Objective: to elaborate and validate the content and appearance of an educational game regarding healthy lifestyle habits for adolescents Method: methodological study carried, in 2016 out in Picos (Brazil). In order to develop the initial version of the game, a survey gathered scientific information and the content. In the end, the game was divided into: home screen and the living room, kitchen and the square scene, where the sports scene was inserted. To validate the instrument, 15 experts were selected, who evaluated the content, didactics and appearance of the game. Ten adolescents evaluated the appearance and usability. In addition, one questionnaire was used for the specialists and another for the adolescents. The Content Validation Index was used to validate the game, which considered the instrument and the items as validated when an index ≥ 0.78 was obtained. Results: the room scene had access to the kitchen and the square. The sport scene was designed for physical activity and the purchase of food, which would be consumed in the kitchen scene. The experts gave the game a Content Validation Index of 0.88 for the item "objectives", 0.87 for "structure and presentation" and 0.99 for "relevance", reaching a overall Content Validation Index of 0.89. For adolescents, the item "appearance" reached the index 0.97 and for the other items, 1.0 and overal index of 0.99. Conclusion: the game was validated in terms of content and appearance and could be validated clinically with adolescents as an incentive to adopt protective behaviors for their health.
This article proposes a system for Content-Based Image Retrieval (CBIR) using stochastic distance for Synthetic-Aperture Radar (SAR) images. The methodology consists of three essential steps for image retrieval. First, it estimates the roughness () and scale () parameters of the distribution that models SAR data in intensity. The parameters of the model were estimated using the Maximum Likelihood Estimation and the fast approach of the Log-Cumulants method. Second, using the triangular distance, CBIR-SAR evaluates the similarity between a query image and images in the database. The stochastic distance can identify the most similar regions according to the image features, which are the estimated parameters of the data model. Third, the performance of our proposal was evaluated by applying the Mean Average Precision (MAP) measure and considering clippings from three radar sensors, i.e., UAVSAR, OrbiSaR-2, and ALOS PALSAR. The CBIR-SAR results for synthetic images achieved the highest MAP value, retrieving extremely heterogeneous regions. Regarding the real SAR images, CBIR-SAR achieved MAP values above 0.833 for all polarization channels for image samples of forest (UAVSAR) and urban areas (ORBISAR). Our results confirmed that the proposed method is sensitive to the degree of texture, and hence, it relies on good estimates. They are inputs to the stochastic distance for effective image retrieval.
O glaucoma é uma das principais causas de cegueira no mundo, afetando o nervo óptico e causando danos progressivos. A detecção precoce é essencial para evitar a perda da visão, mas o diagnóstico manual baseado em imagens retinianas pode ser subjetivo e propenso a erros. Propõe-se o uso de deep learning com o modelo YOLOv8n para detectar o disco óptico, seguido da classificação com uma ResNet-50 refinada. O método obteve 91,7% de precisão, 91,1% de recall, 91,5% de acurácia e 82,7% no coeficiente kappa. Conclui-se que o uso da imagem completa da retina teve melhor desempenho do que o uso exclusivo da região do disco óptico, conforme confirmado pela análise de ativação do Grad-CAM.