UNASSIGNED: To evaluate the performance of a federated learning framework for deep neural network-based retinal microvasculature segmentation and referable diabetic retinopathy (RDR) classification using OCT and OCT angiography (OCTA).
UNASSIGNED: Retrospective analysis of clinical OCT and OCTA scans of control participants and patients with diabetes.
UNASSIGNED: The 153 OCTA en face images used for microvasculature segmentation were acquired from 4 OCT instruments with fields of view ranging from 2 × 2-mm to 6 × 6-mm. The 700 eyes used for RDR classification consisted of OCTA en face images and structural OCT projections acquired from 2 commercial OCT systems.
UNASSIGNED: OCT angiography images used for microvasculature segmentation were delineated manually and verified by retina experts. Diabetic retinopathy (DR) severity was evaluated by retinal specialists and was condensed into 2 classes: non-RDR and RDR. The federated learning configuration was demonstrated via simulation using 4 clients for microvasculature segmentation and was compared with other collaborative training methods. Subsequently, federated learning was applied over multiple institutions for RDR classification and was compared with models trained and tested on data from the same institution (internal models) and different institutions (external models).
UNASSIGNED: For microvasculature segmentation, we measured the accuracy and Dice similarity coefficient (DSC). For severity classification, we measured accuracy, area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve, balanced accuracy, F1 score, sensitivity, and specificity.
UNASSIGNED: For both applications, federated learning achieved similar performance as internal models. Specifically, for microvasculature segmentation, the federated learning model achieved similar performance (mean DSC across all test sets, 0.793) as models trained on a fully centralized dataset (mean DSC, 0.807). For RDR classification, federated learning achieved a mean AUROC of 0.954 and 0.960; the internal models attained a mean AUROC of 0.956 and 0.973. Similar results are reflected in the other calculated evaluation metrics.
UNASSIGNED: Federated learning showed similar results to traditional deep learning in both applications of segmentation and classification, while maintaining data privacy. Evaluation metrics highlight the potential of collaborative learning for increasing domain diversity and the generalizability of models used for the classification of OCT data.

Drug repurposing has become a widely used strategy to accelerate the process of finding treatments. While classical de novo drug development involves high costs, risks, and time-consuming paths, drug repurposing allows to reuse already-existing and approved drugs for new indications. Numerous research has been carried out in this field, both in vitro and in silico. Computational drug repurposing methods make use of modern heterogeneous biomedical data to identify and prioritize new indications for old drugs. In the current paper, we present a new complete methodology to evaluate new potentially repurposable drugs based on disease-gene and disease-phenotype associations, identifying significant differences between repurposing and non-repurposing data. We have collected a set of known successful drug repurposing case studies from the literature and we have analysed their dissimilarities with other biomedical data not necessarily participating in repurposing processes. The information used has been obtained from the DISNET platform. We have performed three analyses (at the genetical, phenotypical, and categorization levels), to conclude that there is a statistically significant difference between actual repurposing-related information and non-repurposing data. The insights obtained could be relevant when suggesting new potential drug repurposing hypotheses.

The wind energy sector has seen an increasing growth in the last decade and this is foreseen to continue in the next years. This has posed several challenges in terms of skilled and prepared professionals that have always to be up to date in an industry that is constantly changing. Thus, teaching tools have gained an increasing interest. The present research reviewed the state of the art in terms of digital interactive training tools pinpointing that the existing options do not feature the user involvement in the development of the training material. Hence, the main aim of this paper is to develop and test an innovative method based on gamification to increase wind energy sector industrial skills, providing a digital interactive environment in the form of a new user-friendly software that can allow its users to train and contribute to the teaching and learning contents. The first methodological step deals with the associated background studies that were required at strategy implementation and development stages, including market analysis and technology trade-offs, as well as the general structure and the implementation steps of the software design. Obtained results pinpointed that with minimal use of web-based database and network connectivity, a mobile phone application could work in the form of a time-scored quiz application that remotely located staff at wind energy farms could benefit from. The technological innovation brought by this research will substantially improve the service of training, allowing a more dynamic formative management contributing to an improvement in the competitiveness and a step towards excellence for the whole sector.

BACKGROUND: Post-stroke depression (PSD) is a neuropsychiatric sequela that causes serious adverse effects on the prognosis of stroke patients. Our developed iPad application is a very innovative approach designed to improve participants\' depressive symptoms by presenting positive words stimuli in a video. Although this application has fewer side effects than existing pharmacological and non-pharmacological interventions and is likely less burdensome for patients and caregivers, its efficacy for PSD has not been investigated. Here we present a pilot randomized controlled trial (RCT) protocol to investigate the therapeutic potential of this application intervention for PSD patients.
METHODS: This study is designed as a 5-week, single-center, open-label, parallel-group, pilot RCT. Thirty-two patients with PSD will be randomly assigned to a combination of the iPad application and usual rehabilitation or usual rehabilitation alone (1:1 allocation ratio). The iPad application intervention lasts 3 min a day, and the usual rehabilitation lasts 3 h a day. The primary outcome is the change from baseline in The Center for Epidemiologic Studies Depression Scale score at the end of the 5-week intervention.
CONCLUSIONS: This pilot RCT is the first study to investigate the potential of iPad application interventions to reduce depressive symptoms in PSD patients. This pilot RCT determines whether this is a viable and effective intervention and informs the design for a full-scale trial. If our hypothesis is correct, this trial can provide evidence to augment the standard practice of iPad application interventions to improve depressive symptoms in patients with PSD.

Nanotechnology has enabled the discovery of a multitude of novel materials exhibiting unique physicochemical (PChem) properties compared to their bulk analogues. These properties have led to a rapidly increasing range of commercial applications; this, however, may come at a cost, if an association to long-term health and environmental risks is discovered or even just perceived. Many nanomaterials (NMs) have not yet had their potential adverse biological effects fully assessed, due to costs and time constraints associated with the experimental assessment, frequently involving animals. Here, the available NM libraries are analyzed for their suitability for integration with novel nanoinformatics approaches and for the development of NM specific Integrated Approaches to Testing and Assessment (IATA) for human and environmental risk assessment, all within the NanoSolveIT cloud-platform. These established and well-characterized NM libraries (e.g. NanoMILE, NanoSolutions, NANoREG, NanoFASE, caLIBRAte, NanoTEST and the Nanomaterial Registry (>2000 NMs)) contain physicochemical characterization data as well as data for several relevant biological endpoints, assessed in part using harmonized Organisation for Economic Co-operation and Development (OECD) methods and test guidelines. Integration of such extensive NM information sources with the latest nanoinformatics methods will allow NanoSolveIT to model the relationships between NM structure (morphology), properties and their adverse effects and to predict the effects of other NMs for which less data is available. The project specifically addresses the needs of regulatory agencies and industry to effectively and rapidly evaluate the exposure, NM hazard and risk from nanomaterials and nano-enabled products, enabling implementation of computational \'safe-by-design\' approaches to facilitate NM commercialization.

The Lipid Annotation Service (LAS) is a representational state transfer (REST) application programming interface (API) service designed to aid researchers performing lipid annotation. It assigns certainty levels (very unlikely, unlikely, likely, and very likely) to the putative annotations received as input and explains the rationale of such assignments. Its rules, obtained from the Centre for Metabolomics and Bioanalysis (CEMBIO) and from a literature review, enable LAS to extract evidence to support or refute the annotations automatically by checking the inter-rule relationships. LAS is the first metabolite annotation tool capable of explaining in natural language (English) the evidence that supports or refutes the annotations. This facilitates the understanding of the results by the user and, thus, increases the user\'s confidence in the results. Concerning its performance, in an evaluation of blood plasma samples whose compounds had previously been identified using well-established standards, LAS yielded an F-measure higher than 80%.

UNASSIGNED: For decades, moderate intensity continuous training (MICT) has been the cornerstone of exercise prescription for cardiac rehabilitation (CR). High intensity interval training (HIIT) is now recognized in CR exercise guidelines as an appropriate and efficient modality for improving cardiorespiratory fitness, a strong predictor of mortality. However, the clinical application of HIIT in a real world CR setting, in terms of feasibility, safety, and long-term adherence, needs further investigation to address ongoing reservations. Furthermore, studies using objective measures of exercise intensity (such as heart rate; HR) have produced variable outcomes. Therefore we propose investigating the use of subjective measures (such as rating of perceived exertion (RPE)) for prescribing exercise intensity.
UNASSIGNED: One hundred adults with coronary artery disease (CAD) attending a hospital-initiated CR program will be randomized to 1) HIIT: 4 × 4 min high intensity intervals at 15-18 RPE interspersed with 3-min active recovery periods or 2) MICT: usual care exercise including 40 min continuous exercise at a moderate intensity corresponding to 11-13 RPE. Primary outcome is change in exercise capacity (peak VO2) following 4 weeks of exercise training. Secondary outcome measures are: feasibility, safety, exercise adherence, body composition, vascular function, inflammatory markers, intrahepatic lipid, energy intake, and dietary behavior over 12-months; and visceral adipose tissue (VAT) following 12 weeks of exercise training.
UNASSIGNED: This study aims to address the ongoing concerns regarding the practicality and safety of HIIT in CR programs. We anticipate study findings will lead to the development of a standardized protocol to facilitate CR programs to incorporate HIIT as a standard exercise option for appropriate patients.