A novel comparative study for detection of Covid-19 on CT lung images using texture analysis,machine learning,and deep learning methods

The Covid-19 virus outbreak that emerged in China at the end of 2019 caused a huge and devastating effect worldwide. In patients with severe symptoms of the disease, pneumonia develops due to Covid-19 virus. This causes intense involvement and damage in lungs. Although the emergence of the disease occurred a short time ago, many literature studies have been carried out in which these effects of the disease on the lungs were revealed by the help of lung CT imaging. In this study, 1.396 lung CT images in total (386 Covid-19 and 1.010 Non-Covid-19) were subjected to automatic classification. In this study, Convolutional Neural Network (CNN), one of the deep learning methods, was used which suggested automatic classification of CT images of lungs for early diagnosis of Covid-19 disease. In addition, k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM) was used to compare the classification successes of deep learning with machine learning. Within the scope of the study, a 23-layer CNN architecture was designed and used as a classifier. Also, training and testing processes were performed for Alexnet and Mobilenetv2 CNN architectures as well. The classification results were also calculated for the case of increasing the number of images used in training for the first 23-layer CNN architecture by 5, 10, and 20 times using data augmentation methods. To reveal the effect of the change in the number of images in the training and test clusters on the results, two different training and testing processes, 2-fold and 10-fold cross-validation, were performed and the results of the study were calculated. As a result, thanks to these detailed calculations performed within the scope of the study, a comprehensive comparison of the success of the texture analysis method, machine learning, and deep learning methods in Covid-19 classification from CT images was made. The highest mean sensitivity, specificity, accuracy, F-1 score, and AUC values obtained as a result of the study were 0,9197, 0,9891, 0,9473, 0,9058, 0,9888; respectively for 2-fold cross-validation, and they were 0,9404, 0,9901, 0,9599, 0,9284, 0,9903; respectively for 10-fold cross-validation.

     

Medical thermograms’ classification using deep transfer learning models and methods

Multimedia Tools and Applications - Infrared thermal imaging and deep learning provide intelligent monitoring systems that detect diseases in early phases. However, deep learning models require...     

A new application for the valorisation of pomegranate seed oil: nanoencapsulation of pomegranate seed oil into electrospun nanomats for food preservation

Pomegranate seed oil (PSO) contains many bioactive materials including antimicrobials, antioxidants, tocopherol and unsaturated fatty acids such as punicic acids. Utilising PSO with nanotechnological ways is a novel approach. Therefore, in this study, PSO-loaded nanomats having an average diameter of 327 nm with 97.6% encapsulation efficiency were produced. Then, the protection potential of nanomats was determined in terms of the microbial and oxidative deterioration of food samples. On the 1st day of storage, the TMAB load of the control kashar cheese was 4.35 log CFU g⁻¹, while it was 3.05 log CFU g⁻¹ in the coated cheese (change: 1.3 logs). On the 20th day of storage, the TMAB load of the CK sample and PSc sample was 5.52 and 4.22 log CFU g⁻¹, respectively (change: 1.3 logs). For fish fillets, nanoencapsulated PSO enabled a bacterial reduction of 1.22 log cycles after 9 days of storage. The total mould and yeast number of cheese samples increased during storage, but the increase was lower in the coated group. Nanomats also increased the oxidative stability of food samples. Thiobarbituric acid values of coated samples were lower than uncoated samples.     

Effect of an Er,Cr:YSGG laser preparation on dentin bond strength of a universal adhesive

The aim of this study was to evaluate the bond strength of a universal adhesive system to dentin prepared with SiC paper or an Er,Cr:YSGG laser using different bonding strategies (etch-and-rinse versus self-etch mode). Ninety-six extracted caries-free, sound human molars were used. The teeth were longitudinally sectioned in the mesiodistal direction and were wet polished with 600-grit SiC paper to obtain a standardized flat dentin surface. All prepared teeth were randomly divided into two groups, according to the surface preparation method: GroupI:an erbium, chromium:yttrium,scandium, gallium, garnet laser; Group II: silicon carbide paper[SiC] (n = 48). Each group was then assigned into three subgroups according to the universal adhesive’s (Single Bond Universal) bonding strategies: (a) etch-and-rinse mode with phosphoric acid, (b) etch-and-rinse mode with a laser, (c) self-etch mode (n = 16). For surface preparation, the Er,Cr:YSGG laser was used at 3 W, 30 Hz with 140 μs pulse duration for 25 s. For etching mode, the laser was used at 1.5 W (60% air, 70% water). Cylinders of composite were fabricated on the bonding area and shear bond strength was determined using a universal testing machine. Failure modes were evaluated using a stereomicroscope. The data were analyzed using two-way ANOVA followed by the Bonferroni test (p < 0.05). Bonding strategies showed statistically significant differences in both the SiC-and laser-prepared groups (p < 0.05).Universal adhesive used in etch-and-rinse mode with acid showed significantly higher bond strength values than in self-etch mode (p < 0.05). The bond strength values did not differ according to the surface preparation method (p > 0.05). Irrespective of preparation method, using universal adhesive in etch-and-rinse mode with acid might improve dentin bond strength. Laser preparation did not affect the bond strength of the universal adhesive tested.     

Characterization of titanium aluminide alloy components fabricated by additive manufacturing using electron beam melting

Intermetallic, γ-TiAl, equiaxed, small-grain (～2 μm) structures with lamellar γ/α₂-Ti₃Al colonies with average spacing of 0.6 μm have been fabricated by additive manufacturing using electron beam melting (EBM) of precursor, atomized powder. The residual microindentation (Vickers) hardness (HV) averaged 4.1 GPa, corresponding to a nominal yield strength of ～1.4 GPa (～HV/3), and a specific yield strength of 0.37 GPa cm³ g^?1 (for a density of 3.76 g cm^?3), in contrast to 0.27 GPa cm³ g^?1 for EBM-fabricated Ti–6Al–4V components. These results demonstrate the potential to fabricate near net shape and complex titanium aluminide products directly using EBM technology in important aerospace and automotive applications.     

Synthesis of ZnO: Ge nanocomposite thin films by plasma gas condensation

In this work, we introduce a new method for the synthesis of Ge nanoparticles embedded ZnO thin films that are considered to be a potential candidate for photovoltaic applications. As opposed to current techniques, for the independent preparation of Ge nanoparticles, we propose using Cluster Deposition Source (CDS), which utilizes gas condensation of sputtered Ge atoms. For the synthesis of ZnO thin film host material conventional sputtering technique is employed. In the proposed technique independently synthesized Ge nanoparticles and ZnO thin films are combined into a composite structure on (100) oriented Si substrates. X-ray diffraction (XRD) patterns of the samples have revealed that Ge nanoparticles preferentially settle on (113) planes on top of the (002) oriented ZnO layer. It is realized that Ge nanoparticles with sizes ranging from 16 nm to 20 nm could be embedded into a well-defined ZnO matrix. In fact, transmission electron microscopy (TEM) studies performed on Ge nanoparticles captured on a Cu grids placed just above the substrate during deposition for about 60 s have manifested that Ge nanoparticles reach to ZnO matrix as clusters composed of particles with sizes of about 7–8 nm and then eventually grow larger due to substrate heating implemented during capping layer deposition. Optical absorption measurements have revealed that Ge nanoparticle inclusion lead to an additional absorption edge at about 2.75 eV along with 3.17 eV edge resulting from ZnO host.     

Boron removal from aqueous solutions by activated carbon impregnated with salicylic acid

In this study, the removal of boric acid from aqueous solution by activated carbon impregnated with salicylic acid was studied in batch system. pH, adsorbent amount, initial boron concentration, temperature, shaking rate and salicylic acid film thickness were chosen as parameters. Boron removal efficiencies increased with increasing adsorbent amount, temperature and pH, decreasing initial boron concentration. As thickness of salicylic acid film on activated carbon becomes thin up to 0.088nm, the efficiency increased, and then, the efficiency decreased with becoming thinner than 0.088nm of salicylic acid film. Shaking rate was no effect on removal efficiency. In result, it was determined that the use of salicylic acid as an impregnant for activated carbon led to the increase of the amount of boron adsorbed. A lactone ring, being the most appropriate conformation, forms between boric acid and -COOH and -OH groups of salicylic acid.     

Evaluation of Pyrite in a Tailings Dam by Flotation

Pyrite is presented as a gangue mineral in most of complex sulphide ores and most of time it is removed as a tailing. However, beneficiation of pyrite from railings could be important because of its economical value and the environmental concern. This paper presents a research for the enrichment of pyrite from copper plant tailings, Siirt-Madenkoy/Turkey, by flotation. Siirt-Madenkoy copper mine is operated by Park Electric Production Mining and Industry since 2005. In order to obtain the pyrite from Siirt-Madenkoy copper plant tailings, laboratory test work has to be done to determine the type of reagents, the flowsheet and the size of plant. The objective of the company is to develop a pyrite flotation circuit for concentration of pyrite from tailing dam to obtain 750,000 tons/year pyrite concentrate. There are about 4 million tons tailings that containing pyrite in the railings dam. Experiments were carried out with three different types of tailings from the plant and the pyrite concentrate with 46%Fe and 90% recovery has been obtained. Therefore, this study shows that high grade pyrite with high recovery can be recovered from the tailings.     

A new conducting polymer of 2,5-bis(2-thienyl)-1H-(pyrrole) (SNS) containing carbazole subunit: Electrochemical, optical and electrochromic properties

We report here the synthesis of a new electroactive monomer 3,6-di-tert-butyl-[4-(2,5-di-2-thienyl-1H-pyrrol-1-yl)phenyl]-9H-carbazole in five steps and polymerization of this monomer by two different procceses: Kumada Coupling and electro-oxidative. The product obtained by chemical polymerization exhibits a high thermal stability and narrow molecular weight distribution. While, in the UV–vis absorption spectrum of monomer the absorption peaks appear at 338 nm, the chemically synthesized polymer absorbs at 428 nm with 90 nm red shift. Cyclic voltammogram of monomer shows two separate oxidation processes which reflect the first oxidation at +0.84 V and a second one at +1.43 V. When the polymeric film prepared by electrochemical process was subjected to a repeated cyclic scan between −0.2 V and +1.0 V, it switches among three different colors (orange, green and blue). The oxidation and reduction response times were calculated as 2.0 s for this polymer film and exhibits high redox stability. The results anticipate that this kind of polymers can be used for designing electrochromics based on the use of one molecule for the generation of three basic colors (RGB).