average accuracy of 96%.

What Radio Waves Tell Us about Sleep!

The ability to assess sleep at home, capture sleep stages, and detect the occurrence of apnea (without on-body sensors) simply by analyzing the radio waves bouncing off people's bodies while they sleep is quite powerful. Such a capability would allow for longitudinal data collection in patients' homes, informing our understanding of sleep and its interaction with various diseases and their therapeutic responses, both in clinical trials and routine care. In this article, we develop an advanced machine learning algorithm for passively monitoring sleep and nocturnal breathing from radio waves reflected off people while asleep. Validation results in comparison with the gold standard (i.e., polysomnography) (n=880) demonstrate that the model captures the sleep hypnogram (with an accuracy of 80.5% for 30-second epochs categorized into Wake, Light Sleep, Deep Sleep, or REM), detects sleep apnea (AUROC = 0.89), and measures the patient's Apnea-Hypopnea Index (ICC=0.90; 95% CI = [0.88, 0.91]). Notably, the model exhibits equitable performance across race, sex, and age. Moreover, the model uncovers informative interactions between sleep stages and a range of diseases including neurological, psychiatric, cardiovascular, and immunological disorders. These findings not only hold promise for clinical practice and interventional trials but also underscore the significance of sleep as a fundamental component in understanding and managing various diseases.

Skin-interfaced microfluidic biosensors for colorimetric measurements of the concentrations of ketones in sweat.

Ketones, such as beta-hydroxybutyrate (BHB), are important metabolites that can be used to monitor for conditions such as diabetic ketoacidosis (DKA) and ketosis. Compared to conventional approaches that rely on samples of urine or blood evaluated using laboratory techniques, processes for monitoring of ketones in sweat using on-body sensors offer significant advantages. Here, we report a class of soft, skin-interfaced microfluidic devices that can quantify the concentrations of BHB in sweat based on simple and low-cost colorimetric schemes. These devices combine microfluidic structures and enzymatic colorimetric BHB assays for selective and accurate analysis. Human trials demonstrate the broad applicability of this technology in practical scenarios, and they also establish quantitative correlations between the concentration of BHB in sweat and in blood. The results represent a convenient means for managing DKA and aspects of personal nutrition/wellness.

On-body non-invasive glucose monitoring sensor based on high figure of merit (FoM) surface plasmonic microwave resonator

High-figure of merit (FoM) plasmonic microwave resonator is researched as a non-invasive on-body sensor to monitor the human body's blood glucose variation rate in adults for biomedical applications, e.g., diabetic patients. The resonance frequencies of the proposed sensor are measured to be around {f}_{01}=3.25 GHz and {f}_{01}=4.67 GHz over the frequency band of DC to 6GHz which are suitable for monitoring interstitial fluid (ISF) changing rate. The 40times 40 {mathrm{mm}}^{2} sensor is experimentally wrapped on the human body arm to monitor the blood glucose changing rate via amplitude and frequency variations of the sensor. Amplitude variation and frequency shift are measured to be around 7 dB and 30 MHz, respectively. The measured results demonstrate the high precision of the proposed approach to depict a valid diagram for glucose changing rate due to good impedance matching of the designed microwave sensor and human body. The sensor is shown to enhance the sensitivity by a factor of 5 compared to the conventional ones.

SmarCyPad

Cycling is an efficient and effective way to improve one's overall fitness level, such as cardiovascular fitness, stamina, lower body strength, and body fat percentage. To improve fitness performance, real-time cycling fitness tracking can not only allow cyclists to better control their energy outputs but also help push workout intensity and keep users accountable for their fitness progress. However, existing bike sensors (e.g., the ones mounted to bike's wheel hub or crank arm) are only limited to measuring cycling cadence and speed. Although several recent studies relying on on-body sensors or cameras can provide more fine-grained information (e.g., riding position and knee joint angle), they would either require inconvenient setups or raise serious privacy concerns. To circumvent these limitations, in this paper, we propose SmarCyPad, an innovative smart seat pad that can continuously and unobtrusively track five cycling-specific metrics, including cadence, per-leg stability, leg strength balance, riding position, and knee joint angle of the cyclist. Specifically, we embed conductive fabric sensors in the seat pad to sense the pressure applied to the bike's seat exerted by the cyclist's gluteal muscles. A series of signal processing algorithms are developed to estimate the pedaling period from the sensed pressure signal and further derive the cycling cadence, per-leg stability, and leg strength balance. Additionally, we leverage a deep learning model to detect the cyclist's riding position and reconstruct the cyclist's knee joint angles via linear regression. The sensors and the system prototype are manufactured from scratch leveraging off-the-shelf materials, and the total cost is less than $50. Extensive experiments involving 15 participants demonstrate that SmarCyPad can accurately estimate the cycling cadence with an average error of 1.13 rounds per minute, quantify the cycling stability for each leg, detect cycling imbalance, distinguish five riding positions with an accuracy of 96.60%, and continuously track the knee joint angle with an average mean error as low as 9.58 degrees.

Flexible Antenna with Circular/Linear Polarization for Wideband Biomedical Wireless Communication.

A wideband low-profile radiating G-shaped strip on a flexible substrate is proposed to operate as biomedical antenna for off-body communication. The antenna is designed to produce circular polarization over the frequency range 5-6 GHz to communicate with WiMAX/WLAN antennas. Furthermore, it is designed to produce linear polarization over the frequency range 6-19 GHz for communication with the on-body biosensor antennas. It is shown that an inverted G-shaped strip produces circular polarization (CP) of the opposite sense to that produced by G-shaped strip over the frequency range 5-6 GHz. The antenna design is explained and its performance is investigated through simulation, as well as experimental measurements. This antenna can be viewed as composed of a semicircular strip terminated with a horizontal extension at its lower end and terminated with a small circular patch through a corner-shaped strip extension at its upper end to form the shape of "G" or inverted "G". The purpose of the corner-shaped extension and the circular patch termination is to match the antenna impedance to 50 Ω over the entire frequency band (5-19 GHz) and to improve the circular polarization over the frequency band (5-6 GHz). To be fabricated on only one face of the flexible dielectric substrate, the antenna is fed through a co-planar waveguide (CPW). The antenna and the CPW dimensions are optimized to obtain the most optimal performance regarding the impedance matching bandwidth, 3dB Axial Ratio (AR) bandwidth, radiation efficiency, and maximum gain. The results show that the achieved 3dB-AR bandwidth is 18% (5-6 GHz). Thus, the proposed antenna covers the 5 GHz frequency band of the WiMAX/WLAN applications within its 3dB-AR frequency band. Furthermore, the impedance matching bandwidth is 117% (5-19 GHz) which enables low-power communication with the on-body sensors over this wide range of the frequency. The maximum gain and radiation efficiency are 5.37 dBi and 98%, respectively. The overall antenna dimensions are 25 × 27 × 0.13 mm3 and the bandwidth-dimension ratio (BDR) is 1733.

ViSig

Visual body signals are designated body poses that deliver an application-specific message. Such signals are widely used for fast message communication in sports (signaling by umpires and referees), transportation (naval officers and aircraft marshallers), and construction (signaling by riggers and crane operators), to list a few examples. Automatic interpretation of such signals can help maintaining safer operations in these industries, help in record-keeping for auditing or accident investigation purposes, and function as a score-keeper in sports. When automation of these signals is desired, it is traditionally performed from a viewer's perspective by running computer vision algorithms on camera feeds. However, computer vision based approaches suffer from performance deterioration in scenarios such as lighting variations, occlusions, etc., might face resolution limitations, and can be challenging to install. Our work, ViSig, breaks with tradition by instead deploying on-body sensors for signal interpretation. Our key innovation is the fusion of ultra-wideband (UWB) sensors for capturing on-body distance measurements, inertial sensors (IMU) for capturing orientation of a few body segments, and photodiodes for finger signal recognition, enabling a robust interpretation of signals. By deploying only a small number of sensors, we show that body signals can be interpreted unambiguously in many different settings, including in games of Cricket, Baseball, and Football, and in operational safety use-cases such as crane operations and flag semaphores for maritime navigation, with > 90% accuracy. Overall, we have seen substantial promise in this approach and expect a large body of future follow-on work to start using UWB and IMU fused modalities for the more general human pose estimation problems.

Challenges of Emerging Wearable Sensors for Remote Monitoring toward Telemedicine Healthcare.

Digitized telemedicine tools with the Internet of Things (IoT) started advancing into our daily lives and have been incorporated with commercial wearable gadgets for noninvasive remote health monitoring. The newly established tools have been steered toward a new era of decentralized healthcare. The advancement of a telemedicine wearable monitoring system has attracted enormous interest in the multimodal big data acquisition of real-time physiological and biochemical information via noninvasive methods for any health-related industries. The expectation of telemedicine wearable creation has been focused on early diagnosis of multiple diseases and minimizing the cost of high-tech and invasive treatments. However, only limited progress has been directed toward the development of telemedicine wearable sensors. This Perspective addresses the advancement of these wearable sensors that encounter multiple challenges on the forefront and technological gaps hampering the realization of health monitoring at molecular levels related to smart materials mostly limited to single use, issues of selectivity to analytes, low sensitivity to targets, miniaturization, and lack of artificial intelligence to perform multiple tasks and secure big data transfer. Sensor stability with minimized signal drift, on-body sensor reusability, and long-term continuous health monitoring provides key analytical challenges. This Perspective also focuses on, promotes, and highlights wearable sensors with a distinct capability to interconnect with telemedicine healthcare for physical sensing and multiplex sensing at deeper levels. Moreover, it points out some critical challenges in different material aspects and promotes what it will take to advance the current state-of-art wearable sensors for telemedicine healthcare. Ultimately, this Perspective is to draw attention to some potential blind spots of wearable technology development and to inspire further development of this integrated technology in mitigating multimorbidity in aging societies through health monitoring at molecular levels to identify signs of diseases.

Semi-Supervised Adversarial Auto-Encoder to Expedite Human Activity Recognition.

The study of human activity recognition concentrates on classifying human activities and the inference of human behavior using modern sensing technology. However, the issue of domain adaptation for inertial sensing-based human activity recognition (HAR) is still burdensome. The existing requirement of labeled training data for adapting such classifiers to every new person, device, or on-body location is a significant barrier to the widespread adoption of HAR-based applications, making this a challenge of high practical importance. We propose the semi-supervised HAR method to improve reconstruction and generation. It executes proper adaptation with unlabeled data without changes to a pre-trained HAR classifier. Our approach decouples VAE with adversarial learning to ensure robust classifier operation, without newly labeled training data, under changes to the individual activity and the on-body sensor position. Our proposed framework shows the empirical results using the publicly available benchmark dataset compared to state-of-art baselines, achieving competitive improvement for handling new and unlabeled activity. The result demonstrates SAA has achieved a 5% improvement in classification score compared to the existing HAR platform.

Contrastive Accelerometer–Gyroscope Embedding Model for Human Activity Recognition

Recently, the widespread use of smart devices has invoked more interest in sensor-based applications. Human activity recognition (HAR) with body-worn sensors is an underlying task that aims to recognize a person’s physical activity from on-body sensor readings. In this article, we address the HAR problem that utilizes accelerometers and gyroscopes. While deep-learning-based feature extraction methods are advancing, recent HAR systems with multimodal sensors mostly use data-level fusion, aggregating different sensor signals into one multichannel input. However, they neglect the fact that different sensors capture different physical properties and produce distinct patterns. In this article, we propose a two-stream convolutional neural network (CNN) model that processes the accelerometer and gyroscope signals separately. The modality-specific features are fused in feature-level and jointly used for the recognition task. Furthermore, we introduce a self-supervised learning (SSL) task that pairs the accelerometer and the gyroscope embeddings acquired from the same activity instance. This auxiliary objective allows the feature extractors of our model to communicate during training and exploit complementary information, achieving better representations for HAR. We name our end-to-end multimodal HAR system the Contrastive Accelerometer–Gyroscope Embedding (CAGE) model. CAGE outperforms preceding HAR models in four publicly available benchmarks. The code is available at github.com/quotation2520/CAGE4HAR.

An observational study to assess validity and reliability of smartphone sensor-based gait and balance assessments in multiple sclerosis: Floodlight GaitLab protocol.

Gait and balance impairments are often present in people with multiple sclerosis (PwMS) and have a significant impact on quality of life and independence. Gold-standard quantitative tools for assessing gait and balance such as motion capture systems and force plates usually require complex technical setups. Wearable sensors, including those integrated into smartphones, offer a more frequent, convenient, and minimally burdensome assessment of functional disability in a home environment. We developed a novel smartphone sensor-based application (Floodlight) that is being used in multiple research and clinical contexts, but a complete validation of this technology is still lacking. This protocol describes an observational study designed to evaluate the analytical and clinical validity of Floodlight gait and balance tests. Approximately 100 PwMS and 35 healthy controls will perform multiple gait and balance tasks in both laboratory-based and real-world environments in order to explore the following properties: (a) concurrent validity of the Floodlight gait and balance tests against gold-standard assessments; (b) reliability of Floodlight digital measures derived under different controlled gait and balance conditions, and different on-body sensor locations; (c) ecological validity of the tests; and (d) construct validity compared with clinician- and patient-reported assessments. The Floodlight GaitLab study (ISRCTN15993728) represents a critical step in the technical validation of Floodlight technology to measure gait and balance in PwMS, and will also allow the development of new test designs and algorithms.

Machine Learning Approach to Model Physical Fatigue during Incremental Exercise among Firefighters

Physical fatigue is a serious threat to the health and safety of firefighters. Its effects include decreased cognitive abilities and a heightened risk of accidents. Subjective scales and, recently, on-body sensors have been used to monitor physical fatigue among firefighters and safety-sensitive professionals. Considering the capabilities (e.g., noninvasiveness and continuous monitoring) and limitations (e.g., assessed fatiguing tasks and models validation procedures) of current approaches, this study aimed to develop a physical fatigue prediction model combining cardiorespiratory and thermoregulatory measures and machine learning algorithms within a firefighters' sample. Sensory data from heart rate, breathing rate and core temperature were recorded from 24 participants during an incremental running protocol. Various supervised machine learning algorithms were examined using 21 features extracted from the physiological variables and participants' characteristics to estimate four physical fatigue conditions: low, moderate, heavy and severe. Results showed that the XGBoosted Trees algorithm achieved the best outcomes with an average accuracy of 82% and accuracies of 93% and 86% for recognising the low and severe levels. Furthermore, this study evaluated different methods to assess the models' performance, concluding that the group cross-validation method presents the most practical results. Overall, this study highlights the advantages of using multiple physiological measures for enhancing physical fatigue modelling. It proposes a promising health and safety management tool and lays the foundation for future studies in field conditions.

A comprehensive ultra-wideband dataset for non-cooperative contextual sensing

Nowadays, an increasing amount of attention is being devoted towards passive and non-intrusive sensing methods. The prime example is healthcare applications, where on-body sensors are not always an option or in other applications which require the detection and tracking of unauthorized (non-cooperative) targets within a given environment. Therefore, in this paper we present a dataset consisting of measurements obtained from Radio-Frequency (RF) devices. Essentially, the dataset consists of Ultra-Wideband (UWB) data in the form of Channel Impulse Response (CIR), acquired via a Commercial Off-the-Shelf (COTS) UWB equipment. Approximately 1.6 hours of annotated measurements are provided, which are collected in a residential environment. This dataset can be used to passively track a target’s location in an indoor environment. Additionally, it can also be used to advance UWB-based Human Activity Recognition (HAR) since three basic human activities were recorded, namely, sitting, standing and walking. We anticipate that such datasets may be utilized to develop novel algorithms and methodologies for healthcare, smart homes and security applications.

Deep Learning-Based Energy Expenditure Estimation in Assisted and Non-Assisted Gait Using Inertial, EMG, and Heart Rate Wearable Sensors.

Energy expenditure is a key rehabilitation outcome and is starting to be used in robotics-based rehabilitation through human-in-the-loop control to tailor robot assistance towards reducing patients’ energy effort. However, it is usually assessed by indirect calorimetry which entails a certain degree of invasiveness and provides delayed data, which is not suitable for controlling robotic devices. This work proposes a deep learning-based tool for steady-state energy expenditure estimation based on more ergonomic sensors than indirect calorimetry. The study innovates by estimating the energy expenditure in assisted and non-assisted conditions and in slow gait speeds similarly to impaired subjects. This work explores and benchmarks the long short-term memory (LSTM) and convolutional neural network (CNN) as deep learning regressors. As inputs, we fused inertial data, electromyography, and heart rate signals measured by on-body sensors from eight healthy volunteers walking with and without assistance from an ankle-foot exoskeleton at 0.22, 0.33, and 0.44 m/s. LSTM and CNN were compared against indirect calorimetry using a leave-one-subject-out cross-validation technique. Results showed the suitability of this tool, especially CNN, that demonstrated root-mean-squared errors of 0.36 W/kg and high correlation (ρ > 0.85) between target and estimation ( = 0.79). CNN was able to discriminate the energy expenditure between assisted and non-assisted gait, basal, and walking energy expenditure, throughout three slow gait speeds. CNN regressor driven by kinematic and physiological data was shown to be a more ergonomic technique for estimating the energy expenditure, contributing to the clinical assessment in slow and robotic-assisted gait and future research concerning human-in-the-loop control.

In-Body to On-Body Channel Characterization and Modeling Based on Heterogeneous Human Models at HBC-UWB Band

In this work, the in-body to on-body (IB2OB) channel characteristics of the ultrawideband (UWB) link from 10 to 50 MHz at the human body communication (HBC) band are studied for developing high-speed and large-capacity intrabody wireless communication systems. Incorporating the time-domain finite integration technique (FIT) with three types of anatomical human models (adult male, adult female, and teenager) and a simple geometric human model with homogeneous or multilayer biological tissues, the 10–50-MHz signal propagation characteristics are investigated, and a comprehensive path loss model is proposed based on typical in-body links of the heart, brain, and intestine to the on-body sensor nodes. The statistical results show that a linear regulation term related to the surface wave propagation should be added to construct a more accurate logarithmic-distance path loss model in the sense of least-mean-square error (LMSE), indicating that the surface wave propagation mechanism can describe the 10–50-MHz signal propagation more accurately. The shadow fading in decibel is normally distributed around the mean path loss with a standard deviation in the range of 7–9 dB. And the average group delay of the IB2OB channel fluctuates around 0–7 ns, illustrating that the channel can be approximated as a linear phase response. Moreover, the experimental verification is conducted with a tissue-equivalent liquid phantom and an adult man, demonstrating the validity of our proposed statistical path loss model. The key parameters for channel characterization in simulations and measurements are summarized to provide useful insights for HBC-UWB communication system design and performance analysis.

Translational gaps and opportunities for medical wearables in digital health.

A confluence of advances in biosensor technologies, enhancements in health care delivery mechanisms, and improvements in machine learning, together with an increased awareness of remote patient monitoring, has accelerated the impact of digital health across nearly every medical discipline. Medical grade wearables-noninvasive, on-body sensors operating with clinical accuracy-will play an increasingly central role in medicine by providing continuous, cost-effective measurement and interpretation of physiological data relevant to patient status and disease trajectory, both inside and outside of established health care settings. Here, we review current digital health technologies and highlight critical gaps to clinical translation and adoption.

Congestive heart failure detection based on attention mechanism-enabled bi-directional long short-term memory model in the internet of medical things

As congestive heart failure (CHF) has become an important medical problem facing the world, the detection of the CHF is very important. Internet of medical things (IoMT) has been increasing popularity as its application and function in industry 4.0 era. IoMT systems provide an opportunity for innovations and cross-discipline applications in the medical fields, that can significantly improve the efficiency. With the help of IoMT systems, patients can have in-home and on-body sensors that monitor their vitals easily and constantly, which can provide massive data for detecting disease. However, the most important thing is to find a good model to the automatic detection for massive medical data collected by IoMT systems. In this case, the paper proposes attention mechanism-enabled bi-directional long short-term memory (ABLSTM) model based on Electrocardiogram (ECG) signals, to automatically detect the CHF in IoMT systems. This method makes full use of the model features that can record long-term and short-term signal information, as well as the functions of attention mechanism with adaptive learning in local features, and effectively extracts complex features of ECG signals and performs detection. ECG signal data is from two public datasets, to train and test the proposed ABLSTM model. At the same time, for the cases with noise and data differences, we propose a preprocessing process for ECG signals, and discuss the impact of different data segmentation methods on the model performance. The experimental results show that the proposed ABLSTM model has the highest accuracy rate with 96.6% for the CHF detection, which is higher than other four baseline methods. Therefore, this proposed method can achieve a good result in the detection of the CHF.

Priority-Based Resource Allocation and Energy Harvesting for WBAN Smart Health

With the emergence of new viral infections and the rapid spread of chronic diseases in recent years, the demand for integrated short-range wireless technologies is becoming a major bottleneck. Implementation of advanced medical telemonitoring and telecare systems for on-body sensors needs frequent recharging or battery replacement. This paper discusses a priority-based resource allocation scheme and smart channel assignment in a wireless body area network capable of energy harvesting. We investigate our transmission scheme in regular communication, where the access point transmits energy and command while the sensor simultaneously sends the information to the access point. A priority scheduling nonpreemptive algorithm to keep the process running for all the users to achieve the maximum reliability of access by the decision-maker or hub during critical situations of users has been proposed. During an emergency or critical situation, the process does not stop until the decision-maker or the hub takes a final decision. The objective of the proposed scheme is to get all the user processes executed with minimum average waiting time and no starvation. By allocating a higher priority to emergency and on data traffic signals such as critical and high-level signals, the proposed transmission scheme avoids inconsistent collisions. The results demonstrate that the proposed scheme significantly improves the quality of the network service in terms of data transmission for higher priority users.

Objective Measurement of Walking Activity Using Wearable Technologies in People with Parkinson Disease: A Systematic Review

Parkinson’s disease (PD) is a complex neurodegenerative disease with a multitude of disease variations including motor and non-motor symptoms. Quality of life and symptom management may be improved with physical activity. Due to technological advancement, development of small new wearable devices recently emerged and allowed objective measurement of walking activity in daily life. This review was specifically designed to synthesize literature on objective walking activity measurements using wearable devices of patients with PD. Inclusion criteria included patients with a diagnosis of PD and exclusion criteria included studies using animal models or mixed syndromes. Participants were not required to undergo any type of intervention and the studies must have reported at least one output that quantifies daily walking activity. Three databases were systematically searched with no limitation on publication date. Twenty-six studies were eligible and included in the systematic review. The most frequently used device was the ActiGraph GT3X which was used in 10 studies. Duration of monitoring presented a range from 8 h to one year. Nevertheless, 11 studies measured walking activity during a 7-day period. On-body sensor wearing location differed throughout the included studies showing eight positions, with the waist, ankle, and wrist being the most frequently used locations. The main procedures consisted of measurement of walking hours during a 2-day period or more, equipped with a triaxial accelerometer at the dominant hip or ankle. It is also important for further research to take care of different factors such as the population, their pathology, the period, and the environment.

UNIPD-BPE: Synchronized RGB-D and Inertial Data for Multimodal Body Pose Estimation and Tracking

The ability to estimate human motion without requiring any external on-body sensor or marker is of paramount importance in a variety of fields, ranging from human–robot interaction, Industry 4.0, surveillance, and telerehabilitation. The recent development of portable, low-cost RGB-D cameras pushed forward the accuracy of markerless motion capture systems. However, despite the widespread use of such sensors, a dataset including complex scenes with multiple interacting people, recorded with a calibrated network of RGB-D cameras and an external system for assessing the pose estimation accuracy, is still missing. This paper presents the University of Padova Body Pose Estimation dataset (UNIPD-BPE), an extensive dataset for multi-sensor body pose estimation containing both single-person and multi-person sequences with up to 4 interacting people. A network with 5 Microsoft Azure Kinect RGB-D cameras is exploited to record synchronized high-definition RGB and depth data of the scene from multiple viewpoints, as well as to estimate the subjects’ poses using the Azure Kinect Body Tracking SDK. Simultaneously, full-body Xsens MVN Awinda inertial suits allow obtaining accurate poses and anatomical joint angles, while also providing raw data from the 17 IMUs required by each suit. This dataset aims to push forward the development and validation of multi-camera markerless body pose estimation and tracking algorithms, as well as multimodal approaches focused on merging visual and inertial data.