We introduce a wearable-based system for real-time ECG anomaly detection and contextual interpretation within a mobile-health framework. Twenty-four-hour Holter ECG data are synchronized over wireless/mobile networks with e.g. Apple Health streams (iPhone + iWatch), including activity states (walking, running, resting, sleeping) and heart rate history. A hybrid preprocessing pipeline extracts instantaneous frequency (Hilbert), spectral entropy, and RMS energy, concatenated into fixed-length multichannel tensors for deep-learning models deployed via edge or cloud SaaS. The model detects critical cardiac anomalies correlating each with user activity and exertion context. This multimodal approach distinguishes physiological deviations during motion from pathological events at rest or sleep and suppresses motion artifacts. Experiments with subjects wearing both Holter and Apple devices demonstrate improved sensitivity and specificity versus ECG-only baselines. Our system exemplifies wearable computing, mobile health, ML-enabled mobile systems, and edge/cloud mobile analytics. Fig. 1 shows a complete system for recording and classifying ECG signals, including a Holter ECG with electrodes, a smartphone and a smartwatch [1].

Electrical impedance tomography (EIT) is a non-invasive imaging technique with promising applications in mobile and embedded healthcare systems. It reconstructs internal conductivity distributions from boundary voltage measurements but suffers from ill-posedness, sensitivity to noise and limited spatial resolution when solved using traditional iterative methods. This work proposes a hybrid reconstruction architecture that combines a Vision Transformer (ViT) with a Physics-Informed Neural Network (PINN). The transformer extracts global contextual features from voltage measurements, while the PINN enforces the quasistatic conduction equation with Neumann boundary conditions through a physics-informed loss function. Although trained and evaluated on synthetic data, the proposed PINN-ViT model demonstrates improved reconstruction accuracy and noise robustness over classical algorithms and purely data-driven networks. These results indicate its potential for future deployment in energy-efficient, real-time EIT systems for mobile healthcare applications.

Thermal infrared (TIR) and terahertz (THz) spectra of electromagnetic radiation offer promising capabilities for non-intrusive detection of hidden items, providing insights into material properties and temperature distribution of objects. However, individually each modality shows limitations in resolution, penetration depth, and spectral characteristics . To address these limitations, we propose a novel approach that combines the complementary strengths of TIR and THz waves for enhanced detection performance. Specifically, we employ an attention-based fusion network to integrate TIR and THz imagery. The proposed algorithm uses a convolutional neural network (CNN) backbone to extract features from both spectra and involves an attention-based fusion mechanism to learn a set of weights that determine the importance of each feature representation for object detection. Our experimental results demonstrate that the proposed fusion network achieves state-of-the-art performance in detecting hidden items, outperforming conventional single-modality approaches. The combination of TIR and THz waves, coupled with the attention-based fusion network, provides a robust and versatile platform for non-intrusive detection of hidden items in a variety of scenarios. The method achieves state-of-the-art performance with an average accuracy (ACC) of 92.34 % and a detection rate (DR) of 91.12 % at zero false detection rate (FDR). This is the first known work to show combined capabilities of passive THz and TIR imaging for more adequate detection and classification of concealed objects.

The aim of the research was to develop a new device for monitoring and diagnosing functional disorders of the lower urinary tract, which will enable measurement of muscle tension (EMG) and electrical impedance tomography (EIT) with the possibility of electrostimulation of bladder reconstruction.

Numerous epidemic lung diseases such as COVID-19, tuberculosis (TB), and pneumonia have spread over the world, killing millions of people. Medical specialists have experienced challenges in correctly identifying these diseases due to their subtle differences in Chest X-ray images (CXR). To assist the medical experts, this study proposed a computer-aided lung illness identification method based on the CXR images. For the first time, 17 different forms of lung disorders were considered and the study was divided into six trials with each containing two, two, three, four, fourteen, and seventeen different forms of lung disorders. The proposed framework combined robust feature extraction capabilities of a lightweight parallel convolutional neural network (CNN) with the classification abilities of the extreme learning machine algorithm named CNN-ELM. An optimistic accuracy of 90.92% and an area under the curve (AUC) of 96.93% was achieved when 17 classes were classified side by side. It also accurately identified COVID-19 and TB with 99.37% and 99.98% accuracy, respectively, in 0.996 microseconds for a single image. Additionally, the current results also demonstrated that the framework could outperform the existing state-of-the-art (SOTA) models. On top of that, a secondary conclusion drawn from this study was that the prospective framework retained its effectiveness over a range of real-world environments, including balanced-unbalanced or large-small datasets, large multiclass or simple binary class, and high- or low-resolution images. A prototype Android App was also developed to establish the potential of the framework in real-life implementation.

Diabetic retinopathy (DR) is an incurable retinal condition caused by excessive blood sugar that, if left untreated, can result in even blindness. A novel automated technique for DR detection has been proposed in this paper. To accentuate the lesions, the fundus images (FIs) were preprocessed using Contrast Limited Adaptive Histogram Equalization (CLAHE). A parallel convolutional neural network (PCNN) was employed for feature extraction and then the extreme learning machine (ELM) technique was utilized for the DR classification. In comparison to the similar CNN structure, the PCNN design uses fewer parameters and layers, which minimizes the time required to extract distinctive features. The effectiveness of the technique was evaluated on two datasets (Kaggle DR 2015 competition (Dataset 1; 34,984 FIs) and APTOS 2019 (3,662 FIs)), and the results are promising. For the two datasets mentioned, the proposed technique attained accuracies of 91.78 % and 97.27 % respectively. However, one of the study's subsidiary discoveries was that the proposed framework demonstrated stability for both larger and smaller datasets, as well as for balanced and imbalanced datasets. Furthermore, in terms of classifier performance metrics, model parameters and layers, and prediction time, the suggested approach outscored existing state-of-the-art models, which would add significant benefit for the medical practitioners in accurately identifying the DR.

This work presents a novel multimodal biometric dataset with emerging biometric traits including 3D face, thermal face, iris on-the-move, iris mobile, somatotype and smartphone sensors. This dataset was created to resemble on-the-move characteristics in applications such as border control. The five types of biometric traits were selected as they can be captured while on-the-move, are contactless, and show potential for use in a multimodal fusion verification system in a border control scenario. Innovative sensor hardware was used in the data capture. The data featuring these biometric traits will be a valuable contribution to advancing biometric fusion research in general. Baseline evaluation was performed on each unimodal dataset. Multimodal fusion was evaluated based on various scenarios for comparison. Real-time performance is presented based on an Automated Border Control (ABC) scenario.

This paper presents the results of research on the identification of moisture inside the walls of buildings with the use of non-invasive electrical impedance tomography (EIT). The novelty and contribution of this research is the development of an original algorithmic method to solve the ill posedness, inverse problem. Since the new algorithm optimizes the method for each pixel of the tomographic image, taking into account a specific measurement vector, regardless of what and how many homogeneous methods are included in the algorithm, the obtained results are more accurate than those obtained with the use of homogeneous methods. As part of the research, prototypes of the EIT tomograph and electrodes for examining walls were designed and manufactured.

We consider concept of approximate query in RDBMS i.e. query that returns results which may differ from common (exact) query results in a way but its evaluation requires less resources.In the work we focus mostly on time and storage space aspects.We follow one of the state-of-the-art trends using synopses of data as the input of approximate query evaluation.We propose some measures of approximate query results quality.Basing on them we present steps of adaptive elaboration of synopses quality measure that should be mutually corresponding.

We compared the possibility of detecting hidden objects covered with various types of clothing by using passive imagers operating in a terahertz (THz) range at 1.2 mm (250 GHz) and a mid-wavelength infrared at 3-6 μm (50-100 THz). We investigated theoretical limitations, performance of imagers, and physical properties of fabrics in both the regions. In order to investigate the time stability of detection, we performed measurements in sessions each lasting 30 min. We present a theoretical comparison of two spectra, as well as the results of experiments. In order to compare the capabilities of passive imaging of hidden objects, we combined the properties of textiles, performance of imagers, and properties of radiation in both spectral ranges. The paper presents the comparison of the original results of measurement sessions for the two spectrums with analysis.