Digital Health and Wireless Solutions

Serum lactate levels are considered a biomarker of tissue hypoxia. In sepsis or septic shock patients, as suggested by The Surviving Sepsis Campaign, early lactate clearance-directed therapy is associated with decreased mortality; thus, serum lactate levels should be assessed. Monitoring a patient’s vital parameters and repetitive blood analysis may have deleterious effects on the patient and also bring an economic burden. Machine learning and trend analysis are gaining importance to overcome these issues. In this context, we aimed to investigate if a machine learning approach can predict lactate trends from non-invasive parameters of patients with sepsis. This retrospective study analyzed adult sepsis patients in the Medical Information Mart for Intensive Care IV (MIMIC-IV) dataset. Inclusion criteria were two or more lactate tests within 6 h of diagnosis, an ICU stay of at least 24 h, and a change of ≥1 mmol/liter in lactate level. Naïve Bayes, J48 Decision Tree, Logistic Regression, Random Forest, and Logistic Model Tree (LMT) classifiers were evaluated for lactate trend prediction. LMT algorithm outperformed other classifiers (AUC = 0.803; AUPRC = 0.921). J48 decision tree performed worse than the other methods when predicting constant trend. LMT algorithm with four features (heart rate, oxygen saturation, initial lactate, and time interval variables) achieved 0.80 in terms of AUC (AUPRC = 0.921). We can say that machine learning models that employ logistic regression architectures, i.e., LMT algorithm achieved good results in lactate trend prediction tasks, and it can be effectively used to assess the state of the patient, whether it is stable or improving.

Image artefacts in computed tomography (CT) limit the diagnostic quality of the images. The objective of this proof-of-concept study was to apply deep learning (DL) for automated CT artefact classification. Openly available Head CT data from Johns Hopkins University was used. Three common artefacts (patient movement, beam hardening, and ring artefacts (RAs)) and artefact free images were simulated using 2D axial slices. Simulated data were split into a training set (Ntrain = 1040 × 4(4160)), two validation sets (Nval1 = 130 × 4(520) and Nval2 = 130 × 4(520)), and a separate test set (Ntest = 201 × 4(804); two individual subjects). VGG-16 model architecture was used as a DL classifier, and the Grad-CAM approach was used to produce attention maps. Model performance was evaluated using accuracy, average precision, area under the receiver operating characteristics (ROC) curve, precision, recall, and F1-score. Sensitivity analysis was performed for two test set slice images in which different RA radiuses (4 pixels to 245) and movement artefacts, i.e., head tilt with rotation angles (0.2° to 3°), were generated. Artefact classification performance was excellent on the test set, as accuracy, average precision, and ROC area under curve over all classes were 0.91, 0.86, and 0.99, respectively. The precision, recall, and F1-scores were over 0.84, 0.71, and 0.85 for all class-wise cases. Sensitivity analysis revealed that the model detected movement at all rotation angles, yet it failed to detect the smallest RAs (4-pixel radius). DL can be used for effective detection of CT artefacts. In future, DL could be applied for automated quality assurance of clinical CT.

The ability to regularly assess Parkinson’s disease (PD) symptoms outside of complex laboratories supports remote monitoring and better treatment management. Multimodal sensors are beneficial for sensing different motor and non-motor symptoms, but simultaneous analysis is difficult due to complex dependencies between different modalities and their different format and data properties. Multimodal machine learning models can analyze such diverse modalities together, thereby enhancing holistic understanding of the data and overall patient state. The Unified Parkinson’s Disease Rating Scale (UPDRS) is commonly used for PD symptoms severity assessment. This study proposes a Perceiver-based multimodal machine learning framework to predict UPDRS scores.We selected a gait dataset of 93 PD patients and 73 control subjects from the PhysioNet repository. This dataset includes two-minute walks from each participant using 16 Ground Reaction Force (GRF) sensors, placing eight on each foot. This experiment used both raw gait timeseries signals and extracted features from these GRF sensors. The Perceiver architecture’s hyperparameters were selected manually and through Genetic Algorithms (GA). The performance of the framework was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and linear Correlation Coefficient (CC).Our multimodal approach achieved a MAE of 2.23 ± 1.31, a RMSE of 5.75 ± 4.16 and CC of 0.93 ± 0.08 in predicting UPDRS scores, outperforming previous studies in terms of MAE and CC.This multimodal framework effectively integrates different data modalities, in this case illustrating by predicting UPDRS scores using sensor data. It can be applied to diverse decision support applications of similar natures where multimodal analysis is needed.

Patient pathway has become a key concept in the organization of healthcare. However, the materialization and operationalization of pathways often focus on work processes of health personnel, clinical decision-making, and deadlines, contradicting the strong patient-oriented perspective that is inherent in their definition. In this paper, we introduce a patient-centered perspective of kidney cancer pathways, reporting on a dual-perspective strategy to map and model patient pathways. Utilizing a multi-method approach, we map and model pathways from the perspectives of both healthcare personnel and patients and investigate the feasibility of the Customer Journey Modeling Language (CJML) for modeling patient pathways. To prevent confusion, the planned pathway as seen from the hospital perspective and the actual pathway experienced by the patient are referred to as ‘pathway’ and ‘journey’, respectively. In the paper, we describe methods to engage with healthcare professionals and patients to collect the necessary information to create precise models, and we show how precise modeling of patient pathways requires the integration of several information sources. Moreover, the study underlines the value of examining pathways from a dual perspective, as the two perspectives corroborate and supplement each other, illustrating the complexity of patient journeys. Finally, the findings provide insights into the feasibility of CJML, firstly underlining that the usefulness of visual models is context-dependent, and secondly, suggesting that the methods and subsequent visualizations may be useful as organizational, instructional, and communicative tools.

Digital services in healthcare and social services have increased due to national promotion and Covid19 pandemic. However, the regional differences may exist. Successful implementation and sustainability of digital services requires that attention is paid to addressing barriers and supporting facilitators at all levels in health care provision.The purpose of this study was to investigate the effects of employee status, form of organization and organizational size on the views related to current state and the role of digital services, development barriers, development plans and the support needed for development in welfare, social and health service organizations operating in the South Ostrobothnia region. The study was carried out in the era of exceptional circumstances created by the Covid19 pandemic in the summer of 2021. The study was a quantitative cross-sectional study using an electronic survey. Respondents (n = 121) were managers, entrepreneurs and employees of welfare, social and health service organizations operating in the South Ostrobothnia region.The results suggested that in more than four out of five welfare, social and health service organizations operating in the region of South Ostrobothnia, part of the services were already digital in the summer of 2021. These services had been extensively developed during the previous year, which was lived in exceptional circumstances caused by the Covid19 pandemic. Digital services were seen to function especially as enablers of customers in exceptional circumstances. However, managers or entrepreneurs also saw digital services as reaching new customers more important than employees. The acquisition of technology and human resources were felt to be the most significant barriers in the development of digital services, regardless of the employee status, form of organization and organization size. Regarding the use and development of digital services, information was felt to be necessary, especially about the characteristics of digital services, and financial support was also felt to be necessary for the development. However, the support needs were significant in many aspects related to digital service development. In particular, large organizations needed information on the cost-effectiveness of digital services.The results can be used to support welfare, social and health service organizations in digital service development.

The global burden of cardiovascular diseases (CVD) is a worldwide public health problem. In 2019, 18.6 million people died from CVD, representing a 17.1% increase compared to 2010. Also, some individuals who experience a cardiovascular event will require some form of cardiovascular procedure, such as a pacemaker or implantable cardioverter-defibrillator insertion, aneurysm repair, or heart valve replacement. Mobile health (mHealth) is a valuable tool for supporting individuals with CVD in self-management, providing medical recommendations, virtual consultations, reminders, and disease monitoring notifications. The main objective of this research was to enhance postoperative care for cardiac procedures. To achieve this, the research involved the development of a new mHealth application and the subsequent evaluation of its usability. The study constituted technological and usability research by using Design Science Research Methodology (DSRM). The design of the mobile application followed the principles of Persuasive Systems Design (PSD) model, which encompass a clear definition of the main task, user interaction through dialogue, system credibility, and social support, aiming to help change user behavior. The sample was non-probabilistic for convenience, and System Usability Scale (SUS) was applied to physicians and nurses as well as individuals in the information technology field. The sample comprised 18 participants, of whom 55.6% were female. The participants rated the application positively, with a median final SUS score of 95 (IQR 90–97.5). Finally, the mobile application presented high usability and user acceptance.

The objective of this study was to evaluate the effectiveness of robot-assisted lower-limb rehabilitation on balance in stroke patients and to explore the covariates associated with these effects.A systematic literature search was carried out in four databases (MEDLINE (Ovid), CINAHL, PsycINFO, and ERIC) for studies published from inception to 25th of March 2022. Studies on robot-assisted lower-limb rehabilitation with a randomized controlled trial (RCT) design, participants with stroke, a comparison group with conventional training, and balance-related outcomes were included. Studies were assessed for Cochrane Risk of Bias 2 and quality of evidence. Meta-analysis and meta-regression were performed.A total of 48 (RCT) with 1472 participants were included. The overall risk of bias in the included studies was unclear (n = 32), high (n = 15) or low (n = 1). Compared to conventional rehabilitation, robot-assisted lower-limb rehabilitation interventions were more effective for balance improvement (Hedges’ g = 0.25, 95% CI: 0.10 0.41). In meta-regression, a relationship between the training effect was observed with the time since stroke, explaining 56% of the variance (p = 0.001), and with the ankle robots, explaining 16% of the variance (p = 0.048). No serious adverse events related to robot-assisted training were reported.Robot-assisted lower-limb rehabilitation may improve balance more than conventional training in people with stroke, especially in the acute stage. Robot-assisted lower-limb rehabilitation seems to be a safe rehabilitation method for patients with stroke. To strengthen the evidence, more high-quality RCTs with adequate sample sizes are needed.

Children with attention and executive function disabilities often have a long-lasting need for rehabilitation to support their functional ability. Yet the availability of rehabilitation services is insufficient, regionally unevenly distributed, and unequal in terms of access to rehabilitation. There is a need for easily accessible services. In this paper, we present the VREALFUN project where the major aim is to develop a novel Virtual Reality (VR) rehabilitation method for children with deficits in attention and executive functions. This ongoing Randomized Control Study (RCT) includes two arms, one in children with attention deficit hyperactivity disorder (ADHD) and the other in children with mild to moderate traumatic brain injury (TBI).

In recent times, several studies have presented single-modality systems for non-contact biosignal monitoring. While these systems often yield estimations correlating with clinical-grade devices, their practicality is limited due to constraints in real-time processing, scalability, and interoperability. Moreover, these studies have seldom explored the combined use of multiple modalities or the integration of various sensors. Addressing these gaps, we introduce a distributed computing architecture designed to remotely acquire biosignals from both radars and cameras. This architecture is supported by conceptual blocks that distribute tasks across sensing, computing, data management, analysis, communication, and visualization. Emphasizing interoperability, our system leverages RESTful APIs, efficient video streaming, and standardized health-data protocols. Our framework facilitates the integration of additional sensors and improves signal analysis efficiency. While the architecture is conceptual, its feasibility has been evaluated through simulations targeting specific challenges in networked remote photoplethysmography (rPPG) systems. Additionally, we implemented a prototype to demonstrate the architectural principles in action, with modules and blocks operating in independent threads. This prototype specifically involves the analysis of biosignals using mmWave radars and RGB cameras, illustrating the potential for the architecture to be adapted into a fully distributed system for real-time biosignal processing.

This paper proposes a passive reconfigurable antenna by using liquid metal and gravity mechanism. When the antenna rotates to different angles, the liquid metal will flow to the lowest point of the container due to gravitational force, thereby acting as a reflector to redirect the main antenna radiation pattern at predefined angles at 5.8 GHz (ISM band). The upper section patterns can be maintained when antenna is tilted at 45°, 90°, 135°, 225°, 275° and 315° whereas the rotated antenna at 0° and 180° causes a forward radiation. More importantly, additional directors surrounding the main patch can be used to increase the directivity of these radiation patterns. Simulation results in terms of reflection coefficient, radiation patterns and gain indicate that the proposed antenna operated passively with a consistent bandwidth of approximately 400 MHz.

Harmonics is an unavoidable phenomenon, even before we knew about digital circuits. In our sleep study, we found harmonic artefacts (HA) in our functional near-infrared spectroscopy (fNIRS) signal. Interestingly, it was neither device- nor subject-dependent. The fundamental frequency was around either 0.5 Hz or 1 Hz. It appeared to be very sharp peaks and they were within the band of interest, i.e., respiratory (0.1–0.6 Hz) and cardiac (0.6–5 Hz) bands. Since the exact location might change, we proposed a skewness-based harmonic filter (sbHF) to identify the fundamental frequency and attenuate HA. Since suppressing certain frequencies may change signal characteristic, spectral entropy was used to evaluate it based on Wilcoxon-test at a 0.05 significant level. 25 controls (6 females, age: 39.0 ± 8.5 years, height: 175.6 ± 8.0 cm, weight: 80.3 ± 10.8 kg) and 16 sleep apnea patients (1 female, age: 48.3 ± 12.4 years, height: 177.3 ± 6.0 cm, weight: 93.6 ± 17.1 kg) were recruited for our sleep study. sbHF showed good performance to identify fundamental frequency and attenuate HA from our raw fNIRS signals and 5% of the signal experienced changes in signal characteristics based on the spectral entropy analysis. Combining sbHF with a certain motion artefact reduction, we found that specific order of operation to get appropriate chromophore concentration was needed. This method is not only for problems in wearable fNIRS, but also can be modified for other problems by adjusting the suspected area or sweeping the frequency range to identify a fundamental frequency.

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder with inattention, hyperactivity, and impulsivity as core symptoms. Current diagnostic methods of ADHD consisting of interviews and self-ratings come with a risk of subjective bias and are dependent on the limited availability of healthcare professionals. However, recent technological advances have opened new opportunities to develop objective and scalable methods for precision diagnostics. The present critical review covers the current literature concerning one of the promising technologies, the use of motion sensors or accelometers for detecting ADHD, particularly evaluating the related clinical potential. Several studies in this field, especially recent studies with advanced computational methods, have demonstrated excellent accuracy in detecting individual participants with ADHD. Machine learning methods provide several benefits in the analysis of rich sensor data, but the existing studies still have critical limitations in explaining the underlying cognitive functions and demonstrating the capacity for differential diagnostics is still underway. Clinical utility of sensor-based diagnostic methods could be improved by conducting rigorous cross-validation against other methods in representative samples and employing multi-sensor solutions with sophisticated analysis methods to improve interpretation of the symptom manifestation. We conclude that motion sensors provide cost-effective and easy-to-use solutions with strong potential to increase the precision and availability of ADHD diagnostics. Nevertheless, these methods should be employed with caution, as only a fraction of ADHD symptoms relate to hyperactivity captured by motion sensors. At best, this technique could complement the existing assessment methods or be used along with other digital tools such as virtual reality.

Modelling relation between Pulse Transit Time (PTT) and blood pressure (BP) is a critical step in BP estimation for wearable technology. Recognizing the limitation of assuming constant vessel and blood conditions, we developed a simplified pulsatile flow model to analyze how various factors affect PTT values. Our research focuses on the impact of mechanical characteristics, such as vessel diameter, wall thickness, blood viscosity, and pressure, on PTT measurements and subsequent BP estimation. Measurements were conducted using accelerometer sensors within a custom-designed mock circulatory loop. This setup allowed for the testing of a wide range of pressure values and pulsation rates, as well as the modification of viscosity in blood-mimicking liquids across different vessel models. We employed the Moens-Korteweg conversion model for pressure estimation, initially trained on PTT data from a specific setup parameter combination, and subsequently tested with data from varied setup parameters. We observed high correlation levels (r = 0.93 ± 0.09) paired with high error (RMSE = 163 ± 100 mHg), suggesting potential inaccuracies in pressure estimation. We present the recorded signals and discuss how alterations in physical conditions influence PTT values and the precision of BP estimation.

Retinal disorders, including diabetic retinopathy and macular degeneration due to aging, can lead to preventable blindness in diabetics. Vision loss caused by diseases that affect the retinal fundus cannot be reversed if not diagnosed and treated on time. This paper employs deep-learned feature extraction with ensemble learning models to improve the multi-disease classification of fundus images. This research presents a novel approach to the multi-classification of fundus images, utilizing deep-learned feature extraction techniques and ensemble learning to diagnose retinal disorders and diagnosing eye illnesses involving feature extraction, classification, and preprocessing of fundus images. The study involves analysis of deep learning and implementation of image processing. The ensemble learning classifiers have used retinal photos to increase the classification accuracy. The results demonstrate improved accuracy in diagnosing retinal disorders using DL feature extraction and ensemble learning models. The study achieved an overall accuracy of 87.2%, which is a significant improvement over the previous study. The deep learning models utilized in the study, including NASNetMobile, InceptionResNetV4, VGG16, and Xception, were effective in extracting relevant features from the Fundus images. The average F1-score for Extra Tree was 99%, while for Histogram Gradient Boosting and Random Forest, it was 98.8% and 98.4%, respectively. The results show that all three algorithms are suitable for the classification task. The combination of DenseNet feature extraction technique and RF, ET, and HG classifiers outperforms other techniques and classifiers. This indicates that using DenseNet for feature extraction can effectively enhance the performance of classifiers in the task of image classification.

The concern has been raised regarding errors in drugs prescription and medical diagnostics that need to be carefully thought through. Both patient diagnosis and medication prescription are the responsibilities of healthcare providers. As the number of people with health issues rises, the healthcare professionals’ burden is increased. Medical errors may occur in the healthcare sector as a result of healthcare professionals prescribing drugs medicines based on inadequate information related to patient history and drug side effects. Therefore, this study aims to propose a drug recommender system to assist healthcare providers in decision making when prescribing drugs for patients depending on their diagnoses. Drug reviews sentiments are analyzed to find the drug effectiveness among the users. Furthermore, the most suitable recommender algorithm for recommending drugs based on the data from healthcare professionals are selected for this study. Opinion mining is applied on drug reviews, and a hybrid method is implemented to overcome the limitations of content-based and collaborative filtering methods, such as the cold start problem and increasing client preference. The system is developed and tested successfully. The proposed system can assist healthcare professionals in drug decision making and sustain the whole digital care pathway for various diseases.

Accurate and early prediction of arrhythmias using Electrocardiograms (ECG) presents significant challenges due to the non-stationary nature of ECG signals and inter-patient variability, posing difficulties even for seasoned cardiologists. Deep Learning (DL) methods offer precision in identifying diagnostic ECG patterns for arrhythmias, yet they often lack the transparency needed for clinical application, thus hindering their broader adoption in healthcare. This study introduces an explainable DL-based prediction model using ECG signals to classify nine distinct arrhythmia categories. We evaluated various DL architectures, including ResNet, DenseNet, and VGG16, using raw ECG data. The ResNet34 model emerged as the most effective, achieving an Area Under the Receiver Operating Characteristic (AUROC) of 0.98 and an F1-score of 0.826. Additionally, we explored a hybrid approach that combines raw ECG signals with Heart Rate Variability (HRV) features. Our explainability analysis, utilizing the SHAP technique, identifies the most influential ECG leads for each arrhythmia type and pinpoints critical signal segments for individual disease prediction. This study emphasizes the importance of explainability in arrhythmia prediction models, a critical aspect often overlooked in current research, and highlights its potential to enhance model acceptance and utility in clinical settings.

Anomaly detection and fall prevention represent one of the key research areas within gait analysis for patients suffering from neurological disorders. Deep Learning has penetrated into healthcare applications, encompassing disease diagnosis and anomaly prediction. Connected wearable medical sensors are emerging due to computationally expensive machine learning tasks, which traditionally require use of remote PC or cloud computing. However, to reduce needs for wireless communication channel throughput, for data processing latency, and increase service reliability and safety, on device machine learning is gaining attention. This paper presents an innovative approach that leverages one dimensional convolutional neural network (1D-CNN) and long-short term memory (LSTM) neural network for the real-time detection of abnormal gait patterns during the step. Real-time anomaly detection pertains to the algorithm’s ability to promptly detect true gait abnormality occurrence during the swing phase of an ongoing step.For the experiments, we have collected eight different common gait anomalies, simulated by 22 persons, using motion sensors containing multidimensional inertial measurement units (IMUs).Results have demonstrated that the proposed 1D-CNN-AD algorithm achieves an average accuracy of 95% and an average F1-score of 88% for all gait types and can run in true real-time. Average earliness for 1D-CNN-AD algorithm was 0.6 s, which is mid-swing phase of the step. Proposed LSTM-AD algorithm achieved average accuracy of 87% and average F1-score of 70% for all gait types.

As people live longer, the incidence and severity of health problems increases, placing strain on healthcare systems. There is an urgent need for resource-wise approaches to healthcare. We present a system built using open-source tools that allows health and functional capacity data to be collected remotely. The app records performance on functional tests using the phone’s built-in camera and provides users with immediate feedback. Pose estimation is used to detect the user in the video. The x, y coordinates of key body landmarks are then used to compute further metrics such as joint angles and repetition durations. In a proof-of-concept study, we collected data from 13 patients who had recently undergone knee ligament or knee replacement surgery. Patients performed the sit-to-stand test twice, with an average difference in test duration of 1.12 s (range: 1.16–3.2 s). Y-coordinate locations allowed us to automatically identify repetition start and end times, while x, y coordinates were used to compute joint angles, a common rehabilitation outcome variable. Mean difference in repetition duration was 0.1 s (range: −0.4–0.4 s) between trials 1 and 2. Bland-Altman plots confirmed general test-retest consistency within participants. We present a mobile app that enables functional tests to be performed remotely and without supervision. We also demonstrate real-world feasibility, including the ability to automate the entire process, from testing to analysis and the provision of real-time feedback. This approach is scalable, and could form part of national health strategies, allowing healthcare providers to minimise the need for in-person appointments whilst yielding cost savings.

Digital transformation and digitalisation are rapidly affecting the society. The gradually increasing applications of different types of AI into solutions and services are welcome, but there are associated risks. These include, for example, within human aspects of care undermining fundamental rights, ethical considerations, sustainability, and policies and regulations. This change permeates every societal level, but it is especially evident in the healthcare sector due to the ageing population and shortage of professionals. This situation also places pressure on the development of competencies among healthcare professionals. A human-centered approach in design and design methods can promote the development of AI-based solutions in transdisciplinary and cross-disciplinary processes encompassing numerous stakeholders, scientific orientations, and perspectives. There is a need for research and evaluation of Human-Centered Design (HCD) processes and design methods to develop and gain more insights for future development.This study was conducted as research through design. It aimed to elucidate the application and insights gained from the adopted Service design process for AI-enabled services and HCD approach while developing AI-empowered solution, Voima-chatbot. One of this research's main conclusions and realization is the shift from purely HCD towards Life-Centered design of AI-enabled solutions with a human-in-the-loop. In addition, this project increased the understanding of the deep importance of having a transdisciplinary dialogue with developers during the process of developing digital well-being devices and combining different professional competencies to achieve the best working solutions.

New health technology assessment (HTA) models for digital health are continuously being developed and are already in use. In Finland, the HTA model for digital health, named Digi-HTA, has been employed since 2020. Internationally and also in Finland, the need for harmonization of these HTA models has been recognized. In order to harmonize the models, it is necessary to first identify the key features and requirements of existing models. In this study, three key assessment models for digital health identified as central in the Finnish context were analyzed. After the analysis, the results were compared to the Finnish Digi-HTA assessment model, and a final synthesis was created regarding the similarities and differences between the assessment models. The comparison includes German DiGA model, the global CEN-ISO/TS 82304-2:2021 technical specification, and the Nordic-designed NordDEC assessment model. There was a great deal of similarity in the evaluated models, although certain differences in emphasis were found. The key differences relate to reimbursement process, maturity of the assessment process and supported product categories as well as cost and effectiveness evaluation. The results of this study can be utilized in harmonizing assessment models for digital health.

Chronic diseases strain global healthcare economically, and integrating digital solutions are proposed to help in meeting the rising demand. Digital health interventions (DHIs) offer promise for personalized, and cost-effective health services, however, factors influencing their uptake remain unclear. We examined whether the probability of lifestyle DHI uptake varies among individuals with different educational levels and lifestyles, based on their attitudes and usage of e-services. We also examined the effect of sex and age, and the association between DHI uptake and both educational attainment and overall lifestyle. A possibility to start using a web-based lifestyle DHI was offered to a subgroup (n = 6978) of Healthy Finland survey participants and adjusted logistic regression models were used to investigate the factors affecting uptake. We found that higher education and healthier lifestyle, as indicated by lifestyle score, were related to higher odds of DHI uptake. However, the effects of age, sex, independence of e-service use, and competence to use online services varied across lifestyle score groups. No significant interactions were observed related to educational attainment. These results imply that lifestyle DHIs are less likely to reach individuals with less-healthy lifestyle habits and lower educational attainment. In addition, some predictors affected the uptake differently across lifestyle score groups, suggesting that implementations of DHIs might attempt strategies to optimize the participation rates in especially targeted subgroups.

Background: Digital preparation programs for day surgery are now available through smartphones; however, research on the effectiveness of digital interventions among parents is lacking.Aim: This study aimed to assess the effectiveness of a mobile application intervention in preparing parents for pediatric day surgery and to describe the correlations between parents’ anxiety, stress, and satisfaction.Methods: A total of 70 parents of preschool children who were scheduled for elective day surgery were randomly divided into two groups: the intervention group (IG; n = 36) and the control group (CG; n = 34). The study took place in the pediatric day surgical department of a university hospital in Finland. The IG used a mobile application, while the CG used routine methods. Parents’ anxiety, stress and satisfaction were measured using validated instruments.Results: There was no significant difference in parental anxiety levels between the two groups, both before and after the surgery. After the surgery, both groups of parents reported feeling less anxious while at home. Pre-surgery, most parents experienced no/mild stress at home. However, post-surgery, intervention group parents reported significantly less stress at home than control group parents. The mean VAS score for parents’ satisfaction in both groups was high: 8.8 for the intervention group (SD 1.9) and 8.6 for the control group (SD 0.9). These mean scores did not significantly differ. Anxiety, stress, and satisfaction showed a significant correlation in most cases at both T1 and T4.Conclusions: A mobile application can serve as an alternative to the traditional method of preparing parents for pediatric day surgery.

Evidence on the effects of robotic technology is required to develop rehabilitation services. This study aimed to evaluate the effects of robot-assisted walking training on walking and functional independence in everyday life in persons with spinal cord injury (SCI) and explore the covariates associated with these effects.We searched the MEDLINE (Ovid), CINAHL, PsycINFO, and ERIC databases until March 25, 2022. Two reviewers independently assessed the studies for inclusion. We included RCTs on people with SCI receiving robotic training. The Cochrane RoB2, meta-analysis, meta-regression, and Grading of Recommendations Assessment, Development, and Evaluation were performed.We included 23 RCTs focusing on SCI with outcomes of walking or functional independence, of which 14 were included in the meta-analysis and meta-regression analyses. Small improvements were observed in functional independence in favor of robot-assisted walking training compared to other physical exercises (Hedges’ g 0.31, 95% CI 0.02 to 0.59; I2 = 19.7%, 9 studies, 419 participants, low certainty evidence). There were no significant differences in walking ability, speed, endurance, or independence between the groups.Robot-assisted walking training may slightly improve functional independence, but its effects on walking ability in SCI patients is uncertain compared to other exercise. Evidence suggests little to no difference in walking independence, and the effects on walking speed and endurance are unclear. No clear evidence exists whether positive effects are linked to personal, clinical, or intervention characteristics. Robot-assisted gait training may be a viable option for improving functional independence in individuals with SCI.

This paper explores novel strategies to strengthen the security of Hybrid Wireless Body Area Networks (HyWBANs), which are essential in smart healthcare and Internet of Things (IoT) applications. Recognizing the vulnerability of HyWBAN to sophisticated cyber-attacks, we propose an innovative combination of semantic communications and jamming receivers. This dual-layered security mechanism protects against unauthorized access and data breaches, particularly in scenarios involving in-body to on-body communication channels. We conduct comprehensive laboratory measurements to understand hybrid (radio and optical) communication propagation through biological tissues. We utilize these insights to refine a dataset for training a Deep Learning (DL) model. These models, in turn, generate semantic concepts linked to cryptographic keys for enhanced data confidentiality and integrity using a jamming receiver. The proposed model significantly reduces energy consumption compared to traditional cryptographic methods, like Elliptic Curve Diffie-Hellman (ECDH), especially when supplemented with jamming. Our approach addresses the primary security concerns and sets the baseline for future secure biomedical communication systems advancements.

Ensuring the provision of sustainable and secure electrical power for ingestible/implantable medical devices (IMDs) is crucial for facilitating the multifaceted capabilities of these IMDs and preventing the need for recurrent battery replacements. Using photovoltaic (PV) energy harvesting in conjunction with an external light source can be advantageous for an optical wireless power transfer (OWPT) system to enable energy self-sufficiency in IMDs. This study investigates the performance of OWPT using commercial monocrystalline silicon PV cells exposed to an 810 nm Near-infrared (NIR) LED light. The ethical concerns are addressed by utilizing porcine samples (ex vivo approach), eliminating the need for live animal experimentation. The experimental setup employs porcine meat samples with several compositions, e.g., pure fat, pure muscle, and different layers of fat-muscle. The primary goal of this initial study is to analyze the open-circuit voltage output (VOC) of the PV against received optical power in the presence of biological tissue. Our study demonstrates that PV cells can generate voltage even when exposed to light passing through porcine samples with a thickness of up to 30 mm. Furthermore, the VOC values of PV cells attained in this study meet the required voltage input level for supplying current IMDs, typically ranging from 2V to 3V. The findings of this study provide valuable insights into OWPT systems in the future, where monocrystalline silicon PV cells can be employed as energy harvester devices to supply various IMDs utilizing NIR light.

Cough is the most common symptom prompting individuals to seek medical advice. However, the widespread adoption of autonomous cough monitoring using wearable devices remains limited. This paper introduces a wireless cough monitoring device utilizing piezoelectric energy harvesting technology. The design emphasizes cost-effectiveness and energy efficiency, allowing simple attachment onto human skin using medical-grade tapes. The device's standout feature lies in its departure from continuously recording real-time acoustic data at a high sampling rate, as commonly employed in prior works. Instead, it capitalizes on the energy harvesting capability, utilizing harvested energy from muscle movements induced by coughing as crucial information. The energy harvested within specific intervals translates into a historical record of cough occurrences during that timeframe. This Energy-as-Data protocol substantially reduces the device's duty cycle, resulting in a remarkable extension of battery life by up to 2100%. Notably, this extension is achieved while maintaining reasonable accuracy in cough monitoring. With this capability, the device can autonomously monitor and analyze cough data from both in- and outpatients, serving daily, research, and clinical purposes. Its potential extends to enhancing prediction and management of severe respiratory diseases.

Microwave technology is emerging as a promising candidate in the field of medical diagnosis and imaging and has paved the way for a transition from invasive to non-invasive methods of monitoring various biological phenomena inside the human body. Intracranial Pressure (ICP) is considered to be a very important parameter by medical practitioners for assessing the health of a subject. Accurate, prolonged, and noninvasive measurement of ICP is still an open area of research with no clinical success so far. Therefore, in this paper, a microwave-based method for non-invasive monitoring of ICP is proposed. The setup utilizes flexible, thin, small, and lightweight planner antennas that are very suitable for non-invasive monitoring of ICP from the skin without compromising the comfort of subject. The proposed microwave method is tested on a realistic head phantom model which imitates the functioning of hydrodynamics in a real human head. The measurement results from the proposed method are verified using invasive pressure sensors. It is deduced from numerous trials that the proposed microwave system can detect small changes in ICP pressure and its response is analogous to actual pressure values measured by invasive pressure sensors.

Early cancer detection is crucial, especially for intestinal cancer with subtle early symptoms. While camera-based Wireless Capsule Endoscopy (WCE) systems are efficient, patient-friendly, and safe investigating gastrointestinal (GI) track thoroughly, some limitations persist in visualizing only the inner part of the GI regions. Our study introduces a radio channel analysis -based approach to detect intestinal/abdominal tumors which are not visible for the WCE camera, i.e., the tumors which have started to grow on the outer parts of the intestinal track. Focused on S-parameter patterns in realistic human voxel models, our simulation-based method discerns dielectric property variations in normal and tumorous tissues, replicating intricate tissue characteristics. Preliminary simulation results in different intestine locations demonstrate our technique’s efficacy in differentiating normal and tumor cases based on S-parameter patterns. With a 98% accuracy rate, simple logistic regression classification model excels in distinguishing normal from tumor tissues, significantly enhancing diagnostic precision in GI health monitoring showcasing its potential to revolutionize early cancer detection and advance diagnostic accuracy within simulated human anatomy. This represents a substantial stride toward improving healthcare outcomes through cutting-edge technology.

Bioimpedance analysis (BIA) is a non-invasive and safe method to measure body composition. Nowadays, due to technological progress, smaller and cheaper devices allow the implementation of BIA into wearable devices. In this pilot study, we analyzed the measurement precision of a cheap BIA solution for wearable devices. Intra-session, intra-day, and inter-day reproducibility of raw impedance values from three subjects at three different body locations (hand-to-hand, hand-to-torso, torso-to-torso), and for three different frequencies (6, 54, and 500 kHz) were analyzed using the coefficient of variation (CV%). Hand-to-hand and hand-to-torso measurements resulted, on average, in high intra-session (CV% = 0.14% and CV% = 0.11%, respectively), intra-day (CV% = 1.67% and CV% = 1.26%, respectively), and inter-day (CV% = 1.53% and CV% = 1.31%) precision. Absolute impedance values for the torso-to-torso measurements showed a larger mean variation (intra-session CV% = 0.68%; intra-day CV% = 5.53%, inter-day CV% = 3.13%). Overall, this cheap BIA solution shows high precision and promising usability for further integration into a wearable measurement environment.

Optical wireless communication (OWC) has emerged as a promising technology for implantable medical devices because it provides private and secure wireless links for patients, low-power consumption, and high-speed data transmission. The OWC system’s receiving end typically relies on a photodetector with a limited field-of-view, necessitating direct line-of-sight connections for effective transmission. The directional nature of light-tissue interaction on the in-body communication can be problematic as the quality of the optical signal is rapidly deteriorated due to the properties of biological tissues, including scattering, absorption, and reflection, leading to a substantial loss of optical beam power reaching the photodetector’s sensitive area. In this sense, any misalignment that occurs in the in-body device can directly impact the power level and further degrade the received signal quality. Numerous studies have been conducted on this topic in free-space environments; nevertheless, only a few results have been found for in-body cases. In this work, we experimentally demonstrate the impact of the in-body device misalignment on the OWC-based in-body communication system. Three cases were investigated: aligned systems, as well as lateral and angular misalignments. We considered an 810 nm Near-infrared (NIR) LED as a transmitter because the optical signal of the mentioned wavelength propagates better than other wavelengths through biological tissues. For the experiments, we used pure muscle and fat tissues with 15 mm thickness at different temperatures (23 ℃ and 37 ℃). We also tested with thicker meat samples (30 mm, 38 mm, and 40 mm, consisting of muscle + fat layers) at 37 ℃. This study adhered to ANSI.Z136.1–2007 safety standards. First, the results reveal that optical power still reaches the receiver in an aligned reference case at a meat thickness of 40 mm. Second, the in-body device misalignment significantly degrades the optical power density received, which is more pronounced under lateral than angular conditions. These misalignment effects must be carefully considered for further system enhancement when using OWC for the in-body communication system.

This paper investigates fat tissue as a medium for communication in implantable/ingestible medical device (IMD) systems based on optical wireless communication (OWC). The findings emphasize the importance of tissue characteristics (temperature in particular) for optimizing OWC performance. This study considered Near-infrared (NIR) light with 810 nm wavelength and fresh porcine samples to mimic the human tissue. The study employs a realistic measurement approach in an ex vivo setting using various porcine samples: pure fat and flesh tissues and samples with different thicknesses. This study also investigates the influence of porcine temperature on the optical communication channels, which are measured by comparing the received optical power at 23 °C and 37 °C. In general, tissue samples at warmer temperatures (37 °C) receive higher optical power than colder samples. The results also demonstrate the superior optical power transmission capabilities of pure fat compared to pure flesh in porcine tissue samples in warm conditions. We also found that porcine with multiple layers of fat (fatty sample) yields higher received optical power than porcine with multiple layers of flesh (muscular). The results of this study provide valuable insights and relevant considerations for OWC-based in-body communication conducted using porcine samples.

The last decade has witnessed significant improvements in vehicular technology, especially in providing a safer and more enjoyable environment for drivers and passengers. Fully autonomous vehicles are no longer a dream but are now a successful technology across the globe. Features such as autopilot, assisted parking, speed warning, and lane change assistance have improved the quality of user experience while using an automobile. Apart from this, e-health services have also become a prime aspect of the modern vehicular industry. Therefore, this research presents preliminary studies on mm-wave radar setup based on Frequency Modulated Continuous Wave (FMCW) technology in the 76 to 81 GHz band for vital sign monitoring of drivers and passengers in a vehicular environment. The effect of system parameters and the driver’s location with respect to radar is studied using human subjects to determine the optimum setup for vital sign monitoring. Measurement results showcase that mm-wave radars can be utilized for accurate and efficient measurement of the vital signs of drivers in vehicular environments.