Thyroid nodules classification and diagnosis in ultrasound images using fine-tuning deep convolutional neural network

Ultrasonography AKA diagnostic sonography is a noninvasive imaging technique that allows the analysis of an organic structure, thanks to the ultrasonic waves. It is a valuable diagnosis method and is also seen as the evidence-based diagnostic method for thyroid nodules. The diagnosis, however, is visually made by the practitioner. The automatic discrimination of benign and malignant nodules would be very useful to report Thyroid Imaging Reporting. In this paper, we propose a fine-tuning approach based on deep learning using a Convolutional Neural Network model named resNet-50. This approach allows improving the effectiveness of the classification of thyroid nodules in ultrasound images. Experiments have been conducted on 814 ultrasound images and the results show that our proposed approach dramatically improves the accuracy of the classification of thyroid nodules and outperforms The VGG-19 model.     

Natural products as a source of environmentally friendly corrosion inhibitors of mild steel in dilute sulphuric acid: Experimental and computational studies

Protection of Metals and Physical Chemistry of Surfaces - The inhibitory effect of Acacia cyanophylla leaves extract (ACLE) on Mild steel corrosion in 1 M H2SO4 solution was investigated using...     

Crystalline properties of poly(ε‐caprolactone)‐block‐poly(vinyl acetate) block copolymers: influence of synthesis route

Poly(ε‐caprolactone)‐block‐poly(vinyl acetate) (PCL‐b‐PVAc) block copolymers were synthesized using two approaches: a ‘coupling’ approach using click chemistry reaction and a ‘macroinitiator’ route. Different copolymers, varying by their block lengths, were prepared with both methods. PCL is a semi‐crystalline polymer, and consequently PCL blocks of PCL‐b‐PVAc are able to crystallize. The purpose of this work was to analyse the influence of the method of copolymer synthesis on the crystallinity of the PCL blocks. The results indicate a significant decrease of the crystallinity of the PCL blocks in copolymers obtained using the coupling method, compared to PCL homopolymers, in contrast to copolymers obtained through the macroinitiator approach for which the crystallinity of PCL is much less affected. This influence of the synthesis method is explained by the presence, in the copolymers obtained using the click reaction, of a rigid triazol cycle binding the two blocks, limiting their mobility and decreasing the tendency of PCL to crystallize. © 2013 Society of Chemical Industry     

Preliminary chemical characterization of Tunisian monovarietal virgin olive oils and comparison with Sicilian ones

Work was carried out on the characterization of monovarietal virgin olive oils (VOO) from Tunisia and Sicily (Italy). The two main Tunisian VOO (cvv. Chétoui of the North and cv. Chemlali grown in the Center and some regions of the South) and three principal Sicilian VOO (cvv. Nocellara del Belice, Biancolilla and Cerasuola) were studied. Moreover, the Chétoui oils were tested in a rain‐fed control and an irrigation regime. All olive samples were picked at three different stages of ripeness. Analyses of major components (fatty acids and triacylglycerols) and minor ones (squalene, tocopherols and phenolic compounds) were carried out. Chétoui oils had a higher level of phenolic compounds followed by Chemlali. Generally, in the Sicilian oils these natural antioxidant contents were lower. These preliminary results indicate that it was possible to classify the Tunisian and Sicilian oils tested in their original growing area based on their chemical composition.     

Audio texturedness indicator based on a direct and reverse short listening time analysis

Multimedia Tools and Applications - In this paper, we present an objective evaluation of audio texturedness level motivated by a subjective study which emphasizes the relevance of a direct and...     

Characterization of two prickly pear species flowers growing in Tunisia at four flowering stages

Opuntia flowers belonging to two species of prickly pear were evaluated at four flowering stages: vegetative, initial flowering, full flowering and post-flowering stage, for possible use as a potential source in the food enrichment. Chemical composition including moisture, ash, protein, fat, soluble sugars, total fibers, mineral amounts and fatty acid profiles of Opuntia ficus-indica and Opuntia stricta flowers were investigated. Functional properties (swelling capacity (SWC), water holding capacity (WHC), water solubility index (WSI) and oil holding capacity (OHC)) were also studied. Results showed that during the maturation of flowers, there is a decrease in protein contents whereas fat contents increase, for both species. High minerals amounts were noticed in Opuntia flowers and they can be considered as an excellent source of minerals. The fatty acids profiles were dominated by the palmitic acid (38–59%). The techno-functional properties (SWC, WSI, WHC and OHC) were found to be important and they can be modulated according to the temperatures. Owing to its chemical profile and functional properties, Opuntia flowers could be used as an ingredient to improve different physical and nutritional properties of foods.     

Unsupervised learning of prototypes and attribute weights

In this paper, we introduce new algorithms that perform clustering and feature weighting simultaneously and in an unsupervised manner. The proposed algorithms are computationally and implementationally simple, and learn a different set of feature weights for each identified cluster. The cluster dependent feature weights offer two advantages. First, they guide the clustering process to partition the data set into more meaningful clusters. Second, they can be used in the subsequent steps of a learning system to improve its learning behavior. An extension of the algorithm to deal with an unknown number of clusters is also proposed. The extension is based on competitive agglomeration, whereby the number of clusters is over-specified, and adjacent clusters are allowed to compete for data points in a manner that causes clusters which lose in the competition to gradually become depleted and vanish. We illustrate the performance of the proposed approach by using it to segment color images, and to build a nearest prototype classifier.     

A genetic algorithm based grey goal programming (G3) approach for parts supplier evaluation and selection

The problem of part supplier selection is a major concern for all manufacturers when seeking to enhance the products’ quality and productivity. The objective of this paper is to propose an integrated genetic algorithm based grey goal programming (G³) approach to solve the part supplier selection problem. The main factor in part supplier selection is the assembly relation of the parts so as to find the suitable suppliers combination for the parts of a product. We first identify the main factors affected on supplier selection. We then present a grey-based goal programming model to work as the fitness function to evaluate the suppliers with respect to the total deviation the factors have from the ideal values. Since the objective is to find the best solution, a genetic algorithm is used to solve this problem for faster and better evaluation. The novelty of this integrated approach is to apply both qualitative and quantitative factors at once in one model and to use the grey theory to cover the lack of information of qualitative factors in order to find a solution in a near real situation.     

Effect of diethanolamine and triethanolamine on the properties of electroplated Zn–Ni alloy coatings from acid bath

In the present work, zinc–nickel alloy coatings have been produced under direct current conditions from an acid bath with different concentrations of diethanolamine (DEA) and triethanolamine (TEA). The produced coatings were analysed through chronopotentiometry, microhardness, X‐ray diffraction and MEB techniques. The compositional analysis of the films showed that the Zn–Ni electrodeposition is anomalous for all the systems. Ethanolamines augmented the anomalous behaviour. The hindering in the Ni (II) reduction will be more effective due to complexation of Ni (II) catalyst with ethanolamines. Electrochemical and structure analysis of deposits indicated the presence of γ and highly Zn‐enriched phases. The presence of these additives resulted also in coatings with finer grains. The behaviour of modified Zn–Ni alloy coatings in corrosion solution of 3% NaCl was investigated through potentiodynamic polarisation and electrochemical impedance spectroscopy. It was found that the obtained Zn–Ni/3 mM TEA alloy exhibited better corrosion resistance compared to pure Zn–Ni alloy electrodeposited at similar conditions. © 2012 Canadian Society for Chemical Engineering