Review on Human Action Recognition in Smart Living: Sensing Technology, Multimodality, Real-Time Processing, Interoperability, and Resource-Constrained Processing.

Smart living, a concept that has gained increasing attention in recent years, revolves around integrating advanced technologies in homes and cities to enhance the quality of life for citizens. Sensing and human action recognition are crucial aspects of this concept. Smart living applications span various domains, such as energy consumption, healthcare, transportation, and education, which greatly benefit from effective human action recognition. This field, originating from computer vision, seeks to recognize human actions and activities using not only visual data but also many other sensor modalities. This paper comprehensively reviews the literature on human action recognition in smart living environments, synthesizing the main contributions, challenges, and future research directions. This review selects five key domains, i.e., Sensing Technology, Multimodality, Real-time Processing, Interoperability, and Resource-Constrained Processing, as they encompass the critical aspects required for successfully deploying human action recognition in smart living. These domains highlight the essential role that sensing and human action recognition play in successfully developing and implementing smart living solutions. This paper serves as a valuable resource for researchers and practitioners seeking to further explore and advance the field of human action recognition in smart living.

Human Postures Recognition by Accelerometer Sensor and ML Architecture Integrated in Embedded Platforms: Benchmarking and Performance Evaluation.

Embedded hardware systems, such as wearable devices, are widely used for health status monitoring of ageing people to improve their well-being. In this context, it becomes increasingly important to develop portable, easy-to-use, compact, and energy-efficient hardware-software platforms, to enhance the level of usability and promote their deployment. With this purpose an automatic tri-axial accelerometer-based system for postural recognition has been developed, useful in detecting potential inappropriate behavioral habits for the elderly. Systems in the literature and on the market for this type of analysis mostly use personal computers with high computing resources, which are not easily portable and have high power consumption. To overcome these limitations, a real-time posture recognition Machine Learning algorithm was developed and optimized that could perform highly on platforms with low computational capacity and power consumption. The software was integrated and tested on two low-cost embedded platform (Raspberry Pi 4 and Odroid N2+). The experimentation stage was performed on various Machine Learning pre-trained classifiers using data of seven elderly users. The preliminary results showed an activity classification accuracy of about 98% for the four analyzed postures (Standing, Sitting, Bending, and Lying down), with similar accuracy and a computational load as the state-of-the-art classifiers running on personal computers.

Multi-Sensor Platform for Predictive Air Quality Monitoring.

Air quality monitoring is a very important aspect of providing safe indoor conditions, and carbon dioxide (CO2) is one of the pollutants that most affects people's health. An automatic system able to accurately forecast CO2 concentration can prevent a sudden rise in CO2 levels through appropriate control of heating, ventilation and air-conditioning (HVAC) systems, avoiding energy waste and ensuring people's comfort. There are several works in the literature dedicated to air quality assessment and control of HVAC systems; the performance maximisation of such systems is typically achieved using a significant amount of data collected over a long period of time (even months) to train the algorithm. This can be costly and may not respond to a real scenario where the habits of the house occupants or the environment conditions may change over time. To address this problem, an adaptive hardware-software platform was developed, following the IoT paradigm, with a high level of accuracy in forecasting CO2 trends by analysing only a limited window of recent data. The system was tested considering a real case study in a residential room used for smart working and physical exercise; the parameters analysed were the occupants' physical activity, temperature, humidity and CO2 in the room. Three deep-learning algorithms were evaluated, and the best result was obtained with the Long Short-Term Memory network, which features a Root Mean Square Error of about 10 ppm with a training period of 10 days.

Benchmarking of Contactless Heart Rate Measurement Systems in ARM-Based Embedded Platforms.

Heart rate monitoring is especially important for aging individuals because it is associated with longevity and cardiovascular risk. Typically, this vital parameter can be measured using wearable sensors, which are widely available commercially. However, wearable sensors have some disadvantages in terms of acceptability, especially when used by elderly people. Thus, contactless solutions have increasingly attracted the scientific community in recent years. Camera-based photoplethysmography (also known as remote photoplethysmography) is an emerging method of contactless heart rate monitoring that uses a camera and a processing unit on the hardware side, and appropriate image processing methodologies on the software side. This paper describes the design and implementation of a novel pipeline for heart rate estimation using a commercial and low-cost camera as the input device. The pipeline's performance was tested and compared on a desktop PC, a laptop, and three different ARM-based embedded platforms (Raspberry Pi 4, Odroid N2+, and Jetson Nano). The results showed that the designed and implemented pipeline achieved an average accuracy of about 96.7% for heart rate estimation, with very low variance (between 1.5% and 2.5%) across processing platforms, user distances from the camera, and frame resolutions. Furthermore, benchmark analysis showed that the Odroid N2+ platform was the most convenient in terms of CPU load, RAM usage, and average execution time of the algorithmic pipeline.

Behavioral Change Prediction from Physiological Signals Using Deep Learned Features.

Predicting change from multivariate time series has relevant applications ranging from the medical to engineering fields. Multisensory stimulation therapy in patients with dementia aims to change the patient's behavioral state. For example, patients who exhibit a baseline of agitation may be paced to change their behavioral state to relaxed. This study aimed to predict changes in one's behavioral state from the analysis of the physiological and neurovegetative parameters to support the therapist during the stimulation session. In order to extract valuable indicators for predicting changes, both handcrafted and learned features were evaluated and compared. The handcrafted features were defined starting from the CATCH22 feature collection, while the learned ones were extracted using a temporal convolutional network, and the behavioral state was predicted through bidirectional long short-term memory auto-encoder, operating jointly. From the comparison with the state of the art, the learned features-based approach exhibits superior performance with accuracy rates of up to 99.42% with a time window of 70 seconds and up to 98.44% with a time window of 10 seconds.

Ambient and Wearable Sensor Technologies for Energy Expenditure Quantification of Ageing Adults.

COVID-19 has affected daily life in unprecedented ways, with dramatic changes in mental health, sleep time and level of physical activity. These changes have been especially relevant in the elderly population, with important health-related consequences. In this work, two different sensor technologies were used to quantify the energy expenditure of ageing adults. To this end, a technological platform based on Raspberry Pi 4, as an elaboration unit, was designed and implemented. It integrates an ambient sensor node, a wearable sensor node and a coordinator node that uses the information provided by the two sensor technologies in a combined manner. Ambient and wearable sensors are used for the real-time recognition of four human postures (standing, sitting, bending and lying down), walking activity and for energy expenditure quantification. An important first aim of this work was to realize a platform with a high level of user acceptability. In fact, through the use of two unobtrusive sensors and a low-cost processing unit, the solution is easily accessible and usable in the domestic environment; moreover, it is versatile since it can be used by end-users who accept being monitored by a specific sensor. Another added value of the platform is the ability to abstract from sensing technologies, as the use of human posture and walking activity for energy expenditure quantification enables the integration of a wide set of devices, provided that they can reproduce the same set of features. The obtained results showed the ability of the proposed platform to automatically quantify energy expenditure, both with each sensing technology and with the combined version. Specifically, for posture and walking activity classification, an average accuracy of 93.8% and 93.3% was obtained, respectively, with the wearable and ambient sensor, whereas an improvement of approximately 4% was reached using data fusion. Consequently, the estimated energy expenditure quantification always had a relative error of less than 3.2% for each end-user involved in the experimentation stage, classifying the high level information (postures and walking activities) with the combined version of the platform, justifying the proposed overall architecture from a hardware and software point of view.

Comparative Analysis of Supervised Classifiers for the Evaluation of Sarcopenia Using a sEMG-Based Platform.

Sarcopenia is a geriatric condition characterized by a loss of strength and muscle mass, with a high impact on health status, functional independence and quality of life in older adults. [d=TT, ]To reduce the effects of the disease, just the diagnostic is not enough, it is necessary more than recognition.To reduce the effects of the disease, it is important to recognize the level and progression of sarcopenia early. Surface electromyography is becoming increasingly relevant for the prevention and diagnosis of sarcopenia, also due to a wide diffusion of smart and minimally invasive wearable devices suitable for electromyographic monitoring. The purpose of this work is manifold. The first aim is the design and implementation of a hardware/software platform. It is based on the elaboration of surface electromyographic signals extracted from the Gastrocnemius Lateralis and Tibialis Anterior muscles, useful to analyze the strength of the muscles with the purpose of distinguishing three different "confidence" levels of sarcopenia. The second aim is to compare the efficiency of state of the art supervised classifiers in the evaluation of sarcopenia. The experimentation stage was performed on an "augmented" dataset starting from data acquired from 32 patients. The latter were distributed in an unbalanced manner on 3 "confidence" levels of sarcopenia. The obtained results in terms of classification accuracy demonstrated the ability of the proposed platform to distinguish different sarcopenia "confidence" levels, with highest accuracy value given by Support Vector Machine classifier, outperforming the other classifiers by an average of 7.7%.

Remaining Useful Life Prediction from 3D Scan Data with Genetically Optimized Convolutional Neural Networks.

In the current industrial landscape, increasingly pervaded by technological innovations, the adoption of optimized strategies for asset management is becoming a critical key success factor. Among the various strategies available, the "Prognostics and Health Management" strategy is able to support maintenance management decisions more accurately, through continuous monitoring of equipment health and "Remaining Useful Life" forecasting. In the present study, convolutional neural network-based deep neural network techniques are investigated for the remaining useful life prediction of a punch tool, whose degradation is caused by working surface deformations during the machining process. Surface deformation is determined using a 3D scanning sensor capable of returning point clouds with micrometric accuracy during the operation of the punching machine, avoiding both downtime and human intervention. The 3D point clouds thus obtained are transformed into bidimensional image-type maps, i.e., maps of depths and normal vectors, to fully exploit the potential of convolutional neural networks for extracting features. Such maps are then processed by comparing 15 genetically optimized architectures with the transfer learning of 19 pretrained models, using a classic machine learning approach, i.e., support vector regression, as a benchmark. The achieved results clearly show that, in this specific case, optimized architectures provide performance far superior (MAPE = 0.058) to that of transfer learning, which, instead, remains at a lower or slightly higher level (MAPE = 0.416) than support vector regression (MAPE = 0.857).

Vision-Based Road Rage Detection Framework in Automotive Safety Applications.

Drivers' road rage is among the main causes of road accidents. Each year, it contributes to more deaths and injuries globally. In this context, it is important to implement systems that can supervise drivers by monitoring their level of concentration during the entire driving process. In this paper, a module for Advanced Driver Assistance System is used to minimise the accidents caused by road rage, alerting the driver when a predetermined level of rage is reached, thus increasing the transportation safety. To create a system that is independent of both the orientation of the driver's face and the lighting conditions of the cabin, the proposed algorithmic pipeline integrates face detection and facial expression classification algorithms capable of handling such non-ideal situations. Moreover, road rage of the driver is estimated through a decision-making strategy based on the temporal consistency of facial expressions classified as "anger" and "disgust". Several experiments were executed to assess the performance on both a real context and three standard benchmark datasets, two of which containing non-frontal-view facial expression and one which includes facial expression recorded from participants during driving. Results obtained show that the proposed module is competent for road rage estimation through facial expression recognition on the condition of multi-pose and changing in lighting conditions, with the recognition rates that achieve state-of-art results on the selected datasets.

Modeling, Fabrication and Integration of Wearable Smart Sensors in a Monitoring Platform for Diabetic Patients.

The monitoring of some parameters, such as pressure loads, temperature, and glucose level in sweat on the plantar surface, is one of the most promising approaches for evaluating the health state of the diabetic foot and for preventing the onset of inflammatory events later degenerating in ulcerative lesions. This work presents the results of sensors microfabrication, experimental characterization and FEA-based thermal analysis of a 3D foot-insole model, aimed to advance in the development of a fully custom smart multisensory hardware-software monitoring platform for the diabetic foot. In this system, the simultaneous detection of temperature-, pressure- and sweat-based glucose level by means of full custom microfabricated sensors distributed on eight reading points of a smart insole will be possible, and the unit for data acquisition and wireless transmission will be fully integrated into the platform. Finite element analysis simulations, based on an accurate bioheat transfer model of the metabolic response of the foot tissue, demonstrated that subcutaneous inflamed lesions located up to the muscle layer, and ischemic damage located not below the reticular/fat layer, can be successfully detected. The microfabrication processes and preliminary results of functional characterization of flexible piezoelectric pressure sensors and glucose sensors are presented. Full custom pressure sensors generate an electric charge in the range 0-20 pC, proportional to the applied load in the range 0-4 N, with a figure of merit of 4.7 ± 1 GPa. The disposable glucose sensors exhibit a 0-6 mM (0-108 mg/dL) glucose concentration optimized linear response (for sweat-sensing), with a LOD of 3.27 µM (0.058 mg/dL) and a sensitivity of 21 µA/mM cm2 in the PBS solution. The technical prerequisites and experimental sensing performances were assessed, as preliminary step before future integration into a second prototype, based on a full custom smart insole with enhanced sensing functionalities.

Thin Film Encapsulation for RF MEMS in 5G and Modern Telecommunication Systems.

In this work, SiNx/a-Si/SiNx caps on conductive coplanar waveguides (CPWs) are proposed for thin film encapsulation of radio-frequency microelectromechanical systems (RF MEMS), in view of the application of these devices in fifth generation (5G) and modern telecommunication systems. Simplification and cost reduction of the fabrication process were obtained, using two etching processes in the same barrel chamber to create a matrix of holes through the capping layer and to remove the sacrificial layer under the cap. Encapsulating layers with etch holes of different size and density were fabricated to evaluate the removal of the sacrificial layer as a function of the percentage of the cap perforated area. Barrel etching process parameters also varied. Finally, a full three-dimensional finite element method-based simulation model was developed to predict the impact of fabricated thin film encapsulating caps on RF performance of CPWs.

A Low-Cost Breath Analyzer Module in Domiciliary Non-Invasive Mechanical Ventilation for Remote COPD Patient Monitoring.

Smart Breath Analyzers were developed as sensing terminals of a telemedicine architecture devoted to remote monitoring of patients suffering from Chronic Obstructive Pulmonary Disease (COPD) and home-assisted by non-invasive mechanical ventilation via respiratory face mask. The devices based on different sensors (CO2/O2 and Volatile Organic Compounds (VOCs), relative humidity and temperature (R.H. & T) sensors) monitor the breath air exhaled into the expiratory line of the bi-tube patient breathing circuit during a noninvasive ventilo-therapy session; the sensor raw signals are transmitted pseudonymized to National Health Service units by TCP/IP communication through a cloud remote platform. The work is a proof-of-concept of a sensors-based IoT system with the perspective to check continuously the effectiveness of therapy and/or any state of exacerbation of the disease requiring healthcare. Lab tests in controlled experimental conditions by a gas-mixing bench towards CO2/O2 concentrations and exhaled breath collected in a sampling bag were carried out to test the realized prototypes. The Smart Breath Analyzers were also tested in real conditions both on a healthy volunteer subject and a COPD suffering patient.

AI-Based Early Change Detection in Smart Living Environments.

In the smart environments we live today, a great variety of heterogeneous sensors are being increasingly deployed with the aim of providing more and more value-added services. This huge availability of sensor data, together with emerging Artificial Intelligence (AI) methods for Big Data analytics, can yield a wide array of actionable insights to help older adults continue to live independently with minimal support of caregivers. In this regard, there is a growing demand for technological solutions able to monitor human activities and vital signs in order to early detect abnormal conditions, avoiding the caregivers' daily check of the care recipient. The aim of this study is to compare state-of-the-art machine and deep learning techniques suitable for detecting early changes in human behavior. At this purpose, specific synthetic data are generated, including activities of daily living, home locations in which such activities take place, and vital signs. The achieved results demonstrate the superiority of unsupervised deep-learning techniques over traditional supervised/semi-supervised ones in terms of detection accuracy and lead-time of prediction.

Chromatographic analysis of VOC patterns in exhaled breath from smokers and nonsmokers.

Cigarette smoking harms nearly every organ of the body and causes many diseases. The analysis of exhaled breath for exogenous and endogenous volatile organic compounds (VOCs) can provide fundamental information on active smoking and insight into the health damage that smoke is creating. Various exhaled VOCs have been reported as typical of smoking habit and recent tobacco consumption, but to date, no eligible biomarkers have been identified. Aiming to identify such potential biomarkers, in this pilot study we analyzed the chemical patterns of exhaled breath from 26 volunteers divided into groups of nonsmokers and subgroups of smokers sampled at different periods of withdrawal from smoking. Solid-phase microextraction technique and gas chromatography/mass spectrometry methods were applied. Many breath VOCs were identified and quantified in very low concentrations (ppbv range), but only a few (toluene, pyridine, pyrrole, benzene, 2-butanone, 2-pentanone and 1-methyldecyclamine) were found to be statistically significant variables by Mann-Whitney test. In our analysis, we did not consider the predictive power of individual VOCs, as well as the criterion of uniqueness for biomarkers suggests, but we used the patterns of the only statistically significant compounds. Probit prediction model based on statistical relevant VOCs-patterns showed that assessment of smoking status is heavily time dependent. In a two-class classifier model, it is possible to predict with high specificity and sensitivity if a subject is a smoker who respected 1 hour of abstinence from smoking (short-term exposure to tobacco) or a smoker (labelled "blank smoker") after a night out of smoking (long-term exposure to tobacco). On the other side, in our study "blank smokers" are more like non-smokers so that the two classes cannot be well distinguished and the corresponding prediction results showed a good sensitivity but low selectivity.

Improving holographic reconstruction by automatic Butterworth filtering for microelectromechanical systems characterization.

Digital holographic microscopy is an important interferometric tool in optical metrology allowing the investigation of engineered surfaces with microscale lateral resolution and nanoscale axial precision. In particular, microelectromechanical systems (MEMS) surface analysis, conducted by holographic characterization, requires high accuracy for functional testing. The main issues related to MEMS inspection are the superficial roughness and the complex geometry resulting from the several fabrication steps. Here, an automatic procedure, particularly suited in the case of high-roughness surfaces, is presented to selectively filter the spectrum, providing very low-noise reconstructed images. The numerical procedure is based on Butterworth filtering, and the obtained results demonstrate a significant increase in the images' quality and in the accuracy of the measurements, making our technique highly applicable for quantitative phase imaging in MEMS analysis. Furthermore, our method is fully tunable to the spectrum under investigation and automatic. This makes it highly suitable for real-time applications. Several experimental tests show the suitability of the proposed approach.

Eco-Friendly Production of Polyvinyl Alcohol/Carboxymethyl Cellulose Wound Healing Dressing Containing Sericin.

Wound dressing production represents an important segment in the biomedical healthcare field, but finding a simple and eco-friendly method that combines a natural compound and a biocompatible dressing production for biomedical application is still a challenge. Therefore, the aim of this study is to develop wound healing dressings that are environmentally friendly, low cost, and easily produced, using natural agents and a physical crosslinking technique. Hydrogel wound healing dressings were prepared from polyvinyl alcohol/carboxymethyl cellulose and sericin using the freeze-thawing method as a crosslinking method. The morphological characterization was carried out by scanning electron microscopy (SEM), whereas the mechanical analysis was carried out by dynamic mechanical analysis (DMA) to test the tensile strength and compression properties. Then, the healing property of the wound dressing material was tested by in vitro and ex vivo tests. The results show a three-dimensional microporous structure with no cytotoxicity, excellent stretchability with compressive properties similar to those of human skin, and excellent healing properties. The proposed hydrogel dressing was tested in vitro with HaCaT keratinocytes and ex vivo with epidermal tissues, demonstrating an effective advantage on wound healing acceleration. Accordingly, this study was successful in developing wound healing dressings using natural agents and a simple and green crosslinking method.

TEER and Ion Selective Transwell-Integrated Sensors System for Caco-2 Cell Model.

Monitoring of ions in real-time directly in cell culture systems and in organ-on-a-chip platforms represents a significant investigation tool to understand ion regulation and distribution in the body and ions' involvement in biological mechanisms and specific pathologies. Innovative flexible sensors coupling electrochemical stripping analysis (square wave anodic stripping voltammetry, SWASV) with an ion selective membrane (ISM) were developed and integrated in Transwell™ cell culture systems to investigate the transport of zinc and copper ions across a human intestinal Caco-2 cell monolayer. The fabricated ion-selective sensors demonstrated good sensitivity (1 × 10-11 M ion concentration) and low detection limits, consistent with pathophysiological cellular concentration ranges. A non-invasive electrochemical impedance spectroscopy (EIS) analysis, in situ, across a selected spectrum of frequencies (10-105 Hz), and an equivalent circuit fitting were employed to obtain useful electrical parameters for cellular barrier integrity monitoring. Transepithelial electrical resistance (TEER) data and immunofluorescent images were used to validate the intestinal epithelial integrity and the permeability enhancer effect of ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA) treatment. The proposed devices represent a real prospective tool for monitoring cellular and molecular events and for studies on gut metabolism/permeability. They will enable a rapid integration of these sensors into gut-on-chip systems.

A novel human biomonitoring study by semiconductor gas sensors in Exposomics: investigation of health risk in contaminated sites.

Two areas in central-southern Italy Land of Fires in Campania and Valley of Sacco river in Lazio are known to be contaminated sites, the first due to illegal fly-tipping and toxic fires, and the second due to an intensive industrial exploitation done by no-scruple companies and crooked public administration offices with dramatic consequences for environment and resident people. The work is intended to contribute to Human BioMonitoring (HBM) studies conducted in these areas on healthy young male population by a semiconductor gas sensor array trained by SPME-GC/MS. Human semen, blood and urine were investigated. The fingerprinting of the Volatile Organic Compounds (VOCs) by a gas sensors system allowed to discriminate the different contamination of the two areas and was able to predict the chemical concentration of several VOCs identified by GC/MS.

How Microgels Can Improve the Impact of Organ-on-Chip and Microfluidic Devices for 3D Culture: Compartmentalization, Single Cell Encapsulation and Control on Cell Fate.

The Organ-on-chip (OOC) devices represent the new frontier in biomedical research to produce micro-organoids and tissues for drug testing and regenerative medicine. The development of such miniaturized models requires the 3D culture of multiple cell types in a highly controlled microenvironment, opening new challenges in reproducing the extracellular matrix (ECM) experienced by cells in vivo. In this regard, cell-laden microgels (CLMs) represent a promising tool for 3D cell culturing and on-chip generation of micro-organs. The engineering of hydrogel matrix with properly balanced biochemical and biophysical cues enables the formation of tunable 3D cellular microenvironments and long-term in vitro cultures. This focused review provides an overview of the most recent applications of CLMs in microfluidic devices for organoids formation, highlighting microgels' roles in OOC development as well as insights into future research.

Deep transfer learning approaches for bleeding detection in endoscopy images.

Wireless capsule endoscopy is a non-invasive, wireless imaging tool that has developed rapidly over the last several years. One of the main limiting factors using this technology is that it produces a huge number of images, whose analysis, to be done by a doctor, is an extremely time-consuming process. In this research area, the management of this problem has been addressed with the development of Computer-aided Diagnosis systems thanks to which the automatic inspection and analysis of images acquired by the capsule has clearly improved. Recently, a big advance in classification of endoscopic images is achieved with the emergence of deep learning methods. The proposed expert system employs three pre-trained deep convolutional neural networks for feature extraction. In order to construct efficient feature sets, the features from VGG19, InceptionV3 and ResNet50 models are then selected and fused using the minimum Redundancy Maximum Relevance method and different fusion rules. Finally, supervised machine learning algorithms are employed to classify the images using the extracted features into two categories: bleeding and nonbleeding images. For performance evaluation a series of experiments are performed on two standard benchmark datasets. It has been observed that the proposed architecture outclass the single deep learning architectures, with an average accuracy in detection bleeding regions of 97.65 % and 95.70 % on well-known state-of-the-art datasets considering three different fusion rules, with the best combination in terms of accuracy and training time obtained using mean value pooling as fusion rule and Support Vector Machine as classifier.