Explore the vast range of titles available.

This review uniquely integrates open access dataset taxonomy with methodological trends in stress analysis, outlining the experimental framework and highlighting key gaps in reproducibility and FAIR compliance. In this context, stress induction methods, ground truth labeling approaches, open access datasets, computational advances, and current challenges and limitations are reported. A systematic review over the last decade (2014–2024) identified thirty-two open access affective datasets eligible for stress-related research, encompassing multimodal physiological signals, including electroencephalography (EEG), electrocardiography (ECG), electrodermal activity (EDA), and respiration (Resp), as well as behavioral measures, such as motion, audiovisual, and eye tracking data. Recent developments in signal analysis methods (2023–2025) highlight the predominance of multimodal fusion, advances in deep and self-supervised learning, personalized/adaptive models, and the growing adoption of explainable Artificial Intelligence, while machine learning approaches continue to hold a fundamental role. Despite these advances, several limitations and challenges remain, including heterogeneous experimental designs, sensor variability, limited demographic representation, data synchronization and labeling, and class imbalance. An effective experimental framework for stress research should integrate individual demographics and traits, reliable stressors, and high-quality physiological recordings within a well-defined and bias-controlled protocol, thereby producing reliable data to support and validate computational stress modeling. Continued progress in sensing, experimental standardization, and interpretable modeling is essential to produce reproducible, interpretable, and generalizable models of stress and emotions.

Speech emotion recognition (SER) has gained an increased interest during the last decades as part of enriched affective computing. As a consequence, a variety of engineering approaches have been developed addressing the challenge of the SER problem, exploiting different features, learning algorithms, and datasets. In this paper, we propose the application of the graph theory for classifying emotionally-colored speech signals. Graph theory provides tools for extracting statistical as well as structural information from any time series. We propose to use the mentioned information as a novel feature set. Furthermore, we suggest setting a unique feature-based identity for each emotion belonging to each speaker. The emotion classification is performed by a Random Forest classifier in a Leave-One-Speaker-Out Cross Validation (LOSO-CV) scheme. The proposed method is compared with two state-of-the-art approaches involving well known hand-crafted features as well as deep learning architectures operating on mel-spectrograms. Experimental results on three datasets, EMODB (German, acted) and AESDD (Greek, acted), and DEMoS (Italian, in-the-wild), reveal that our proposed method outperforms the comparative methods in these datasets. Specifically, we observe an average UAR increase of almost 18 % , 8 % and 13 % , respectively.

Pain is a multifaceted phenomenon that significantly affects a large portion of the global population. Objective pain assessment is essential for developing effective management strategies, which in turn contribute to more efficient and responsive healthcare systems. However, accurately evaluating pain remains a complex challenge due to subtle physiological and behavioural indicators, individual-specific pain responses, and the need for continuous patient monitoring. Automatic pain assessment systems offer promising, technology-driven solutions to support and enhance various aspects of the pain evaluation process. Physiological indicators offer valuable insights into pain-related states and are generally less influenced by individual variability compared to behavioural modalities, such as facial expressions. Skin conductance, regulated by sweat gland activity, and the heart’s electrical signals are both influenced by changes in the sympathetic nervous system. Biosignals, such as electrodermal activity (EDA) and electrocardiogram (ECG), can, therefore, objectively capture the body’s physiological responses to painful stimuli. This paper proposes a novel multi-modal ensemble deep learning framework that combines electrodermal activity and electrocardiogram signals for automatic pain recognition. The proposed framework includes a uni-modal approach (FCN-ALSTM-Transformer) comprising a Fully Convolutional Network, Attention-based LSTM, and a Transformer block to integrate features extracted by these models. Additionally, a multi-modal approach (CrossMod-Transformer) is introduced, featuring a dedicated Transformer architecture that fuses electrodermal activity and electrocardiogram signals. Experimental evaluations were primarily conducted on the BioVid dataset, with further cross-dataset validation using the AI4PAIN 2025 dataset to assess the generalisability of the proposed method. Notably, the CrossMod-Transformer achieved an accuracy of 87.52% on Biovid and 75.83% on AI4PAIN , demonstrating strong performance across independent datasets and outperforming several state-of-the-art uni-modal and multi-modal methods. These results highlight the potential of the proposed framework to improve the reliability of automatic multi-modal pain recognition and support the development of more objective and inclusive clinical assessment tools.

Gait analysis is crucial for the detection and management of various neurological and musculoskeletal disorders. The identification of gait events is valuable for enhancing gait analysis, developing accurate monitoring systems, and evaluating treatments for pathological gait. The aim of this work is to introduce the Smart-Insole Dataset to be used for the development and evaluation of computational methods focusing on gait analysis. Towards this objective, temporal and spatial characteristics of gait have been estimated as the first insight of pathology. The Smart-Insole dataset includes data derived from pressure sensor insoles, while 29 participants (healthy adults, elderly, Parkinson’s disease patients) performed two different sets of tests: The Walk Straight and Turn test, and a modified version of the Timed Up and Go test. A neurologist specialized in movement disorders evaluated the performance of the participants by rating four items of the MDS-Unified Parkinson’s Disease Rating Scale. The annotation of the dataset was performed by a team of experienced computer scientists, manually and using a gait event detection algorithm. The results evidence the discrimination between the different groups, and the verification of established assumptions regarding gait characteristics of the elderly and patients suffering from Parkinson’s disease.

Background: Diabetes is a chronic condition that necessitates regular monitoring and self-management of the patient’s blood glucose levels. People with type 1 diabetes (T1D) can live a productive life if they receive proper diabetes care. Nonetheless, a loose glycemic control might increase the risk of developing hypoglycemia. This incident can occur because of a variety of causes, such as taking additional doses of insulin, skipping meals, or overexercising. Mainly, the symptoms of hypoglycemia range from mild dysphoria to more severe conditions, if not detected in a timely manner. Objective: In this review, we aimed to report on innovative detection techniques and tactics for identifying and preventing hypoglycemic episodes, focusing on T1D. Methods: A systematic literature search following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines was performed focusing on the PubMed, GoogleScholar, IEEEXplore, and ACM Digital Library to find articles on technologies related to hypoglycemia detection in patients with T1D. Results: The presented approaches have been used or devised to enhance blood glucose monitoring and boost its efficacy in forecasting future glucose levels, which could aid the prediction of future episodes of hypoglycemia. We detected 19 predictive models for hypoglycemia, specifically on T1D, using a wide range of algorithmic methodologies, spanning from statistics (1.9/19, 10%) to machine learning (9.88/19, 52%) and deep learning (7.22/19, 38%). The algorithms used most were the Kalman filtering and classification models (support vector machine, k-nearest neighbors, and random forests). The performance of the predictive models was found to be satisfactory overall, reaching accuracies between 70% and 99%, which proves that such technologies are capable of facilitating the prediction of T1D hypoglycemia. Conclusions: It is evident that continuous glucose monitoring can improve glucose control in diabetes; however, predictive models for hypo- and hyperglycemia using only mainstream noninvasive sensors such as wristbands and smartwatches are foreseen to be the next step for mobile health in T1D. Prospective studies are required to demonstrate the value of such models in real-life mobile health interventions.

Automatic segmentation of the prostate of and the prostatic zones on MRI remains one of the most compelling research areas. While different image enhancement techniques are emerging as powerful tools for improving the performance of segmentation algorithms, their application still lacks consensus due to contrasting evidence regarding performance improvement and cross-model stability, further hampered by the inability to explain models’ predictions. Particularly, for prostate segmentation, the effectiveness of image enhancement on different Convolutional Neural Networks (CNN) remains largely unexplored. The present work introduces a novel image enhancement method, named RACLAHE, to enhance the performance of CNN models for segmenting the prostate’s gland and the prostatic zones. The improvement in performance and consistency across five CNN models (U-Net, U-Net++, U-Net3+, ResU-net and USE-NET) is compared against four popular image enhancement methods. Additionally, a methodology is proposed to explain, both quantitatively and qualitatively, the relation between saliency maps and ground truth probability maps. Overall, RACLAHE was the most consistent image enhancement algorithm in terms of performance improvement across CNN models with the mean increase in Dice Score ranging from 3 to 9% for the different prostatic regions, while achieving minimal inter-model variability. The integration of a feature driven methodology to explain the predictions after applying image enhancement methods, enables the development of a concrete, trustworthy automated pipeline for prostate segmentation on MR images.

Parkinson’s disease (PD) is characterized by a variety of motor and non-motor symptoms, some of them pertaining to gait and balance. The use of sensors for the monitoring of patients’ mobility and the extraction of gait parameters, has emerged as an objective method for assessing the efficacy of their treatment and the progression of the disease. To that end, two popular solutions are pressure insoles and body-worn IMU-based devices, which have been used for precise, continuous, remote, and passive gait assessment. In this work, insole and IMU-based solutions were evaluated for assessing gait impairment, and were subsequently compared, producing evidence to support the use of instrumentation in everyday clinical practice. The evaluation was conducted using two datasets, generated during a clinical study, in which patients with PD wore, simultaneously, a pair of instrumented insoles and a set of wearable IMU-based devices. The data from the study were used to extract and compare gait features, independently, from the two aforementioned systems. Subsequently, subsets comprised of the extracted features, were used by machine learning algorithms for gait impairment assessment. The results indicated that insole gait kinematic features were highly correlated with those extracted from IMU-based devices. Moreover, both had the capacity to train accurate machine learning models for the detection of PD gait impairment.

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by motor and nonmotor impairment with various implications on patients’ quality of life. Since currently available therapies are only symptomatic, identifying individuals with prodromal, preclinical, or early-stage PD is crucial, as they would be ideal candidates for future disease-modifying therapies. Our analysis aims to develop a robust model for accurate PD detection using accelerometer data collected from PD and non-PD individuals with mild or no tremor during phone conversations. An open-access dataset comprising accelerometer recordings from 22 PD patients and 11 healthy controls (HCs) was utilized. The data were preprocessed to extract relevant time-, frequency-, and energy-related features, and a bidirectional long short-term memory (Bi-LSTM) model with attention mechanism was employed for classification. The performance of the model was evaluated using fivefold cross-validation, and metrics of accuracy, precision, recall, specificity, and f1-score were computed. The proposed model demonstrated high accuracy (98%), precision (99%), recall (98%), specificity (96%), and f1-score (98%) in accurately distinguishing PD patients from HCs. Our findings indicate that the proposed model outperforms existing approaches and holds promise for detection of PD with subtle symptoms, like tremor, in the wild. Such symptoms can present in the early or even prodromal stage of the disease, and appropriate mobile-based applications may be a practical tool in real-life settings to alert individuals at risk to seek medical assistance or give patients feedback in monitoring their symptoms.

Gait analysis refers to the systematic study of human locomotion and finds numerous applications in the fields of clinical monitoring, rehabilitation, sports science and robotics. Wearable sensors for real-time gait monitoring have emerged as an attractive alternative to the traditional clinical-based techniques, owing to their low cost and portability. In addition, 3D printing technology has recently drawn increased interest for the manufacturing of sensors, considering the advantages of diminished fabrication cost and time. In this study, we report the development of a 3D-printed capacitive smart insole for the measurement of plantar pressure. Initially, a novel 3D-printed capacitive pressure sensor was fabricated and its sensing performance was evaluated. The sensor exhibited a sensitivity of 1.19 MPa−1, a wide working pressure range (<872.4 kPa), excellent stability and durability (at least 2.280 cycles), great linearity (R2=0.993), fast response/recovery time (142–160 ms), low hysteresis (DH<10%) and the ability to support a broad spectrum of gait speeds (30–70 steps/min). Subsequently, 16 pressure sensors were integrated into a 3D-printed smart insole that was successfully applied for dynamic plantar pressure mapping and proven able to distinguish the various gait phases. We consider that the smart insole presented here is a simple, easy to manufacture and cost-effective solution with the potential for real-world applications.

Contemporary deep learning-based decision systems are well-known for requiring high-volume datasets in order to produce generalized, reliable, and high-performing models. However, the collection of such datasets is challenging, requiring time-consuming processes involving also expert clinicians with limited time. In addition, data collection often raises ethical and legal issues and depends on costly and invasive procedures. Deep generative models such as generative adversarial networks and variational autoencoders can capture the underlying distribution of the examined data, allowing them to create new and unique instances of samples. This study aims to shed light on generative data augmentation techniques and corresponding best practices. Through in-depth investigation, we underline the limitations and potential methodology pitfalls from critical standpoint and aim to promote open science research by identifying publicly available open-source repositories and datasets.