Search

Breast cancer stands as a prevalent malignancy affecting women globally, and a screening method, mammography, boasts reliability for early diagnosis. Nevertheless, interpretive errors during population screening may result in false negatives and positives. To address this, Computer-Aided Detection systems rooted in deep learning have emerged, aiming to reduce both false positive and negative predictions. This study introduces an open-source toolkit called The Digital Eye for Mammography (DEM) and addressing limitations in mammography screening for mass detection and classification. The DEM comprises 11 state-of-the-art object detection architectures and uses a meticulously labeled dataset. It serves as a transfer learning source, and provides ensemble of models from diverse deep-learning architectures, resulting in a more robust solution. Experiments conducted on widely-used datasets indicate that the DEM outperforms existing transfer learning sources by significant margins in terms of true positive rate (TPR). According to the experimental results, the DEM serves as a better transfer learning source for mass detection in pathology-proven InBreast and CBIS-DDSM datasets, presenting improvements 12% and 5% in TPR performance at 0.1 false positive per image (FPPI), respectively. Compared to literature, the DEM achieves lower FPPI values while maintaining higher sensitivity, indicating its potential usage as a transfer learning source. By employing ensemble strategies, the DEM produces more reliable outcomes in our KETEM dataset, reducing FPPI by 49% for BI-RADS 1-2 (Breast Imaging Reporting and Data System) and 46% for BI-RADS 4-5 compared to the best individual model while preserving TPR values. The DEM’s results suggest its ability to attain better performance without requiring complex model hyperparameters optimization. The GitHub repository of the DEM project is publicly available on: . [ABSTRACT FROM AUTHOR]

Objective: Chemotherapy (CT) is the standard treatment for patients with metastatic esophageal cancer. Primary tumor progression is one of the important reasons for morbidity and mortality. Several studies have shown that aggressive treatment of primary tumors in metastatic patients may contribute to survival. In this study, patients with metastatic esophageal cancer and who received radiotherapy (RT) to the esophagus were evaluated. Methods: Patients with metastatic esophageal cancer treated with esophageal RT were retrospectively evaluated. ECOG 3-4 patients were not included. Analyzes were performed with SPSS 22.0 (SPSS Inc., IBM Corp., Armonk, NY). Kaplan-Meier, Log-rank, and Cox-regression tests were used for the analysis. Results: Forty-seven patients between 2009 and 2016 were evaluated. The median age was 61 (44-85); 60% of the patients were female. Eighty-nine percent of the histological subtype was squamous cell carcinoma. The mean RT dose was 45 Gy (20-68 Gy). Thirty percent of the patients had concurrent CT. The median overall survival (OS) was 14 months. 45 Gy and higher RT doses and concurrent CT were associated with better OS in univariate analysis (p=0.009, p=0.03). The median OS was nine months in patients receiving <45 Gy, and 20 months in patients who received RT ≥45 Gy; 24 months in patients who received concurrent CT, and 13 months in patients who did not. There was no significant prognostic factor in multivariate analysis. Conclusion: Forty-five Gy and higher dose RT with concurrent CT after standard treatment may contribute to survival in metastatic esophageal cancer patients with good performance scores. [ABSTRACT FROM AUTHOR]

Pristine and chromium (Cr) doped WO3 nanoflakes (NFs) with various concentrations were successfully fabricated by a facile hydrothermal technique on alumina (Al2O3) substrates. The structural, optical, and morphological properties of these NFs were investigated by x-ray diffraction, scanning electron microscopy, transmission electron microscopy, UV-Visible diffusion reflectance spectroscopy (UV–Vis DRS), and x-ray photoelectron spectroscopy. Gas sensor tests were performed against various volatile organic compounds such as ethanol, xylene, toluene, and isopropanol gases in the temperature range between 50 °C and 250 °C. According to the results of these tests, the isopropanol sensing ability of WO3 NFs is enhanced with Cr-doping due to the increasing active adsorption sites on the surface and improved surface reactions with the decreasing band gap energy. The highest isopropanol sensing response which was calculated to be 77.1 has been obtained by nominally 2% Cr doping at the optimal operating temperature of 150 °C. [ABSTRACT FROM AUTHOR]

Copyright of Arkeoloji Dergisi is the property of Ege University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Deep learning-based methods are accepted as a viable alternative to conventional statistical and geometrical methods for hyperspectral unmixing in recent years. These methods are however mainly based on linear mixture assumption on the hyperspectral data. The vast majority of presented algorithms process individual hyperspectral pixels while neglecting the spatial relationships between pixels. In order to address these two missing aspects, we propose a convolutional autoencoder-based hyperspectral unmixing method in this paper. The proposed structure incorporates the spatial neighbourhood relation with its convolutional layers in the first stage and possible non-linearities in the observed data with the included non-linear layer in the final stage. The experiments have first revealed that Adam optimizer have the best performance among different optimization methods for the proposed network. Second, the proposed method has indicated about 20–40% accuracy improvement in terms of mean squared error (MSE) metric compared to traditional hyperspectral abundance estimation methods. Third, the contribution of the non-linear layer is verified by comparing the proposed network with the conventional LMM-based autoencoder structure without the non-linear layer. Finally, the accuracy improvement for the proposed network with non-linear layer compared to the state-of-the-art deep learning-based methods using linear mixture assumption is evaluated in terms of MSE and reported as about 10% and 20% for synthetic and real data, respectively. [ABSTRACT FROM AUTHOR]

This study is motivated by the lack of knowledge in the research of mechanical characterization of thermoplastic composites (TPCs) with additional fiber hybridization. To enhance the mechanical properties of long glass fiber-reinforced polypropylene (PP) composite, hybridization via alkaline-treated aramid and carbon fabrics is performed. High performance fabrics modified with 10 wt.% sodium hydroxide (NaOH) aqueous solution are incorporated into the PP composite as reinforcements. Herewith, four arrangements (hybrid composites) for two different reinforcements and two different stacking configurations and the monolithic composite are separately investigated in terms of quasi-static perforation behavior. Failure mechanisms are also evaluated at macro level by visual observations and micro scales through a scanning electron microscopy (SEM). The experimental results provide a basis for selecting fiber-enabled hybridization and lay-up configuration with improved perforation resistance. Moreover, the influence of test temperature is reported for three different values as 20°C, 60°C, and 100°C. Based upon the results, the maximum penetration force of hybrid configuration with single-layered aramid fabric reinforcements is approximately 15.5% higher than that of single-layered carbon fabric reinforcements at 60°C test temperature. It is further observed that the absorbed energy improves as the number of fabrics is increased in both aramid and carbon reinforcements. The test temperatures significantly affect the failure mechanisms of TPCs. A smaller damaged area at the penetrated faces of the hybrid structures is obtained by comparison with the monolithic TPCs. [ABSTRACT FROM AUTHOR]

Copyright of Dokuz Eylul University Muhendislik Faculty of Engineering Journal of Science & Engineering / Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi is the property of Dokuz Eylul Universitesi Muhendislik Fakultesi Fen ve Muhendislik Dergisi and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this study, low velocity impact behavior of E-glass fiber-reinforced thermoplastic composites repaired by pressing external laminated composite patches was investigated by experimental methods. Thermoplastic composites were manufactured from polypropylene granules with two different fiber contents of 40 wt. % and 60 wt. %. Repaired specimens were prepared by using unidirectional E-glass reinforced polypropylene based thermoplastic prepregs. In order to compare the low velocity impact behavior of the repaired and unrepaired specimens, a number of single impact tests (ranging from energy levels of 10 J to 50 J) were carried out through a drop weight impact test machine with a hemispherical impactor. Low-velocity impact response of the specimens was investigated with cross-examining contact force-deformation curves and damaged specimens. Impact damages occurred in the upside and bottom surfaces of the composites were recorded from the visual inspection and compared for repaired and unrepaired specimens. According to experimental results, bending stiffness and maximum contact force of the specimens having fiber content of 60 wt.% are higher than those of 40 wt.%. Moreover, it was concluded that the patch repaired specimens have achieved a better performance in terms of maximum contact force and absorbed energy compared to the intact specimens. [ABSTRACT FROM AUTHOR]

Aim: Dosimetric comparison of three different techniques in female lymphoma patients who had radiotherapy (RT) to the neck and mediastinum. Setting and Design: Retrospective clinical study. Materials and Methods: Computerized tomography‑simulator images of eight patients were obtained retrospectively. Using 6 MV‑X photon energy, RT plans were formed with three different techniques (anterior posterior‑posterior anterior 2‑field three‑dimensional conformal RT [AP‑PA 2‑field 3D‑CRT], 4‑field 3D‑CRT and “forward” plan intensity modulated RT [FPIMRT]). Comparisons were in terms of homogeneity index (HI), conformity index (CI), and inhomogeneity coefficient for planning target volume (PTV); mean lung dose, V5Gy, V10Gy, V20Gy, V30Gy for lung; Dmean, V7.5Gy, V15Gy, V25Gy for heart; Dmean, V3.5Gy, V10Gy, V20Gy for breast; Dmax for spine; Dmean, V10Gy, V18Gy, V25Gy, V30Gy for thyroid. Statistical Analysis Used: Since nonparametric tests had to be used due to the study population being < 30, Friedman and Wilcoxon signed‑rank tests were implemented in trilateral and bilateral comparison of techniques, respectively. For statistical significance, P value was required to be <0.05. Results: When FPIMRT was compared with AP‑PA and 4‑field techniques with respect to, HI (AP‑PA/FPIMRT P : 0.017; 4‑field/ FPIMRT P: 0.03) and CI (AP‑PA/FPIMRT P: 0.018; 4‑field/FPIMRT P: 0.042), FPIMRT was more advantageous. In addition, FPIMRT was found more useful in terms of Dmax (AP‑PA/FPIMRT P: 0.012; 4‑Field/FPIMRT P: 0.012) for spinal cord and Dmean (AP‑PA/FPIMRT P: 0.012; 4‑field/FPIMRT P: 0.012) for thyroid. Conclusion: FPIMRT was superior in terms of PTV homogeneity and conformity. However, it was observed that for normal tissues, FPIMRT was advantageous only for spinal cord and thyroid; but it was not the most advantageous technique for some of the dose‑volume parameters of the breast, lung, and heart. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Textiles & Engineers / Tekstil ve Mühendis is the property of Union of Chambers of Turkish Engineers & Architects, Chamber of Textile Engineers and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this study, multilayer graphene (Gr)-reinforced cellulose composites were synthesized by using 1-ethyl-3-methylimidazolium diethylphosphonate ionic liquid. The composites were fabricated via dissolving the cellulose in 1-ethyl-3-methylimidazolium diethylphosphonate and Gr loading at different ratios (0.2, 0.4, and 0.6 wt.%). Both sides of the composites were coated with gold leaf to generate electrodes. The effect of Gr loading on chemical functional groups, crystallographic properties, thermal stability, and morphological and mechanical properties of cellulose film was investigated by Fourier transform infrared spectroscopy, X-ray diffraction, thermogravimetric analysis, scanning electron microscopy, and tensile test, respectively. Electromechanical behavior of the cellulose composite films reinforced with Gr (0.2, 0.4, and 0.6 wt.%) was investigated under DC excitation voltages of 1, 3, 5 and 7 V. Gr loading of 0.2 wt.% increased maximum tip displacement by 400% when the actuator is excited with 3 V. [ABSTRACT FROM AUTHOR]

In this study, the low-velocity impact response of sandwich composites consisting of different foam core configurations is investigated experimentally. Polyvinyl chloride foam core and glass fibers were used as core material and face sheets, respectively. A number of tests under various impact energy levels for three different sandwich composite configurations were conducted with Ceast 9350 Fractovis Plus impact testing machine in order to improve the energy absorption capacity of sandwich composites panels. Absorbed energies, maximum loads and the maximum deflection of sandwich panels were obtained for each impact energy level. As the impact energy was increased, fiber fractures at face sheets, delaminations between glass-epoxy layers, core fractures, and indentations failures were observed by visual inspection. According to the obtained results, the sandwich composite with proposed new foam core design with two internal face sheets exhibits high energy absorption capacity compared to sandwich panels formed by sandwiching a polyvinyl chloride foam core between glass fabric face sheets. Maximum contact force values decrease by increase of number of core material. [ABSTRACT FROM AUTHOR]

Hyperspectral image processing techniques, with their ability to provide information about the chemical compositions of materials, have great potential for pavement condition assessment. This study introduces a novel age-based pavement assessment method, employing an integrated algorithm with artificial neural network (ANN) and spectral angle mapping (SAM) on hyperspectral images. In the proposed method, the resulting ANN prediction outputs are used to make a new prediction along with the results from SAM scores. Tests are performed on hyperspectral images that have 360 spectral bands between 400 and 900 nm, collected by a specifically designed vehicular system for proximal image acquisition. The acquired images have eight classes, including three different pavement classes (good (5-year), medium (10-year), and poor (25-year)), yellow dye, white dye, soil, paving stone, and shadow. Several experiments are performed to evaluate the robustness of the followed methodology with limited learning data that include 5, 10, 25, and 50 samples per class, selected randomly from our independent spectral database. For a fair comparison, the individual ANN, SAM, support vector machine (SVM), and stacked auto-encoders (SAE) algorithms are also evaluated. The classification performances of individual ANN and SAM are significantly increased with their joint use, demonstrating a 1.2% to 21% classification accuracy improvement in relation to the training sample size. The study proves that the proposed approach is quite robust in cases wherein few training data are available, while SAE and standard ANN algorithms are more successful in cases wherein more learning data are present. [ABSTRACT FROM AUTHOR]