1,721,068 research outputs found
Non-negative factorization methods for extracting semantically relevant features in Intelligent Data Analysis
Full text linkPreface
No full textPreface: Enabling Technologies for Effective Planning and Management in Sustainable Smart Citie
Does Time Matter in Analyzing Educational Data? - A New Dataset for Streaming Learning Analytics
No full textThis research introduces a novel dataset developed for streaming learning analytics, derived from the Open University Learning Analytics Dataset (OULAD). The dataset incorporates essential temporal information that captures the timing of student interactions with the Virtual Learning Environment (VLE). By integrating these time-based interactions, the dataset enhances the capabilities of stream algorithms, which are particularly well-suited for real-time monitoring and analysis of student learning behaviors. Experiments utilizing the Online Bagging algorithm across three temporal units-months, trimesters, and semesters-demonstrated that the dataset contains pertinent information for predicting student outcomes. Despite the variations associated with different temporal units, the classifier effectively identified patterns within the data, especially for the majority class (Pass), achieving high F1 scores. These results indicate that the temporal structure of the data supports accurate predictions; however, challenges remain in accurately identifying the minority class (Fail). This dataset paves the way for more dynamic and responsive educational interventions by enabling timely predictions of student outcomes. Such capabilities facilitate continuous learning support within VLEs, allowing educators to respond promptly to student needs and enhance overall learning experiences
Evaluating the robustness of a contact-less mHealth solution for personal and remote monitoring of blood oxygen saturation
Full text linkMHealth technologies play a fundamental role in epidemiological situations such as the ongoing outbreak of COVID-19 because they allow people to self-monitor their health status (e.g. vital parameters) at any time and place, without necessarily having to physically go to a medical clinic. Among vital parameters, special care should be given to monitor blood oxygen saturation (SpO2), whose abnormal values are a warning sign for potential COVID-19 infection. SpO2 is commonly measured through the pulse oximeter that requires skin contact and hence could be a potential way of spreading contagious infections. To overcome this problem, we have recently developed a contact-less mHealth solution that can measure blood oxygen saturation without any contact device but simply processing short facial videos acquired by any common mobile device equipped with a camera. Facial video frames are processed in real-time to extract the remote photoplethysmographic signal useful to estimate the SpO2 value. Such a solution promises to be an easy-to-use tool for both personal and remote monitoring of SpO2. However, the use of mobile devices in daily situations holds some challenges in comparison to the controlled laboratory scenarios. One main issue is the frequent change of perspective viewpoint due to head movements, which makes it more difficult to identify the face and measure SpO2. The focus of this work is to assess the robustness of our mHealth solution to head movements. To this aim, we carry out a pilot study on the benchmark PURE dataset that takes into account different head movements during the measurement. Experimental results show that the SpO2 values obtained by our solution are not only reliable, since they are comparable with those obtained with a pulse oximeter, but are also insensitive to head motion, thus allowing a natural interaction with the mobile acquisition device
Subtractive clustering for seeding non-negative matrix factorizations
No full textNon-negative matrix factorization is a multivariate analysis method which is proven to be useful in many areas such as bio-informatics, molecular pattern discovery, pattern recognition, document clustering and so on. It seeks
a reduced representation of a multivariate data matrix into the product of
basis and encoding matrices possessing only non-negative elements, in order
to learn the so called part-based representations of data. All algorithms for
computing non-negative matrix factorization are iterative, therefore particular emphasis must be placed on a proper initialization of NMF because of its local convergence. The problem of selecting appropriate starting matrices
becomes more complex when data possess special meaning as in document
clustering. In this paper, we propose the adoption of the subtractive clustering algorithm as a scheme to generate initial matrices for non-negative
matrix factorization algorithms. Comparisons with other commonly adopted
initializations of non-negative matrix factorization algorithms have been performed and the proposed scheme reveals to be a good trade-off between effectiveness and speed. Moreover, the effectiveness of the proposed initialization
to suggest a number of basis for NMF, when data distances are estimated,
is illustrated when NMF is used for solving clustering problems where the
number of groups in which the data are grouped is not known a-priori. The
influence of a proper rank factor on the interpretability and the effectiveness
of the results are also discussed
Computational Intelligence for Digital Health: A brief summary of our research work
Full text linkIn the last few decades, a digitization process has involved various aspects of daily life, and the healthcare sector is one of the fields most heavily affected by this digital transformation. Artificial Intelligence, and in particular Computational Intelligence (CI) techniques, such as Neural Networks and Fuzzy Systems, have proven to be promising methods for extracting meaningful knowledge from medical data and for developing intelligent systems for faster diagnosis, improved monitoring and effective healthcare. CI-based systems can learn models from data that evolve as data changes, taking into account the uncertainty that characterizes health data and processes. Our group working at the Computational Intelligence Laboratory (CILab) of the Department of Computer Science, University of Bari, is currently carrying out scientific research on the application of CI techniques to Digital Health problems
MicroRNA expression classification for pediatric multiple sclerosis identification
Full text linkMicroRNAs (miRNAs) are a set of short non-coding RNAs that play significant regulatory roles in cells. The study of miRNA data produced by Next-Generation Sequencing techniques can be of valid help for the analysis of multifactorial diseases, such as Multiple Sclerosis (MS). Although extensive studies have been conducted on young adults affected by MS, very little work has been done to investigate the pathogenic mechanisms in pediatric patients, and none from a machine learning perspective. In this work, we report the experimental results of a classification study aimed at evaluating the effectiveness of machine learning methods in automatically distinguishing pediatric MS from healthy children, based on their miRNA expression profiles. Additionally, since Attention Deficit Hyperactivity Disorder (ADHD) shares some cognitive impairments with pediatric MS, we also included patients affected by ADHD in our study. Encouraging results were obtained with an artificial neural network model based on a set of features automatically selected by feature selection algorithms. The results obtained show that models developed on automatically selected features overcome models based on a set of features selected by human experts. Developing an automatic predictive model can support clinicians in early MS diagnosis and provide new insights that can help find novel molecular pathways involved in MS disease
Incremental learning and granular computing from evolving data streams: An application to speech-based bipolar disorder diagnosis
No full textWe apply an evolving granular-computing modeling approach, called evolving Optimal Granular System (eOGS), to bipolar mood disorder (BD) diagnosis based on speech data streams. The eOGS online learning algorithm reveals information granules in the flow and design the structure and parameters of a granular rule-based model with a certain degree of interpretability based on acoustic attributes obtained from phone calls made over 7 months to the Psychiatry department of a hospital. A multi-objective programming problem that trades-off information specificity, model compactness, and numerical and granular error indices is presented. Spectral and prosodic attributes are ranked and selected based on a hybrid Pearson-Spearman correlation coefficient. Low attribute-class correlation, ranging from 0.03 to 0.07, is observed, as well as high class overlap, which is typical in the psychiatric field. eOGS models for BD recognition overcome alternative computational-intelligence models, namely, Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) and Fuzzy-set-Based evolving Modeling (FBeM-Gauss), by a small margin in both best and average cases; followed by eXtended Takagi-Sugeno (xTS) and evolving Takagi Sugeno (eTS) types of models. The proposed eOGS model using only 8 of the original acoustic attributes, and about 15 ‘If-Then’ inference rules, has exhibited the best root mean square error, 0.1361, and 91.8% accuracy in sharp BD class estimates. Granules associated to linguistic labels and a granular input-output map offer human understandability with relation to the inherent process of generating class estimates. Linguistically readable eOGS rules may assist physicians in explaining symptoms and making a diagnosis
Improving a Mirror-based Healthcare System for Real-time Estimation of Vital Parameters
No full textContactless methods are widely used to measure vital signs from recorded or live videos using remote photoplethysmography (rPPG), which takes advantage of the slight skin color variation that occurs periodically on specific body regions with each blood pulse.
However, existing rPPG-based solutions are typically expensive and not suitable for daily use at home for personal healthcare. To address this issue, we have recently developed a low-cost device that allows for the real-time estimation of vital signs using rPPG and can be easily integrated into any common home environment. The device consists of a smart mirror equipped with a camera that captures facial videos and extracts rPPG signals by processing video frames.
One major limitation of this solution was its high sensitivity to abrupt head movements during video acquisition. This paper presents some advancements in the development of our smart device aimed at obtaining a more robust measurement of vital signs. Experimental results on live videos show that the new version of our system overcomes the limitations of the previous version, offering a more stable performance. Moreover, the new methodology shows improved performance compared to other state-of-the-art rPPG algorithms when tested on pre-recorded in-house videos from the UBFC-RPPG database
Estimating blood pressure using video-based PPG and deep learning
No full textThis paper introduces a novel pipeline for estimating systolic and diastolic blood pressure using remote photoplethysmographic (rPPG) signals derived from video recordings of subjects’ faces. The pipeline consists of three main stages: rPPG signal extraction, denoising to transform the rPPG signal into a PPG-like waveform, and blood pressure estimation. This approach directly addresses the current lack of datasets that simultaneously include video, rPPG, and blood pressure data. To overcome this, the proposed pipeline leverages the extensive availability of PPG-based blood pressure estimation techniques, in combination with state-of-the-art algorithms for rPPG extraction, enabling the generation of reliable PPG-like signals from video input. To validate the pipeline, we conducted comparative analyses with state-of-the-art methods at each stage and collected a dedicated dataset through controlled laboratory experimentation. The results demonstrate that the proposed solution effectively captures blood pressure information, achieving a mean error of 9.2 ± 11.3 mmHg for systolic and 8.6 ± 9.1 mmHg for diastolic blood pressure. Moreover, the denoised rPPG signals show a strong correlation with conventional PPG signals, supporting the reliability of the transformation process. This non-invasive and contactless method offers considerable potential for long-term blood pressure monitoring, particularly in Ambient Assisted Living (AAL) systems, where unobtrusive and continuous health monitoring is essential
- …
