EAI Endorsed Transactions on Pervasive Health and Technology

Evaluating Physicians’ Satisfaction with Using eHealth System for Managing Patients in Primary and Secondary Care

2025-02-18T14:53:56+00:00

INTRODUCTION: eHealth systems are raising both patient satisfaction and medical care. The proper workflow regulations and data exchange between primary and secondary healthcare are crucial.

OBJECTIVES: Investigation of the major determinants influencing the physicians’ satisfaction while using an eHealth information system.

METHODS: A survey of primary and secondary healthcare medical professionals was conducted in R.N. Macedonia. The categorical variables from the data analysis were presented and a logistic regression was carried out.

RESULTS: The multiple logistic regression model was statistically significant for sufficient evidence to reject the null hypothesis in which the overall satisfaction rating of the eHealth system usage for managing patients and other healthcare services will not be affected by the other variables in favour of the alternative Ha.

CONCLUSION: Various factors between primary and secondary healthcare professionals regarding system’s usage satisfaction are presented and studied. Various issues were revealed between both parties that should serve the policymakers and medical authorities for further improvements.

Use of personal mobile technologies for peer-based assessment of stress: a systematic literature review

2025-03-19T09:32:38+00:00

The use of personal mobile technologies has grown in recent years, providing a method for collecting high-frequency and high-quality data on human behaviors and states, amongst the others, on stress levels. Mobile technologies can play a significant role in peer-based stress assessment, particularly in e-mental health and well-being. It is accessible, convenient, and reliable compared to traditional self-report methods, making it a popular choice for collecting data. This systematic literature review aimed to explore the use of mobile technologies for peer-based assessment of stress. We analyzed existing literature to understand how mobile technologies have been used to assess stress levels through peer feedback—from relatives, friends, or others with close and daily contact with the individual. The results of the review showed that mobile technologies have the potential to be a valuable tool for peer-based stress assessment, as they can provide real-time and convenient data collection. However, although its popularity has grown in recent years, it is worth noting that the use of paper and pen questionnaires has remained prevalent in peer-based stress assessment over the last decade. This indicates that there is still a need for further exploration and evaluation of the benefits and limitations of both methods.

Anomaly Detection in Skull scanning Images based on Multi-sensor Fusion

2025-03-12T09:39:27+00:00

INTRODUCTION: Skull bones typically possess complex structures and features. When scanned with ordinary sensors, they are easily affected by noise due to the small difference between abnormal areas and normal tissue.

OBJECTIVES: In order to solve the problem of small differences between abnormal areas and normal tissues, which make them susceptible to noise interference, this paper proposes a multi-sensor fusion based skull scan image anomaly detection method.

METHODS: Firstly, the frequency correction factor is utilized to modify the frequency domain characteristics of the sensor signal during the skull scanning image acquisition process, aiming to enhance signal quality and reduce noise impact during acquisition. Secondly, bilateral filters and discrete wavelet transform are employed to subject the skull scanning image to dual domain decomposition in spatial and transformation domains, aiding in distinguishing between normal and abnormal regions. Subsequently, the low-frequency fusion algorithm guided by filtering and the high-frequency fusion algorithm based on multi-scale morphological gradients are fused, and the fused dual frequency components are merged back into the original spatial domain to retain important details. The fused reconstructed image aids in improving the accuracy of anomaly detection. Finally, a backbone network with an auto encoder structure is established to learn the feature representation of fused images, and an unsupervised deep neural network is employed to establish a detection model for anomaly detection in skull scanning images.

RESULTS: Through experiments, it has been demonstrated that the F1 score approaches 1, the ROC curve closely approaches the upper left corner, and the AUC value approaches 1 after applying the proposed method for anomaly detection in skull scanning images.

CONCLUSION: This algorithm has high sensitivity and low specificity, achieving high detection accuracy and demonstrating good performance.

Battery signal control model for large-scale IoT medical monitors under multipath interference

2025-02-14T13:49:11+00:00

INTRODUCTION: In large-scale IOT medical monitors, the accurate control of battery signals has been facing the problem of multipath interference. Multipath interference causes the receiver to receive multiple signals propagating through different paths and interfering with each other, which results in an imbalance in the battery signal control based on the time delay of the "transmit-receive" signals.

OBJECTIVES: To solve the multipath interference problem of existing battery signal control, this paper designs a battery signal control model for a large-scale IoT medical monitor.

METHODS: Firstly, this paper uses time synchronization to align the time between the receiver and the transmitter to synchronize the communication signals of the acquisition system; Next, a transverse time-domain filter is used for modulation filtering; Then, a judgment feedback equalization algorithm is introduced in combination with a full-feedback filter to suppress the inter-code interference and improve the signal quality; Finally, a fractional interval equalizer is designed to adjust the weight coefficients of the equalizer taps, and implement intelligent battery signal control in multi-hop communication under multipath interference based on fractional interval and bit error rate (BER) feedback modulation.

RESULTS: Experimental results have shown that after using the method described in this paper to control the communication signal of the monitor battery, the output signal is relatively stable, and the BER reaches 1×10^-4 when the signal-to-noise ratio is equal to 18dB. The BER is low, and the carrier-to-noise ratio of the output signal is 0.73~0.85. The carrier-to-noise ratio always remains above 0.73.

CONCLUSION: The technologies effectively deal with complex network environment and channel condition variation, ensuring balanced system control, and the signal control effect is outstanding.

Sport injury imaging for deep blood flow distribution with laser speckle

2025-02-19T10:47:40+00:00

When laser speckle program technology is used to measure the blood flow distribution of deep tissues (such as subcutaneous tissue) in sports injuries, the deep blood flow characteristics of sports injuries contain a large amount of turbid tissue fluid. Laser passing through turbid tissue fluid will produce strong interference static speckle, masking the dynamic speckle of blood flow distribution, resulting in poor imaging effect of blood flow characteristics. Propose laser speckle imaging optimization technology and apply it to the measurement of deep tissue blood flow distribution in sports injuries. Based on the principle of laser speckle imaging technology, the problems in laser speckle imaging of deep blood flow distribution characteristics in sports injuries are analyzed. An exponential Laplace loss function is introduced to reduce the amplitude of changes in blood flow characteristics in intra class sports injuries and collect deep blood flow distribution characteristics in sports injuries; On the basis of calculating the laser speckle contrast ratio, the blood volume flow rate is determined, and the blood volume flow rate data is combined with the laser speckle contrast ratio to achieve imaging of deep blood flow distribution in sports injuries. The experimental results show that the improved laser speckle imaging technology has better imaging effects in imaging the deep blood flow distribution of sports injuries; Compared with the comparison method, the DICE coefficient, average accuracy MPA, and global imaging index have all improved, indicating that this method can effectively improve the imaging effect and is feasible.

SAM2CLIP2SAM: Vision Language Model for Segmentation of 3D CT Scans for Covid-19 Detection

2025-04-02T08:30:17+00:00

This paper presents a new approach for effective segmentation of images that can be integrated into any model and methodology; the paradigm that we choose is classification of medical images (3-D chest CT scans) for Covid-19 detection. Our approach includes a combination of vision-language models that segment the CT scans, which are then fed to a deep neural architecture, named RACNet, for Covid-19 detection. In particular, a novel framework, named SAM2CLIP2SAM, is introduced for segmentation that leverages the strengths of both Segment Anything Model (SAM) and Contrastive Language-Image Pre-Training (CLIP) to accurately segment the right and left lungs in CT scans, subsequently feeding these segmented outputs into RACNet for classification of COVID-19 and non-COVID-19 cases. At first, SAM produces multiple part-based segmentation masks for each slice in the CT scan; then CLIP selects only the masks that are associated with the regions of interest (ROIs), i.e., the right and left lungs; finally SAM is given these ROIs as prompts and generates the final segmentation mask for the lungs.
Experiments are presented across two Covid-19 annotated databases which illustrate the improved performance obtained when our method has been used for segmentation of the CT scans.

The Future of Fall Prevention: Integrating OpenPose with Cutting-Edge ML Models

2025-04-02T09:30:24+00:00

The research paper aims to assess ML models for video-recorded gaits with an aim of classifying people into high or low risks to fall groups. Several ML algorithms were tried employing OpenPose for CV, with RF showing the best outcomes: 93% accuracy along with F1-score as well as balanced sensitivity (93.50%) as well as specificity (92.50%). Some important determining factors were speed per unit distance, angle among other statistical measures. In comparison to wearables-based DL approaches plus traditional fall detection methods, this study’s approach showed higher accuracy and adaptability within health care settings.

Privacy-Preserving Abnormal Gait Detection Using Computer Vision and Machine Learning

2025-04-16T12:34:31+00:00

Gait analysis plays a pivotal role in diagnosing a spectrum of neurological and musculoskeletal disorders. Variations in gait patterns often serve as early indicators of underlying health conditions, underscoring the importance of precise and timely analysis for effective intervention and treatment. In recent years, computer vision techniques have emerged as robust tools for automated gait analysis, offering non-invasive, costeffective, and scalable solutions. However, existing approaches often overlook the critical aspect of privacy preservation. In this study, we propose the world’s pioneering computer vision-based abnormal gait detection system with a privacy-preserving mechanism. Specifically, we extract 2D skeletons from encrypted images using a deep neural network model, which is facilitated by an optical system incorporating a custom-made refractive optical element. These extracted features are then fed into machine learning models for the detection of normal versus abnormal gait patterns. Evaluations across various models including random forest, decision tree, K-nearest neighbor, support vector machine, neural network, and convolutional neural network reveal that the random forest model attains the highest classification performance based on 2D skeletons extracted from encrypted images.

Privacy-Preserving Human Motion Analysis for Lower Back Pain Stratification through Federated Learning

2025-04-17T08:33:25+00:00

Human Gait Analysis is crucial in healthcare applications, with numerous research works focusing on machine learning and deep learning approaches for tasks such as abnormal gait detection and gait quality assessment. However, developing such models requires collecting and sharing a significant amount of patient data, raising privacy concerns. In this study, we introduce the world’s first technique for constructing a deep neural network model to stratify patients’ pain levels based on video recordings of timed up-and-go activities, while ensuring privacy preservation through modern federated learning algorithms. Our experimental results demonstrate the effectiveness of this technique in accurately stratifying LBP levels without the need for data sharing among local clients to maintain privacy.