Hospital logistics can significantly benefit from autonomous robotic technologies capable of alleviating personnel workload, reducing operational costs, and streamlining internal workflows. This paper introduces HOSBOT (HOSpital roBOT), a modular and cost-effective robotic system designed to enhance hospital logistics and workflow management through autonomous missions and real-time monitoring of transported items. HOSBOT combines a commercially available autonomous mobile robot with a customisable non-motorised cart, the SmartRack, which accommodates standalone, sensorised SmartBoxes equipped with Radio Frequency IDentification technology. This paper describes the overall design of HOSBOT, including its mechanical, electrical, and software interfaces. The system has been validated in three pilot experiments in real hospital environments across Europe, demonstrating good levels of acceptability, usability and efficacy. This innovative robotic solution highlights the potential for scalable integration of autonomous mobile robots in healthcare and other indoor logistics settings.

Monitoring of advanced cancer patients' health, treatment, and supportive care is essential for improving cancer survival outcomes. Traditionally, oncology has relied on clinical metrics such as survival rates, time to disease progression, and clinician-assessed toxicities. In recent years, patient-reported outcome measures (PROMs) have provided a complementary perspective, offering insights into patients' health-related quality of life (HRQoL). However, collecting PROMs consistently requires frequent clinical assessments, creating important logistical challenges. Wearable devices combined with artificial intelligence (AI) present an innovative solution for continuous, real-time HRQoL monitoring. While deep learning models effectively capture temporal patterns in physiological data, most existing approaches are unimodal, limiting their ability to address patient heterogeneity and complexity. This study introduces a multimodal deep learning approach to estimate HRQoL in advanced cancer patients. Physiological data, such as heart rate and sleep quality collected via wearable devices, are analyzed using a hybrid model combining convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) networks with an attention mechanism. The BiLSTM extracts temporal dynamics, while the attention mechanism highlights key features, and CNNs detect localized patterns. PROMs, including the Hospital Anxiety and Depression Scale (HADS) and the Integrated Palliative Care Outcome Scale (IPOS), are processed through a parallel neural network before being integrated into the physiological data pipeline. The proposed model was validated with data from 204 patients over 42 days, achieving a mean absolute percentage error (MAPE) of 0.24 in HRQoL prediction. These results demonstrate the potential of combining wearable data and PROMs to improve advanced cancer care.

Recent advancements in Natural Language Processing (NLP) have substantially enhanced language understanding. However, non-English languages, especially in specialized and low-resource domains like biomedicine, remain largely underrepresented. Bridging this gap is essential for promoting inclusivity and expanding the global applicability of NLP technologies. This study presents a cross-lingual knowledge distillation framework that utilizes sentence transformers to improve domain-specific NLP capabilities in non-English languages. Specifically, the framework focuses on biomedical text classification tasks. By aligning sentence embeddings between a teacher model trained on English biomedical corpora and a multilingual student model, the proposed method effectively transfers both domain-specific and task-specific knowledge. This alignment allows the student model to efficiently process and adapt to biomedical texts in Spanish, French, and German, particularly in low-resource settings with limited tuning data. Extensive experiments with domain-adapted models like BioBERT and multilingual BERT with machine-translated text pairs demonstrate substantial performance improvements in downstream biomedical NLP tasks. The proposed framework proves highly effective in scenarios characterized by limited training data availability. The results highlight the scalability and effectiveness of this approach, facilitating the development of robust multilingual models tailored to the biomedical domain, thus advancing global accessibility and impact in biomedical NLP applications.

As diabetes management becomes more complicated, there is an increasing interest in understanding how to manage diabetes with physical activity. Our study aimed to investigate the role of wearable, non-invasive technologies in collecting data related to physical activity to model them via artificial intelligence methods for efficient diabetes management. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (also known as PRISMA) protocol and searched three databases, namely PubMed, Scopus, and Web of Science. Out of 960 titles, we included 32 in the full-text analysis. Results showed two main methods were used for the analysis, i.e., statistical and classification modeling. Results indicate among the employed regression methods, linear regression was used more than other methods, and the most common classification-based method for analyzing data was the Artificial Neural Network method. Assessing the quality of papers that used the classification method was done through Prediction model Risk Of Bias Assessment Tool (also known as PROBAST) and Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (also known as TRIPOD) tools. Based on PROBAST outcomes, although the risk of bias was low in most of the works, explaining the analyzing method specifically, the method of handling missing data needs more attention. Upon evaluating papers using the TRIPOD, it realized that there is a need to place emphasis on improving the quality of the presentation and explanation of the result. According to our review, the conjunction of non-invasive technologies and artificial intelligence is promising in managing diabetic risk factors for real-time monitoring of physical activities, enabling regular clinical intervention and optimized medical treatment.

The role of ECG segmentation tool has been pivotal in automated analysis of real-time ECG signals for detection of non-invasive cardiovascular and physiological conditions. Most of the existing approaches focus on traditional signal processing and/or traditional machine learning based approaches which are highly dependent on signal noise, inter/intra subject variability, etc. With the advent of deep learning based networks, it is possible to design and develop the classification model based on local features along with spatial and temporal context of the physiological signals. In this paper, we developed the attention based Convolutional Bidirectional Long Short Term Memory (Conv-BiLSTM) architecture network based on local beat features and temporal sequencing while correlating ECG beat across different positions. The performance of our ECG segmentation tool has been evaluated against the state-of-the art approaches in terms of ECG segmentation and fiducial point detection accuracy. The ECG segmentation accuracy was 95% whereas fiducial point detection accuracy was 99.4%.

The majority of people with Autism Spectrum Disorder (ASD) exhibit difficulties in social communication and behavior, which hinder their learning capability, amid others. Among technological solutions for people with ASD, serious games are frequently used to enhance learning of specific skills and instructional contents. However, because of heterogeneity in applications and game design, few studies have investigated their use in training daily activities. This paper presents a 3D personalized serious game we developed and validated to help ASD patients practice with shopping activities. Personalized training is paramount in people with ASD, thus several elements of this game were personalized to improve engagement and therefore the effectiveness of the virtual training. In order to assess the validity of the game, ten subjects (age [Formula: see text], 20% female) with ASD played ten sessions of the serious game, once per week. The participants underwent a real-life experience pre- and post-training in a real-life supermarket. Changes in daily living skills among participants were evaluated through specific tools: a form based on the International Classification of Functioning, Disability and Health for Children and Youth; and the Vineland Adaptive Behavior Scale II. Significant improvements (p 0.05) were detected in the main skills trained with the serious game, especially in learning the shopping procedure, directing attention, and problem-solving skills. These findings suggest that personalized serious games can represent a prominent tool to enhance daily living skills, but future work should clinically validate their efficacy.

The papers in this special focus on enabling technologies for next generation telehealthcare applications. The use of Information and Communication Technology (ICT) for health and well-being is rapidly increasing in the majority of high-income countries. The interest about telehealthcare allows the provisioning of various kinds of health-related services and applications over the Internet. There are several benefits associated with tele-healthcare, including: the reduction of infection risk due to optimized patients access to clinical centers; optimized healthcare workflows; containment of hospital costs; increased patient safety; improves in the quality of life of both patients and their families. Common telehealthcare applications include tele-nursing, tele-rehabilitation, tele-dialog, tele-monitoring, tele-analysis, tele-pharmacy, tele-care, tele-psychiatry, tele-radiology, tele-pathology, teledermatology, tele-dentistry, tele-audiology, tele-ophthalmology, etc. In the past ten years, key enabling technologies (KETs) such as Internet of Things (IoT), tools for big data management and processing, Cloud/Edge/Fog computing, Artificial Intelligence (AI), Blockchain reached an advanced maturity, and therefore the potential for revolutionizing the whole tele-healthcare sector.

Falls are a major problem of later life having severe consequences on quality of life and a significant burden in occidental countries. Many technological solutions have been proposed to assess the risk or to predict falls and the majority is based on accelerometers and gyroscopes. However, very little was done for identifying first time fallers, which are very difficult to recognize. This paper presents a metamodel predicting falls using short term Heart Rate Variability (HRV) analysis acquired at the baseline. About 170 hypertensive patients (age: 72 ± 8 years, 56 female) were investigated, of which 34 fell once in the 3 months after the baseline assessment. This study is focused on hypertensive patients, which were considered as convenient pragmatic sample, as they undergo regular outpatient visits, during which short term Electrocardiogram (ECG) can be easily recorded without significant increase of healthcare costs. For each subject, 11 consecutive excerpts of 5 min each (55 min) were extracted from ECGs recorded between 10:30 and 12:30 and analysed. Linear and nonlinear HRV features were extracted and averaged among the 11 excerpts, which were, then, considered for the statistical and data mining analysis. The best predictive metamodel was based on Multinomial Naïve Bayes, which enabled to predict first-time fallers with sensitivity, specificity, and accuracy rates of 72%, 61%, and 68%, respectively.

This study aims to develop an automatic classifier for risk assessment in patients suffering from congestive heart failure (CHF). The proposed classifier separates lower risk patients from higher risk ones, using standard long-term heart rate variability (HRV) measures. Patients are labeled as lower or higher risk according to the New York Heart Association classification (NYHA). A retrospective analysis on two public Holter databases was performed, analyzing the data of 12 patients suffering from mild CHF (NYHA I and II), labeled as lower risk, and 32 suffering from severe CHF (NYHA III and IV), labeled as higher risk. Only patients with a fraction of total heartbeats intervals (RR) classified as normal-to-normal (NN) intervals (NN/RR) higher than 80% were selected as eligible in order to have a satisfactory signal quality. Classification and regression tree (CART) was employed to develop the classifiers. A total of 30 higher risk and 11 lower risk patients were included in the analysis. The proposed classification trees achieved a sensitivity and a specificity rate of 93.3% and 63.6%, respectively, in identifying higher risk patients. Finally, the rules obtained by CART are comprehensible and consistent with the consensus showed by previous studies that depressed HRV is a useful tool for risk assessment in patients suffering from CHF.

In this study, we investigated the discrimination power of short-term heart rate variability (HRV) for discriminating normal subjects versus chronic heart failure (CHF) patients. We analyzed 1914.40 h of ECG of 83 patients of which 54 are normal and 29 are suffering from CHF with New York Heart Association (NYHA) classification I, II, and III, extracted by public databases. Following guidelines, we performed time and frequency analysis in order to measure HRV features. To assess the discrimination power of HRV features, we designed a classifier based on the classification and regression tree (CART) method, which is a nonparametric statistical technique, strongly effective on nonnormal medical data mining. The best subset of features for subject classification includes square root of the mean of the sum of the squares of differences between adjacent NN intervals (RMSSD), total power, high-frequencies power, and the ratio between low- and high-frequencies power (LF/HF). The classifier we developed achieved sensitivity and specificity values of 79.3 % and 100 %, respectively. Moreover, we demonstrated that it is possible to achieve sensitivity and specificity of 89.7 % and 100 %, respectively, by introducing two nonstandard features ΔAVNN and ΔLF/HF, which account, respectively, for variation over the 24 h of the average of consecutive normal intervals (AVNN) and LF/HF. Our results are comparable with other similar studies, but the method we used is particularly valuable because it allows a fully human-understandable description of classification procedures, in terms of intelligible "if … then …" rules.