Temporal analysis of academic performance in higher education before, during and after COVID-19 confinement using artificial intelligence.

This study provides the profiles and success predictions of students considering data before, during, and after the COVID-19 pandemic. Using a field experiment of 396 students and more than 7400 instances, we have analyzed students' performance considering the temporal distribution of autonomous learning during courses from 2016/2017 to 2020/2021. After applying unsupervised learning, results show 3 main profiles from the clusters obtained in the simulations: students who work continuously, those who do it in the last-minute, and those with a low performance in the whole autonomous learning. We have found that the highest success ratio is related to students that work in a continuous basis. However, last-minute working is not necessarily linked to failure. We have also found that students' marks can be predicted successfully taking into account the whole data sets. However, predictions are worse when removing data from the month before the final exam. These predictions are useful to prevent students' wrong learning strategies, and to detect malpractices such as copying. We have done all these analyses taking into account the effect of the COVID-19 pandemic, founding that students worked in a more continuous basis in the confinement. This effect was still present one year after. Finally, We have also included an analysis of the techniques that could be more effective to keep in a future non-pandemic scenario the good habits that were detected in the confinement.

Reward-Penalty Weighted Ensemble for Emotion State Classification from Multi-Modal Data Streams.

Researchers have shown the limitations of using the single-modal data stream for emotion classification. Multi-modal data streams are therefore deemed necessary to improve the accuracy and performance of online emotion classifiers. An online decision ensemble is a widely used approach to classify emotions in real-time using multi-modal data streams. There is a plethora of online ensemble approaches; these approaches use a fixed parameter ([Formula: see text]) to adjust the weights of each classifier (called penalty) in case of wrong classification and no reward for a good performing classifier. Also, the performance of the ensemble depends on the [Formula: see text], which is set using trial and error. This paper presents a new Reward-Penalty-based Weighted Ensemble (RPWE) for real-time multi-modal emotion classification using multi-modal physiological data streams. The proposed RPWE is thoroughly tested using two prevalent benchmark data sets, DEAP and AMIGOS. The first experiment confirms the impact of the base stream classifier with RPWE for emotion classification in real-time. The RPWE is compared with different popular and widely used online ensemble approaches using multi-modal data streams in the second experiment. The average balanced accuracy, F1-score results showed the usefulness and robustness of RPWE in emotion classification in real-time from the multi-modal data stream.

Application of an Exploratory Knowledge-Discovery Pipeline Based on Machine Learning to Multi-Scale OMICS Data to Characterise Myocardial Injury in a Cohort of Patients with Septic Shock: An Observational Study.

Currently, there is no therapy targeting septic cardiomyopathy (SC), a key contributor to organ dysfunction in sepsis. In this study, we used a machine learning (ML) pipeline to explore transcriptomic, proteomic, and metabolomic data from patients with septic shock, and prospectively collected measurements of high-sensitive cardiac troponin and echocardiography. The purposes of the study were to suggest an exploratory methodology to identify and characterise the multiOMICs profile of (i) myocardial injury in patients with septic shock, and of (ii) cardiac dysfunction in patients with myocardial injury. The study included 27 adult patients admitted for septic shock. Peripheral blood samples for OMICS analysis and measurements of high-sensitive cardiac troponin T (hscTnT) were collected at two time points during the ICU stay. A ML-based study was designed and implemented to untangle the relations among the OMICS domains and the aforesaid biomarkers. The resulting ML pipeline consisted of two main experimental phases: recursive feature selection (FS) assessing the stability of biomarkers, and classification to characterise the multiOMICS profile of the target biomarkers. The application of a ML pipeline to circulate OMICS data in patients with septic shock has the potential to predict the risk of myocardial injury and the risk of cardiac dysfunction.

Real-Time Emotion Classification Using EEG Data Stream in E-Learning Contexts.

In face-to-face and online learning, emotions and emotional intelligence have an influence and play an essential role. Learners' emotions are crucial for e-learning system because they promote or restrain the learning. Many researchers have investigated the impacts of emotions in enhancing and maximizing e-learning outcomes. Several machine learning and deep learning approaches have also been proposed to achieve this goal. All such approaches are suitable for an offline mode, where the data for emotion classification are stored and can be accessed infinitely. However, these offline mode approaches are inappropriate for real-time emotion classification when the data are coming in a continuous stream and data can be seen to the model at once only. We also need real-time responses according to the emotional state. For this, we propose a real-time emotion classification system (RECS)-based Logistic Regression (LR) trained in an online fashion using the Stochastic Gradient Descent (SGD) algorithm. The proposed RECS is capable of classifying emotions in real-time by training the model in an online fashion using an EEG signal stream. To validate the performance of RECS, we have used the DEAP data set, which is the most widely used benchmark data set for emotion classification. The results show that the proposed approach can effectively classify emotions in real-time from the EEG data stream, which achieved a better accuracy and F1-score than other offline and online approaches. The developed real-time emotion classification system is analyzed in an e-learning context scenario.

Influence of COVID-19 confinement on students' performance in higher education.

This study analyzes the effects of COVID-19 confinement on the autonomous learning performance of students in higher education. Using a field experiment with 458 students from three different subjects at Universidad Autónoma de Madrid (Spain), we study the differences in assessments by dividing students into two groups. The first group (control) corresponds to academic years 2017/2018 and 2018/2019. The second group (experimental) corresponds to students from 2019/2020, which is the group of students that had their face-to-face activities interrupted because of the confinement. The results show that there is a significant positive effect of the COVID-19 confinement on students' performance. This effect is also significant in activities that did not change their format when performed after the confinement. We find that this effect is significant both in subjects that increased the number of assessment activities and subjects that did not change the student workload. Additionally, an analysis of students' learning strategies before confinement shows that students did not study on a continuous basis. Based on these results, we conclude that COVID-19 confinement changed students' learning strategies to a more continuous habit, improving their efficiency. For these reasons, better scores in students' assessment are expected due to COVID-19 confinement that can be explained by an improvement in their learning performance.

Biomedical Holistic Ontology for People with Rare Diseases.

This research provides a biomedical ontology to adequately represent the information necessary to manage a person with a disease in the context of a specific patient. A bottom-up approach was used to build the ontology, best ontology practices described in the literature were followed and the minimum information to reference an external ontology term (MIREOT) methodology was used to add external terms of other ontologies when possible. Public data of rare diseases from rare associations were used to build the ontology. In addition, sentiment analysis was performed in the standardized data using the Python library Textblob. A new holistic ontology was built, which models 25 real scenarios of people with rare diseases. We conclude that a comprehensive profile of patients is needed in biomedical ontologies. The generated code is openly available, so this research is partially reproducible. Depending on the knowledge needed, several views of the ontology should be generated. Links to other ontologies should be used more often to model the knowledge more precisely and improve flexibility. The proposed holistic ontology has many benefits, such as a more standardized computation of sentiment analysis between attributes.

Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare.

Precision Medicine implies a deep understanding of inter-individual differences in health and disease that are due to genetic and environmental factors. To acquire such understanding there is a need for the implementation of different types of technologies based on artificial intelligence (AI) that enable the identification of biomedically relevant patterns, facilitating progress towards individually tailored preventative and therapeutic interventions. Despite the significant scientific advances achieved so far, most of the currently used biomedical AI technologies do not account for bias detection. Furthermore, the design of the majority of algorithms ignore the sex and gender dimension and its contribution to health and disease differences among individuals. Failure in accounting for these differences will generate sub-optimal results and produce mistakes as well as discriminatory outcomes. In this review we examine the current sex and gender gaps in a subset of biomedical technologies used in relation to Precision Medicine. In addition, we provide recommendations to optimize their utilization to improve the global health and disease landscape and decrease inequalities.

Mining Facebook Data of People with Rare Diseases: A Content-Based and Temporal Analysis.

This research characterized how Facebook deals with rare diseases. This characterization included a content-based and temporal analysis, and its purpose was to help users interested in rare diseases to maximize the engagement of their posts and to help rare diseases organizations to align their priorities with the interests expressed in social networks. This research used Netvizz to download Facebook data, word clouds in R for text mining, a log-likelihood measure in R to compare texts and TextBlob Python library for sentiment analysis. The Facebook analysis shows that posts with photos and positive comments have the highest engagement. We also observed that words related to diseases, attention, disability and services have a lot of presence in the decalogue of priorities (which serves for all associations to work on the same objectives and provides the lines of action to be followed by political decision makers) and little on Facebook, and words of gratitude are more present on Facebook than in the decalogue. Finally, the temporal analysis shows that there is a high variation between the polarity average and the hour of the day.

Monitoring and Prognosis System Based on the ICF for People with Traumatic Brain Injury.

The objective of this research is to provide a standardized platform to monitor and predict indicators of people with traumatic brain injury using the International Classification of Functioning, Disability and Health, and analyze its potential benefits for people with disabilities, health centers and administrations. We developed a platform that allows automatic standardization and automatic graphical representations of indicators of the status of individuals and populations. We used data from 730 people with acquired brain injury performing periodic comprehensive evaluations in the years 2006-2013. Health professionals noted that the use of color-coded graphical representation is useful for quickly diagnose failures, limitations or restrictions in rehabilitation. The prognosis system achieves 41% of accuracy and sensitivity in the prediction of emotional functions, and 48% of accuracy and sensitivity in the prediction of executive functions. This monitoring and prognosis system has the potential to: (1) save costs and time, (2) provide more information to make decisions, (3) promote interoperability, (4) facilitate joint decision-making, and (5) improve policies of socioeconomic evaluation of the burden of disease. Professionals found the monitoring system useful because it generates a more comprehensive understanding of health oriented to the profile of the patients, instead of their diseases and injuries.

Automatic assessment of socioeconomic impact on cardiac rehabilitation.

Disability-Adjusted Life Years (DALYs) and Quality-Adjusted Life Years (QALYs), which capture life expectancy and quality of the remaining life-years, are applied in a new method to measure socioeconomic impacts related to health. A 7-step methodology estimating the impact of health interventions based on DALYs, QALYs and functioning changes is presented. It relates the latter (1) to the EQ-5D-5L questionnaire (2) to automatically calculate the health status before and after the intervention (3). This change of status is represented as a change in quality of life when calculating QALYs gained due to the intervention (4). In order to make an economic assessment, QALYs gained are converted to DALYs averted (5). Then, by inferring the cost/DALY from the cost associated to the disability in terms of DALYs lost (6) and taking into account the cost of the action, cost savings due to the intervention are calculated (7) as an objective measure of socioeconomic impact. The methodology is implemented in Java. Cases within the framework of cardiac rehabilitation processes are analyzed and the calculations are based on 200 patients who underwent different cardiac-rehabilitation processes. Results show that these interventions result, on average, in a gain in QALYs of 0.6 and a cost savings of 8,000 €.

Circles of Health: towards an advanced social network about disabilities of neurological origin.

OBJECTIVES: This research is concerned with the study of a new social-network platform, which (1) provides people with disabilities of neurological origin, their relatives, health professionals, therapists, carers and institutions with an interoperable platform that supports standard indicators, (2) promotes knowledge democratization and user empowerment, and (3) allows making decisions with a more informed opinion. METHODS: A new social network, Circles of Health, has been designed, developed and tested by end-users. To allow monitoring the evolution of people's health status and comparing it with other users and with their cohort, anonymized data of 2675 people from comprehensive and multidimensional medical evaluations, carried out yearly from 2006 to 2010, have been standardized to the International Classification of Functioning, Disability and Health, integrated into the corresponding medical health records and then used to automatically generate and graphically represent multidimensional indicators. These indicators have been integrated into Circles of Health's social environment, which has been then evaluated via expert and user-experience analyses. RESULTS: Patients used Circles of Health to exchange bio-psycho-social information (medical and otherwise) about their everyday lives. Health professionals remarked that the use of color-coding in graphical representations is useful to quickly diagnose deficiencies, difficulties or barriers in rehabilitation. Most people with disabilities complained about the excessive amount of information and the difficulty in interpreting graphical representations. CONCLUSIONS: Health professionals found Circles of Health useful to generate a more integrative understanding of health based on a comprehensive profile of individuals instead of being focused on patient's diseases and injuries. People with disabilities found enriching personal knowledge with the experiences of other users helpful. The number of descriptors used at the same time in the graphical interface should be reduced in future versions of the social-network platform.