Towards an efficient and robust foot classification from pedobarographic images

O documento em anexo encontra-se na versão post-print (versão corrigida pelo editor).

Image upsampling algorithms applied on Nuclear Medicine: comparision of hqnx and nxSaI algorithms families with two interpolation algorithms - nearest neighbour and bicubic interpolation

Introduction: Image resizing is a normal feature incorporated into the Nuclear Medicine digital imaging. Upsampling is done by manufacturers to adequately fit more the acquired images on the display screen and it is applied when there is a need to increase - or decrease - the total number of pixels. This paper pretends to compare the “hqnx” and the “nxSaI” magnification algorithms with two interpolation algorithms – “nearest neighbor” and “bicubic interpolation” – in the image upsampling operations. Material and Methods: Three distinct Nuclear Medicine images were enlarged 2 and 4 times with the different digital image resizing algorithms (nearest neighbor, bicubic interpolation nxSaI and hqnx). To evaluate the pixel’s changes between the different output images, 3D whole image plot profiles and surface plots were used as an addition to the visual approach in the 4x upsampled images. Results: In the 2x enlarged images the visual differences were not so noteworthy. Although, it was clearly noticed that bicubic interpolation presented the best results. In the 4x enlarged images the differences were significant, with the bicubic interpolated images presenting the best results. Hqnx resized images presented better quality than 4xSaI and nearest neighbor interpolated images, however, its intense “halo effect” affects greatly the definition and boundaries of the image contents. Conclusion: The hqnx and the nxSaI algorithms were designed for images with clear edges and so its use in Nuclear Medicine images is obviously inadequate. Bicubic interpolation seems, from the algorithms studied, the most suitable and its each day wider applications seem to show it, being assumed as a multi-image type efficient algorithm.

Drilling delamination outcomes on glass and sisal reinforced plastics

Nowadays, fibre reinforced plastics are used in a wide variety of applications. Apart from the most known reinforcement fibres, like glass or carbon, natural fibres can be seen as an economical alternative. However, some mistrust is yet limiting the use of such materials, being one of the main reasons the inconsistency normally found in their mechanical properties. It should be noticed that these materials are more used for their low density than for their high stiffness. In this work, two different types of reinforced plates were compared: glass reinforced epoxy plate and sisal reinforced epoxy plate. For material characterization purposes, tensile and flexural tests were carried out. Main properties of both materials, like elastic modulus, tensile strength or flexural modulus, are presented and compared with reference values. Afterwards, plates were drilled under two different feed rates: low and high, with two diverse tools: twist and brad type drill, while cutting speed was kept constant. Thrust forces during drilling were monitored. Then, delamination area around the hole was assessed by using digital images that were processed using a computational platform previously developed. Finally, drilled plates were mechanically tested for bearing and open-hole resistance. Results were compared and correlated with the measured delamination. Conclusions contribute to the understanding of natural fibres reinforced plastics as a substitute to glass fibres reinforced plastics, helping on cost reductions without compromising reliability, as well as the consequence of delamination on mechanical resistance of this type of composites.

Damage analysis of carbon/epoxy plates after drilling

Drilling of composites plates normally uses traditional techniques but damage risk is high. NDT use is important. Damage in a carbon/epoxy plate is evaluated by enhanced X-rays. Four different drills are used. The images are analysed using Computational Vision techniques. Surface roughness is compared. Results suggest strategies for delamination reduction.

High-Content Analysis of Breast Cancer Using Single-Cell Deep Transfer Learning

High-content analysis has revolutionized cancer drug discovery by identifying substances that alter the phenotype of a cell, which prevents tumor growth and metastasis. The high-resolution biofluorescence images from assays allow precise quantitative measures enabling the distinction of small molecules of a host cell from a tumor. In this work, we are particularly interested in the application of deep neural networks (DNNs), a cutting-edge machine learning method, to the classification of compounds in chemical mechanisms of action (MOAs). Compound classification has been performed using image-based profiling methods sometimes combined with feature reduction methods such as principal component analysis or factor analysis. In this article, we map the input features of each cell to a particular MOA class without using any treatment-level profiles or feature reduction methods. To the best of our knowledge, this is the first application of DNN in this domain, leveraging single-cell information. Furthermore, we use deep transfer learning (DTL) to alleviate the intensive and computational demanding effort of searching the huge parameter's space of a DNN. Results show that using this approach, we obtain a 30% speedup and a 2% accuracy improvement.

Aplicação de um método de correlação de imagem para a determinação da tenacidade à fratura em corte de adesivos estruturais

Qualquer estrutura hoje em dia deve ser resistente, robusta e leve, o que aumentou o interesse industrial e investigação nas ligações adesivas, nomeadamente pela melhoria das propriedades de resistência e fratura dos materiais. Com esta técnica de união, o projeto de estruturas pode ser orientado para estruturas mais leves, não só em relação à economia direta de peso relativamente às juntas aparafusas ou soldadas, mas também por causa da flexibilidade para ligar materiais diferentes. Em qualquer área da indústria, a aplicação em larga escala de uma determinada técnica de ligação supõe que estão disponíveis ferramentas confiáveis para o projeto e previsão da rotura. Neste âmbito, Modelos de Dano Coesivo (MDC) são uma ferramenta essencial, embora seja necessário estimar as leis MDC do adesivo à tração e corte para entrada nos modelos numéricos. Este trabalho avalia o valor da tenacidade ao corte (GIIC) de juntas coladas para três adesivos com ductilidade distinta. O trabalho experimental consiste na caracterização à fratura ao corte da ligação adesiva por métodos convencionais e pelo Integral-J. Além disso, pelo integral-J, é possível definir a forma exata da lei coesiva. Para o integral-J, é utilizado um método de correlação de imagem digital anteriormente desenvolvido para a avaliação do deslocamento ao corte do adesivo na extremidade da fenda (δs) durante o ensaio, acoplado a uma sub-rotina em Matlab® para a extração automática de δs. É também apresentado um trabalho numérico para avaliar a adequabilidade de leis coesivas triangulares aproximadas em reproduzir as curvas força-deslocamento (P-δ) experimentais dos ensaios ENF. Também se apresenta uma análise de sensibilidade para compreender a influência dos parâmetros coesivos nas previsões numéricas. Como resultado deste trabalho, foram estimadas experimentalmente as leis coesivas de cada adesivo pelo método direto, e numericamente validadas, para posterior previsão de resistência em juntas adesivas. Em conjunto com a caraterização à tração destes adesivos, é possível a previsão da rotura em modo-misto.