IoT Technologies for HealthCare

Towards implementation of adult hearing screening tests that can be delivered via a mobile app, we have recently designed a novel speech-in-noise test based on the following requirements: user-operated, fast, reliable, accurate, viable for use by listeners of unknown native language and viable for testing at a distance. This study addresses specific models to (i) investigate the ability of the test to identify ears with mild hearing loss using machine learning; and (ii) address the range of the output levels generated using different transducers. Our results demonstrate that the test classification performance using decision tree models is in line with the performance of validated, language-dependent speech-in-noise tests. We observed, on average, 0.75 accuracy, 0.64 sensitivity and 0.81 specificity. Regarding the analysis of output levels, we demonstrated substantial variability of transducers’ characteristics and dynamic range, with headphones yielding higher output levels compared to earphones. These findings confirm the importance of a self-adjusted volume option. These results also suggest that earphones may not be suitable for test execution as the output levels may be relatively low, particularly for subjects with hearing loss or for those who skip the volume adjustment step. Further research is needed to fully address test performance, e.g. testing a larger sample of subjects, addressing different classification approaches, and characterizing test reliability in varying conditions using different devices and transducers.

Drug abuse among pregnant women and the subsequent neonatal illness are very crucial clinical and social problems. Drugs misuse during pregnancy places the mother and her baby at increased risk of severe complications including deformities, low birth weight, and mental disabilities. Pregnancy can motivate a woman to enter into an addiction treatment program to protect her unborn baby from the effects of drug misuse. Despite the availability of several treatment centers, many women do not seek needed help during and after pregnancy. Some of the reasons include stigmatization, fear of their babies being taken away by Child and Family Services, and the fear of confinement to a facility. In this paper, we propose a non-invasive Cloud-based Internet-of-Things (IoT) and Data Analytics framework that will provide support for women seeking addiction treatment during pregnancy. The system will use simplified sensors incorporated into a smartwatch to monitor, collect, and process vital data from pregnant women to identify instances of emergencies. During emergencies, the system automatically contacts specific needed service(s) and sends the processed data to the cloud for storage and Data Analytics to provide deeper insight and necessary decision making. The framework ensures that pregnant women are not confined into a facility and are reachable remotely by healthcare practitioners during addiction treatment. These capabilities guarantee that the system is operational during global pandemics like COVID-19. The framework integrates every patient’s data into a centralized database accessible to all healthcare practitioners thereby preventing multiple prescriptions of the same medication by different doctors.

The design of technological aids to assist older adults in their ageing process and to ensure proper attendance and care, despite the decreasing percentage of young people in the demographic profiles of many developed countries, requires the proper selection of sensing components, in order to come up with devices that can be easily used and integrated into everyday life. This paper addresses the metrological characterization of pressure sensors to be inserted into smart insoles aimed at monitoring the older adult’s physical activity levels. Two types of sensing elements are evaluated and a recommendation provided, based on the main requirement of designing a calibration-free insole: in this case, the pressure sensor should act as a switch, and the FSR 402 Short sensing element appears to be the proper solution to adopt.

In recent years, the application of IoT for health purposes, including the intense use of wearable devices, has been considerably growing. Among the wearable devices, the systems for measuring EMG (electromyography) signals are highly investigated. The possibility of recording different signals in a multichannel approach can lead to reliable data that can be used to improve diagnostic techniques, analyze performance in sports professionals and perform remote rehabilitation. In this work, we describe the design of a novel wearable system for surface EMG using a compact electronic board and a printed matrix of electrodes. The whole system has an estimated maximum current absorption of 55 mA at 3.3 V. We focused on the subsystem integration and on the real-time data transmission through Bluetooth Low Energy (BLE) with a throughput of 28 kB/s with a success rate of 99%. Some preliminary data are collected on a healthy man’s arm to validate the design. The acquired data are then analyzed and processed to improve information quality and extract contraction patterns.

Chronic Kidney Disease (CKD) is currently a worldwide chronic disease with an increasing incidence, prevalence and high cost to health systems. A delayed recognition and prevention often lead to a premature mortality due to progressive and incurable loss of kidney function. Data mining classifiers employment to discover patterns in CKD indicators would contribute to an early diagnosis that allow patients to prevent such kidney severe damage. Adopting the cross Industry Standard Process of Data Mining (CRISP-DM) methodology, this work develops a classifier model that would support healthcare professionals in early diagnosis of CKD patients. By building a data pipeline that manages the different phases of CRISP-DM, an automated data transformation, modelling and evaluation is applied to the CKD dataset extracted from the UCI ML repository. Moreover, the pipeline along with the Scikit-learn package’s GridSearchCV is used to carry out an exhaustive search of the best data mining classifier and the different parameters of the data preparation’s sub-stages like data missing and feature selection. Thus, AdaBoost is selected as the best classifier and it outperforms with a 100% in terms of accuracy, precision, sensivity, specificity, f1-score and roc auc, the classification results obtained by the related works reviewed. Moreover, the application of feature selection reduces up to 12 out of 24 features which are employed in the classifier model developed.

Using a wearable electroencephalogram (EEG) device, this paper introduces a novel method of quantifying and understanding the visual acuity of the human eye with the steady-state visually evoked potential (SSVEP) technique. This method gives users easy access to self-track and to monitor their eye health. The study focuses on how varying the SSVEP stimulus frequency and duration affect the overall representation of a person’s visual perception. The study proposes two methods for this visual representation. The first method is a hardware system that utilizes long-exposure photography to augment reality and collocate the visual map onto the plane of interest. The second is a software implementation that captures the visual field at a set distance. A three-dimensional mapping is created by gathering software-defined visual maps at various set distances. Preliminary results show that these methods can gain some insight into the user’s central vision, peripheral vision, and depth perception.

The problem of stress detection and classification has attracted a lot of attention in the past decade. It has been tackled with mainly two different approaches, where signals were either collected in ambulatory settings, which can be limited to the period of presence in the hospital, or in continuous mode in the field. A sensor-based continuous measurement of stress in daily life has a potential to increase awareness of patterns of stress occurrence. In this work, we first present a data-flow infrastructure suitable for two types of studies that conforms with the data protection requirements of the ethics committee monitoring the research on humans. The detection and binary classification of stress events is compared with three different machine learning models based on the features (meta-data) extracted from physiological signals acquired in laboratory conditions and ground-truth stress level information provided by the subjects themselves via questionnaires associated with these features. The main signals considered in current classification are electro-dermal activity (EDA) and blood volume pulse (BVP) signals. Different models are compared and the best configuration yields an \(F_1\) score of 0.71 (random baseline: 0.48). The importance on prediction of phasic and tonic EDA components is also investigated. Our results also pave the way for further work on this topic with both machine learning approaches and signal processing directions.

This work presents anxiety classification using physiological data, namely, EDA (eletrodermal activity) and HR (heart rate), collected with a sensing wrist-wearable device during a neutral baseline state condition. For this purpose, the WESAD public available dataset was used. The baseline condition was collected for around 20 min on 15 participants. Afterwards, to assess anxiety scores, the shortened 6-item STAI was filled by the participants. Using train and test sets with 70% and 30% of data, respectively, the proposed ensemble of 100 bagged classification trees obtained an overall accuracy of 95.7%. This, along with the high precision and recall obtained, reveal the good performance of the proposed classifier and support the ability of anxiety score classification using physiological data. Such a classification task can be integrated in a mobile application presenting coping strategies to deal and manage anxiety.

Emotions play a key role in everyday life of human beings, and since several years, researchers have investigated the physiological changes caused by external stimuli, looking for methods to automatically classify the emotional involvement of individuals. The Galvanic Skin Response, or ElectroDermal Activity, is one of the most interesting signals used in emotion research. In this preliminary study, a few participants were submitted to auditory stimuli (i.e., pleasant, neutral and unpleasant sounds) and their skin conductance signals were measured by means of a wireless and IoT-enabled wearable device, the Empatica E4. To investigate the impact of the emotional stimuli, data measured as emotion elicitation and retrieved from the Empatica cloud platform, was analysed in the time domain, showing that pleasant and neutral sounds do not produce evident effects, while listening to an unpleasant sound increases the subjective response, with higher impact when the sound duration is shorter. The preliminary outcomes obtained confirm great intra- and inter-subject variability that deserves further investigation, by involving a bigger population of test users.

Obesity is one of the most common problem that can be avoid with the correct education of the teenagers. There are different methods, but the use of the mobile devices to promote the creation of social challenges is important, because the teenagers act mainly in groups. The use of questionnaires, challenges and gamification purposes may promote the use of this type of mobile applications by teenagers. It is a special population that needs the adoption of different interactive technologies. The studies available are not validated by healthcare professionals. First of all, we started to analyze the related work of obesity problem, mobile applications, and different methodologies adopted with teenagers. By the end, seven students participated in the study with the performance of visualization of daily tips and curiosities, answering questionnaires, monitoring of physical activity and gamification. The teenagers were satisfied with the strategies adopted, but this study was affected by the pandemic situation around the world. In general, the participants were satisfied with the use of the mobile, and they would like to use it in the future for the improvement of their nutrition and physical activity habits.

Different wireless sensing methods have been proposed for acquisition and measurement of body signals. In medical healthcare, it is critical that data are received simultaneously, processed, and analyzed in order to diagnose the disease accurately. For instance, to detect a patient with sleep apnea, it is necessary for the biosignals from dozens of biosensors including electroencephalography (EEG), electrocardiogram (ECG), photoplethysmogram (PPG), and peripheral oxygen saturation (Sp\(O_2\)) to be received in sequence it is used for diagnosis. However, it is difficult to accurately received these signals as their measurement frequencies are different from each other. Precise synchronization of the heartbeat with other measuring cycles of each biosensor is a critical attribute for identifying the correlation of each biosignal. Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) used in existing body area networks to guarantee the precise synchronization of multi-biosignals. This paper addressed this issue by proposing a bio-inspired Dynamic Time Division Scheduling Protocol (D-TDSP) based on the Frog Calling Algorithm (FCA) to adjust the timing of data transmission and to guarantee the synchronization of the sensing and receiving of multi-biosignals. The accuracy of the proposed algorithm is compared with the CSMA/CA method using a TelosB and XM1000 sensor nodes.

Remote health monitoring and medication systems are becoming prevalent owing to the advances in sensing and connectivity technologies as well as the social and economical demands due to high health care costs as well as low availability of skilled health care providers. The significance of such devices and coordination are also highlighted in the context of recent pandemic outbreaks underlying the need for physical distancing as well as even lock-downs globally. Though such devices bring forth large scale benefits, being the safety critical nature of such applications, one has to be vigilant regarding the potential risk factors. Apart from the device and application level faults, ensuring the secure operation becomes paramount due to increased network connectivity of these systems and services. In this paper, we present a systematic approach for identification of cyber threats and vulnerabilities and how to mitigate them in the context of remote medication and monitoring devices. We specifically elaborate our approach and present the results using a case study of an electronic medication device.