Search

Valentina Perrone,1 Chiara Veronesi,1 Elisa Giacomini,1 Rita Citraro,2 Stefania Dell’Orco,3 Fabio Lena,4 Arrigo Paciello,5 Anna Maria Resta,6 Mihaela Nica,7 Daniela Ritrovato,7 Luca Degli Esposti1 1Clicon S.r.l. Società Benefit, Health Economics & Outcomes Research, Bologna, Italy; 2Dipartimento di Scienze della Salute, Università Magna Grecia di Catanzaro, Unita’ Operativa di Farmaco-logia Clinica e Farmacovigilanza, Azienda Ospedaliero-Universitaria “Mater Domini”, Catanzaro, Italy; 3ASL Roma 6, Rome, Italy; 4USL Toscana Sud Est, Grosseto, Italy; 5ATS Bergamo, Bergamo, Italy; 6Struttura Complessa di Farmacia Territoriale Area Vasta 1, Fano, Italy; 7Novartis Farma S.p.A, Origgio, Varese, ItalyCorrespondence: Valentina Perrone, CliCon S.r.l. Società Benefit, Health, Economics & Outcomes Research, Via Murri, 9, Bologna, 40137, Italy, Tel +39 3450316494, Email valentina.perrone@clicon.itPurpose: A retrospective analysis of real-world data was performed to assess the epidemiology and economic burden of multiple sclerosis (MS), relapsing-remitting MS (RRMS), and secondary-progressive MS (SPMS) in Italy.Patients and Methods: An observational study on administrative databases from a sample of Italian entities was carried-out. Between 01/2010– 12/2017, patients with ≥ 1 MS diagnosis code (ICD-9-CM:340 and/or exemption code:046) and/or ≥ 1 disease-modifying therapies (DMTs) prescription, were included. Among MS-cohort, SPMS patients were identified by ≥ 2 hospitalizations or by ≥ 2 drug prescriptions related to MS progression. MS patients not fulfilling SPMS criteria were included as RRMS. Mean annual healthcare costs were reported during follow-up and stratified by DMT treatment/untreatment.Results: Overall, 9543 MS patients were included; 8397 with RRMS and 1146 with SPMS. Estimated prevalence of MS was 141.6/100,000 inhabitants (RRMS 124.4/100,000 and SPMS 17.2/100,000). Mean annual cost for untreated and treated patient was respectively: € 3638 and € 11796 (MS-cohort), € 3183 and € 11486 (RRMS-cohort), € 6317 and € 15511 (SPMS-cohort). The first-line DMT treatment duration averaged 27.4 ± 22.8 months; the mean cost was 19004€ for the whole period. The second-line DMT treatment lasted on average 31.1 ± 24.5 months; the mean cost was 47293€ for the whole period.Conclusion: This study provided insights into the MS epidemiology in Italy and its economic burden. Healthcare costs associated with MS management were mainly driven by DMTs expenditure. A trend of higher healthcare-resource consumption was observed among SPMS-cohort.Keywords: real-world evidence, clinical practice, multiple sclerosis, epidemiology, healthcare resource consumption

Obiettivo del presente contributo è di portare l’attenzione su I poeti del Novecento di Franco Fortini, pubblicato in un periodo decisivo nella ridefinizione del canone poetico novecentesco, nonché dei criteri che consentono di riconoscerne il valore paradigmatico. In particolare, il tentativo di Fortini si contraddistingue per l’intenzione di superare la dimensione «editoriale» del genere antologia, proponendola all’interno di una riflessione critica sul «pregiudizio» della modernità

Abstract Background Age-related macular degeneration (AMD) is a common and chronic eye condition characterized by the presence of progressive degenerative abnormalities in the central retina (macula). Notably, neovascular, or wet, AMD (nAMD) occurs when new, abnormal blood vessels grow under the macula causing scarring of the macula itself and resulting in a loss of central vision, visual distortion, and an impaired capacity of perceiving colour contrast and intensity. Brolucizumab, a new generation anti-vascular endothelial growth factor (anti-VEGF) monoclonal antibody, was approved by the European Medicines Agency for the treatment of nAMD. The aim of this analysis is to evaluate the cost-effectiveness profile of brolucizumab, compared to the main therapeutic alternative available (aflibercept), for the treatment of nAMD. Methods The simulation of costs and outcomes was carried out using a Markov model over a time horizon of 15 years. In base-case, treatment effectiveness inputs for brolucizumab and aflibercept were extracted from the HAWK and HARRIER studies and from a network meta-analysis. The Italian National Healthcare Service (NHS) perspective was considered, therefore only healthcare direct costs (treatment acquisition, administration, adverse events, disease monitoring) were analysed. In the alternative scenarios, the societal perspective and a prolonged time horizon were considered. Model robustness was tested through sensitivity analyses. Results In the base-case analysis, brolucizumab was dominant over aflibercept (+ 0.11 years QALY gained and -€15,679 costs). Both one-way deterministic and probabilistic sensitivity analyses confirmed the robustness and reliability of base-case results. The results of the probabilistic sensitivity analysis showed that when the willingness to pay is equal to €50,000 per QALY gained, brolucizumab would be dominant in 84% of simulations and in the remaining simulations brolucizumab would be cost-effective compared to aflibercept. Results of the alternative scenarios and sensitivity analyses confirmed the results of base-case. Conclusion The cost-utility analysis shows that brolucizumab is dominant over aflibercept. Treatment with brolucizumab reduces the economic impact of nAMD and determined a slight increase of quality-adjusted survival. This analysis gives a high level of confidence that the treatment with brolucizumab would reduce the burden of intravitreal injections, compared to aflibercept, a relevant therapeutic alternative in Italy.

Scientific studies have shown that dysfunctional problems associated with dysmorphisms are the base of malocclusion. The position of the mouth and the head is a component of the general postural arrangement, so any head misalignment creates disharmonies and tensions among the neck muscles. Elastodontics is a treatment based on the use of removable elastomeric devices. These functional appliances are made of elastomeric material, acting tridimensionally on a neuromuscular system with orthopaedic and dental effects. Bioactivators have a vestibular and lingual flange with a free central area that permits them to position teeth without constraints. The flanges are linked by an occlusal plane that maintains the two skeletal bases and teeth in first-class normocclusion with propulsive and retropulsive effects on the maxilla and mandible. Elastomeric devices are generally well tolerated by patients requiring simple collaboration and management. This review aim is to describe the indications and limits of orthodontic therapy with elastomeric devices to provide an overview of elastodontics.

Valentina Perrone,1 Elisa Giacomini,1 Diego Sangiorgi,1 Margherita Andretta,2 Anna Michela Menti,2 Mariantonietta Naclerio,3 Daniela Ritrovato,3 Luca Degli Esposti1 1CliCon S.r.l. Health, Economics & Outcomes Research, Ravenna, Italy; 2Health Technology Assessment Unit, Azienda Zero, Padova, Italy; 3Novartis Farma Italy, Origgio, VA, ItalyCorrespondence: Valentina PerroneClicon Srl, Health, Economics and Outcomes Research, Via Salara, 36, 48100, Ravenna, ItalyTel +39 544 38393Fax +39 544 212699Email valentina.perrone@clicon.itPurpose: To analyze the treatment patterns of psoriatic arthritis (PSA) or ankylosing spondylitis (AS) patients under biological therapies and to evaluate in this population the health-care resource consumption and related costs.Patients and Methods: A retrospective analysis was performed on administrative databases of the Veneto region. Patients ≥ 18 years with at least one prescription of biological drugs and a diagnosis at any level for PSA or AS from January 1, 2011 to December 31, 2016 (inclusion period) were included. Index date (ID) was defined as date of first biological drug prescription during inclusion period. Patients were characterized the year before ID and followed-up for one year after ID. The drug utilization profile in terms of adherence, persistence and therapeutic regimen changes, and the health-care resource consumption was analyzed during follow-up.Results: A total of 2602 patients were included: 1857 with PSA and 745 with AS. In the PSA cohort, 40.3% of patients were prescribed adalimumab, 35.6% etanercept, 8.0% golimumab, 7.5% infliximab, 5.6% ustekinumab and 3.0% certolizumab. Percentage of PSA patients adherent to treatment was higher among ustekinumab patients (91.3%) and lower among etanercept users (54.3%). Persistence ranged from 53.2% (infliximab) to 70.3% (etanercept). Regarding AS cohort, 45.5% of patients were prescribed adalimumab, 26% etanercept, 17.3% infliximab, 9.7% golimumab and 1.5% certolizumab. Adherence ranged from 46.9% (etanercept) to 90.9% (certolizumab) and persistence from 62.8% (adalimumab) to 81.8% (certolizumab). Mean annual health-care costs (including costs for drug treatment, diagnostic services, specialist visits and hospital admissions) ranged from € 9727 (certolizumab) to € 14,994 (ustekinumab) among PSA patients and from € 9875 (infliximab) to € 12,991 (golimumab) among AS patients.Conclusion: This study in Veneto region gave a picture of biological treatment patterns among PSA and AS patients in a real-world setting. Our findings showed the high degree of variability concerning utilization of each biological drug and provided insight on the economic burden of both diseases.Keywords: psoriatic arthritis, ankylosing spondylitis, biologic drugs, drug utilization, real-world

There is growing interest in studying human brain connectivity and in modelling the brain functional structure as a network. Brain network creation requires parcellation of the cerebral cortex to define nodes. Parcellation might be affected by possible errors due to inter- and intra-subject variability as a consequence of brain structural and physiological characteristics and shape variations related to ageing and diseases, acquisition noise, and misregistration. These errors could induce a knock-on effect on network measure variability. The aim of this study was to investigate spatial stability, a measure of functional connectivity variations induced by parcellation errors. We simulated parcellation variability with random small spatial changes and evaluated its effects on twenty-seven graph-theoretical measures. The study included subjects from three public online datasets. Two brain parcellations were performed using FreeSurfer with geometric atlases. Starting from these, 100 new parcellations were created by increasing the area of 30% of parcels, reducing the area of neighbour parcels, with a rearrangement of vertices. fMRI data were filtered with linear regression, CompCor, and motion correction. Adjacency matrices were constructed with 0.1, 0.2, 0.3, and 0.4 thresholds. Differences in spatial stability between datasets, atlases, and threshold were evaluated. The higher spatial stability resulted for Characteristic-path-length, Density, Transitivity, and Closeness-centrality, and the lower spatial stability resulted for Bonacich and Katz. Multivariate analysis showed a significant effect of atlas, datasets, and thresholds. Katz and Bonacich centrality, which was subject to larger variations, can be considered an unconventional graph measure, poorly implemented in the clinical field and not yet investigated for reliability assessment. Spatial stability (SS) is affected by threshold, and it decreases with increasing threshold for several measures. Moreover, SS seems to depend on atlas choice and scanning parameters. Our study highlights the importance of paying close attention to possible parcellation-related spatial errors, which may affect the reliability of functional connectivity measures.

Background Siponimod is an effective treatment for patients with secondary progressive multiple sclerosis (SPMS), with active disease evidenced by relapses or imaging features characteristic of multiple sclerosis inflammatory activity, however there is a need to evaluate its economic value and sustainability compared to other disease modifying-therapies (DMTs). Objective To estimate the siponimod cost-effectiveness profile and its relative budget impact compared with other DMTs, by using the Italian National Healthcare System perspective. Methods We performed: 1) a cost-effectiveness analysis (CEA) vs interferon beta-1b using an analytical Markov model and a life time-horizon, and 2) a budget impact analysis by using 3-years time-horizon. The results were reported as incremental cost-effectiveness ratio (ICER) and net-monetary benefit (NMB) for CEA, using a willingness to pay threshold of €40,000 per QALY gained, and as difference in the overall budget (Euro) between the scenario with and without siponimod for budget impact. Results In the base case scenario siponimod resulted cost-effective compared with interferon beta-1b 28,891€ per QALY. Overall, the market access of siponimod was associated to an increased budget of about 3€ millions (+0.9%) in the next 3 years simulated. Conclusion Compared to interferon beta-1b, siponimod seems to be cost-effective in SPMS patients and sustainable, with less than 1% overall budget increased in the next 3 years. Future studies need to confirm our results in the real word setting and in other countries.

Abdominal pain represents a frequent symptom for referral to emergency departments and/or internal medicine outpatient setting. Similarly, fever, fatigue and weight loss are non-specific manifestations of disease. The present case describes the diagnostic process in a patient with abdominal pain and a palpable abdominal mass. Abdominal ultrasonography confirmed the presence of a mass in the mesogastrium. Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scans oriented toward calcific lymphadenopathies with increased metabolism in the positron emission tomography–computed tomography (PET-CT) scan. Laboratory examinations were inconclusive, although serology for Brucella and the Quantiferon test were positive. After multidisciplinary discussion, the patient underwent surgical excision of the abdominal mass. Histological examination excluded malignancies and oriented toward brucellosis in a patient with latent tuberculosis. The patient was treated with rifampin 600 mg qd and doxycycline 100 mg bid for 6 weeks with resolution of the symptoms. In addition, rifampin was continued for a total of 6 months in order to treat latent tuberculosis. This case underlines the need for a multidisciplinary approach in the diagnostic approach to abdominal lymphadenopathies.

Radiomic models outperform clinical data for outcome prediction in high-grade gliomas (HGG). However, lack of parameter standardization limits clinical applications. Many machine learning (ML) radiomic models employ single classifiers rather than ensemble learning, which is known to boost performance, and comparative analyses are lacking in the literature. We aimed to compare ML classifiers to predict clinically relevant tasks for HGG: overall survival (OS), isocitrate dehydrogenase (IDH) mutation, O-6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation, epidermal growth factor receptor vIII (EGFR) amplification, and Ki-67 expression, based on radiomic features from conventional and advanced magnetic resonance imaging (MRI). Our objective was to identify the best algorithm for each task. One hundred fifty-six adult patients with pathologic diagnosis of HGG were included. Three tumoral regions were manually segmented: contrast-enhancing tumor, necrosis, and non-enhancing tumor. Radiomic features were extracted with a custom version of Pyradiomics and selected through Boruta algorithm. A Grid Search algorithm was applied when computing ten times K-fold cross-validation (K=10) to get the highest mean and lowest spread of accuracy. Model performance was assessed as AUC-ROC curve mean values with 95% confidence intervals (CI). Extreme Gradient Boosting (xGB) obtained highest accuracy for OS (74,5%), Adaboost (AB) for IDH mutation (87.5%), MGMT methylation (70,8%), Ki-67 expression (86%), and EGFR amplification (81%). Ensemble classifiers showed the best performance across tasks. High-scoring radiomic features shed light on possible correlations between MRI and tumor histology.

In the biomedical field, there is an ever-increasing number of large, fragmented, and isolated data sources stored in databases and ontologies that use heterogeneous formats and poorly integrated schemes. Researchers and healthcare professionals find it extremely difficult to master this huge amount of data and extract relevant information. In this work, we propose a linked data approach, based on multilayer networks and semantic Web standards, capable of integrating and harmonizing several biomedical datasets with different schemas and semi-structured data through a multi-model database providing polyglot persistence. The domain chosen concerns the analysis and aggregation of available data on neuroendocrine neoplasms (NENs), a relatively rare type of neoplasm. Integrated information includes twelve public datasets available in heterogeneous schemas and formats including RDF, CSV, TSV, SQL, OWL, and OBO. The proposed integrated model consists of six interconnected layers representing, respectively, information on the disease, the related phenotypic alterations, the affected genes, the related biological processes, molecular functions, the involved human tissues, and drugs and compounds that show documented interactions with them. The defined scheme extends an existing three-layer model covering a subset of the mentioned aspects. A client–server application was also developed to browse and search for information on the integrated model. The main challenges of this work concern the complexity of the biomedical domain, the syntactic and semantic heterogeneity of the datasets, and the organization of the integrated model. Unlike related works, multilayer networks have been adopted to organize the model in a manageable and stratified structure, without the need to change the original datasets but by transforming their data “on the fly” to respond to user requests.

This paper is aimed at addressing all the critical aspects linked to the implementation of intensive care ventilators in a pediatric setting, highlighting the most relevant technical features and describing the methodology to conduct health technology assessment (HTA) for supporting the decision-making process. Four ventilator models were included in the assessment process. A decision-making support tool (DoHTA method) was applied. Twenty-eight Key Performance Indicators (KPIs) were identified, defining the safety, clinical effectiveness, organizational, technical, and economic aspects. The Performance scores of each ventilator have been measured with respect to KPIs integrated with the total cost of ownership analysis, leading to a final rank of the four possible technological solutions. The final technologies’ performance scores reflected a deliver valued, contextualized, and shared outputs, detecting the most performant technological solution for the specific hospital context. HTA results had informed and supported the pediatric hospital decision-making process. This study, critically identifying the pros and cons of innovative features of ventilators and the evaluation criteria and aspects to be taken into account during HTA, can be considered as a valuable proof of evidence as well as a reliable and transferable method for conducting decision-making processes in a hospital context.

Abstract Background To test the application of Business Process Management technology to manage clinical pathways, using a pediatric kidney transplantation as case study, and to identify the benefits obtained from using this technology. Methods Using a Business Process Management platform, we implemented a specific application to manage the clinical pathway of pediatric patients, and monitored the activities of the coordinator in charge of the case management during a 6-month period (from June 2015 to November 2015) using two methodologies: the traditional procedure and the one under study. Results The application helped physicians and nurses to optimize the amount of time and resources devoted to management purposes. In particular, time reduction was close to 60%. In addition, the reduction of data duplication, the integrated event management and the efficient data collection improved the quality of the service. Conclusions The use of Business Process Management technology, usually related to well-defined processes with high management costs, is an established procedure in multiple environments; its use in healthcare, however, is innovative. The use of already accepted clinical pathways is known to improve outcomes. The combination of these two techniques, well established in their respective areas of application, could represent a revolution in clinical pathway management. The study has demonstrated that the use of this technology in a clinical environment, using a proper architecture and identifying a well-defined process, leads to real benefits in terms of resources optimization and quality improvement.

abstract

Abstract Understanding temporal variation of threats that cause species endangerment is a key to understand conservation strategies needed to improve species recovery. We assessed temporal variation in the threats to species listed under the United States Endangered Species Act (ESA) as identified by the United States Fish and Wildlife Service (USFWS) and National Marine Fisheries Service (NMFS). Based on initial review of ESA listing decisions and literature, we identified six overarching threat categories: habitat modification, overutilization, pollution, species–species interaction, demographic stochasticity, and environmental stochasticity. We screened listing decision documents to determine threat occurrence (i.e., presence/absence of a given threat in a listing decision) for each threat category for all species listed between 1975 and 2017. We evaluated how the number of threats and specific threat occurrences changed over the past four decades. We found that the number of threats per listing decision increased more than twofold from an average of 1.5 (95% CI: 1.3–1.7) threats in 1975 to 3.7 (95% CI: 3.4–4.0) threats in 2017. Threat occurrence increased for habitat modification, environmental stochasticity and species–species interaction, while it decreased for overutilization since 1975 and for demographic stochasticity and pollution since the mid‐2000s. The documented increase in number of threats at time of listing may be due to a growing human population exerting increased pressure on species persistence, improved scientific advancement in understanding factors influencing species endangerment, or prolonged time taken for more recent species to be listed under the ESA. We believe that key federal and state governmental regulations have resulted in a documented decrease in overutilization, demographic stochasticity, and pollution, and we recommend large‐scale strategies combined with local planning efforts to address the growing threats of habitat loss, environmental stochasticity, and species–species interaction.

Purpose: We aimed to estimate the 3-year budget impact of replacing salmeterol/fluticasone with indacaterol/glycopyrronium for the treatment of chronic obstructive pulmonary disease patients in Italy. Methods: The budget impact model included patients with moderate, severe, and very severe chronic obstructive pulmonary disease and was developed using data from the FLAME study. The budget impact model considered direct medical costs from the National Healthcare System perspective over a 3-year time-horizon. Costs of drugs, adverse events, and exacerbations were estimated using public net prices for drugs and national (inpatient/outpatient) tariffs for patient management. Data on adverse events and exacerbations were derived from FLAME and other clinical trials. The model assumed an increasing uptake of indacaterol/glycopyrronium, substituting it for salmeterol/fluticasone. Two sensitivity analyses were conducted. In the first, every long-acting muscarinic antagonist/inhaled corticosteroid switch to indacaterol/glycopyrronium; in the second, drugs compliance varies from 20% to 60%. Results: The target population was estimated at 572,788 patients in the first year, increasing to 585,491 in the third year. The alternative treatment pattern, with a higher indacaterol/glycopyrronium uptake, was associated with a total cost of €1.69 billion over 3 years, while the total costs in the current treatment pattern amounted to €1.71 billion. Thus, increasing indacaterol/glycopyrronium use provided total savings of €17.3 million over 3 years. Sensitivity analysis confirmed robustness of the results. Conclusion: Increasing the uptake of indacaterol/glycopyrronium was estimated to be cost-saving for the Italian National Healthcare System, both when drug-related expenditures and total costs were considered. Savings increased with the number of patients switching to indacaterol/glycopyrronium.

This study focuses on landslide susceptibility mapping in the Daunia area (Apulian Apennines, Italy) and achieves this by using a multivariate statistical method and data processing in a Geographical Information System (GIS). The Logistic Regression (hereafter LR) method was chosen to produce a susceptibility map over an area of 130 000 ha where small settlements are historically threatened by landslide phenomena. By means of LR analysis, the tendency to landslide occurrences was, therefore, assessed by relating a landslide inventory (dependent variable) to a series of causal factors (independent variables) which were managed in the GIS, while the statistical analyses were performed by means of the SPSS (Statistical Package for the Social Sciences) software. The LR analysis produced a reliable susceptibility map of the investigated area and the probability level of landslide occurrence was ranked in four classes. The overall performance achieved by the LR analysis was assessed by local comparison between the expected susceptibility and an independent dataset extrapolated from the landslide inventory. Of the samples classified as susceptible to landslide occurrences, 85% correspond to areas where landslide phenomena have actually occurred. In addition, the consideration of the regression coefficients provided by the analysis demonstrated that a major role is played by the "land cover" and "lithology" causal factors in determining the occurrence and distribution of landslide phenomena in the Apulian Apennines.

This paper describes how specific blood gas analyzer characteristics can support the emerging clinical and organizational user’s needs while ensuring patient safety. A oneyear data set from two Italian hospitals was analyzed from 10 different blood gas analyzers. Performance measurements in terms of mean down time (MDT) were calculated to show how technical characteristics declared by the manufacturer compare to the analyzer availability in real clinical settings. Results show a high level of reliability for the analyzed technology, associated with very low MDT of each sensor integrated in the cartridge. Moreover, results show a high level of efficiency in cartridge usage. Such results are consistent with the specification of the GEM® Premier™ maintenance- free technology and are particularly relevant in the point-of-care testing setting.

The Learning Objects Repositories are electronic databases able to deliver material on the web allowing instructors sharing and reusing educational units and students accessing and enjoying them. The best way to guarantee these interactions is a good indexing. Each content needs a machine-understandable description able to declare requirements and limits for its right use and to improve any research and delivery action. These descriptions are stored in the metadata. Filling in the metadata is a boring and time-consuming activity but it is very important since it could influence the choice of the best material to deliver. This paper describes a possible methodological approach to automate this activity by extracting metadata directly from the files setting up the learning object itself. In the literature, there are many methods able to automatically characterize the technological aspects of the content, but very few of them are able to provide information about its pedagogical features. The proposed approach tries to draw together information theory, learning models, statistical analysis and ad hoc heuristics to extract a wide set of fields of the metadata. The results of a first experimentation are particularly encouraging to think about this approach as a solution to support learning object repositories and other platforms having needs to manage wide content storage and huge amount of users with various personal features, devices for interaction and goals as in the MOOCs.

This work proposes the definition of an Adaptive Conversation-based Learning System (ACLS) able to foster computer-mediated tutorial dialogues at the workplace in order to increase the probability to generate meaningful learning during conversations. ACLS provides a virtual assistant generating adaptive feedbacks, in the form of recommendations, for the conversation partners. The concepts extracted from the conversation texts trigger the recommendations, while queries on the organizational knowledge, represented by means of Semantic Web technologies, generate their content. Lastly, the Fuzzy Formal Concept Analysis is exploited to conceptualize domain knowledge.

Over the last decade, the interest in disaster education has grown rapidly. Several studies demonstrate that effective results can be obtained in this field only with instructional methods able to motivate the learner and to support them in practicing skills by means of narrative situations. The narrative is a privileged method that can help developing cognitive skills, organize knowledge and support the construction of meaning. In this paper we present a novel adaptive storytelling model defined in the context of ALICE project and its contextualization in the field of disaster education. The defined model aims at maximizing learner’s understanding and development of concepts fostering the “learning in action” and problem solving skills in natural disaster contexts by combining direct experience, observation, discovery and action. In particular the model arises motivation in the story and creatively engage learners in finding solutions to a problem and building personal responsibility. The experimentation results are encouraging and confirm that the storytelling offers higher engagement than the traditional practicing methods.

[Display omitted] • We propose two-stage real-time blood glucose estimation model via ECG signals. • The first stage extracts ECG beat morphology via self-attention deep network. • The second stage utilizes extracted beat morphology to predict blood glucose. • 89 % blood glucose estimation accuracy as per clinical Clarke's Grid Error. • Outperformed blood glucose estimation models based on traditional HRV feature. With the advent of modern digital technology, the physiological signals (such as electrocardiogram) are being acquired from portable wearable devices which are being used for non-invasive chronic disease management (such as Type 1 Diabetes). The diabetes management requires real-time assessment of blood glucose which is cumbersome for paediatric population due to clinical complexity and invasiveness. Therefore, real-time non-invasive blood glucose estimation is now pivotal for effective diabetes management. In this paper, we propose a Self-Attention Deep Neural Network Regressor for real-time non-invasive blood glucose estimation for paediatric population based on automatically extracted beat morphology. The first stage performs Morphological Extractor based on Self-Attention based Long Short-Term Memory driven by Convolutional Neural Network for highlighting local features based on temporal context. The second stage is based on Morphological Regressor driven by multilayer perceptron with dropout and batch normalization to avoid overfitting. We performed feature selection via logit model followed by Spearman's correlation among features to avoid feature redundancy. We trained as tested our model on publicly available MIT/BIH-Physionet databases and physiological signals acquired from a T1D paediatric population. We performed our evaluation via Clarke's Grid error to analyse estimation accuracy on range of blood values under different glycaemic conditions. The results show that our tool outperformed existing regression models with 89% accuracy under clinically acceptable range. The proposed model based on beat morphology significantly outperformed models based on HRV features. [ABSTRACT FROM AUTHOR]