A new deep convolutional neural network model for classifying breast cancer histopathological images and the hyperparameter optimisation of the proposed model

The Journal of Supercomputing - Deep learning algorithms have yielded remarkable results in medical diagnosis and image analysis, besides their contribution to improvements in a number of fields...     

Predicting direction of stock price index movement using artificial neural networks and support vector machines: The sample of the Istanbul Stock Exchange

Prediction of stock price index movement is regarded as a challenging task of financial time series prediction. An accurate prediction of stock price movement may yield profits for investors. Due to the complexity of stock market data, development of efficient models for predicting is very difficult. This study attempted to develop two efficient models and compared their performances in predicting the direction of movement in the daily Istanbul Stock Exchange (ISE) National 100 Index. The models are based on two classification techniques, artificial neural networks (ANN) and support vector machines (SVM). Ten technical indicators were selected as inputs of the proposed models. Two comprehensive parameter setting experiments for both models were performed to improve their prediction performances. Experimental results showed that average performance of ANN model (75.74%) was found significantly better than that of SVM model (71.52%).     

The effect of H. pylori in perforation of duodenal ulcer

BACKGROUND/AIMS: H. pylori has been described as an opportunistic pathogen attracted by changes in the gastric mucosa caused by inflammation and ulceration. However, the role of H. pylori infection in the perforation of duodenal ulcers has not yet been clearly determined. The aim of this study was to assess the prevalence of H. pylori infection in patients undergoing laparotomy for repair of a perforated duodenal ulcer. METHODOLOGY: Patients who underwent surgery for a perforated duodenal ulcer in our Surgical Unit between January 1994 and July 1996 were included in this study. The study population consisted of eighteen patients with a mean age of 32.7 (21-48) years. All of the patients were male. Patients with chronic duodenal ulcer perforation and with no contraindications for definitive surgery, such as peritonitis, shock (blood pressure <90 mm Hg), age >60 years, or more than a 12-hour elapse from the time of perforation, were treated by bilateral truncal vagotomy and Weinberg pyloroplasty. The ulcer was excised with the pyloric ring. The cut was then extented by about 2 cm on both the gastric and duodenal sides. Two biopsies were taken from the antral mucosa by endoscopic biopsy forceps. The defect was closed transversely. The ulcer specimen and the antral biopsies were fixed separately in 10% formalin solution and sent to the department of Histopathology. The specimens were stained with Hematoxylin-Eosin and examined for H. pylori . Sections of the ulcer specimen were especially investigated for the presence of H. pylori through all layers of the ulcer. RESULTS: H. pylori was found in the antral biopsies of 16 patients (88.8%). In seven of the ulcer specimens (38.8%), H. pylori was present in the mucosa and also extended through the wall of the ulcer. H. pylori was positive in the antral biopsies of all patients with H. pylori present in the ulcer wall. CONCLUSIONS: In our study, H. pylori was present at a high ratio in the antral biopsies of patients with duodenal ulcer perforation. The presence of H. pylori throughout the ulcer wall to a considerable extent emphasizes the fact that eradication of H. pylori is important in the treatment of perforated duodenal ulcer.     

Percutaneous treatment of a congenital splenic cyst with alcohol: a new therapeutic approach

Percutaneous treatment of a huge congenital splenic cyst in a 23-year-old man is presented. The cyst had been catheterized and drained two times within a 3-month period without injecting any sclerosing agent into the cavity. On the third attempt, catheter drainage and injection of alcohol into the cyst cavity were performed because of insufficient response to drainage alone. He was discharged symptom-free after the procedure. The cyst diminished in size considerably 9 months after the treatment with alcohol. The volume of the cyst was reduced from 5200 to 8 ml. Although percutaneous treatment of a congenital splenic cyst with tetracyclin has been reported, to our knowledge this is the first case of a congenital splenic cyst treated with alcohol as a sclerosing agent. Percutaneous treatment of splenic cyst can obviate the need for partial or total splenectomy and may be an alternative to surgical treatment.     

Encountering the Urban Crisis: The Gezi Event and the Politics of Urban Design

This article addresses an experimental urban design studio conducted in Bilkent University in Ankara, which problematized the protests that initially started in Gezi Park in Istanbul and shook Turkey in the summer of 2013. As will be argued in detail below, we claim that the Gezi event represents an urban crisis. The particular event was the rapid escalation of a small protest against the destruction of a public space into a nationwide anti-government insurrection. But it also represented a larger urban crisis: the increasing influence of neoliberalism on the city, as the protests were the outcome of a period marked by zealous commodification of urban space. In both instances, such urban conflicts have to be addressed by urban designers, since they produce the renewal projects that gentrify urban spaces and, at the same time, seek possible alternatives for a better urban environment.     

Energy Spectra of the Local States in the Forbidden Gap of Monoclinic TlInS 2x Se 2(1−x) Crystals

Near IR properties of the mixed TlInS_2xSe_2(1?x) have been studied previously by the present authors. In this work the temperature and frequency dependence's of the conductivity and the current-voltage characteristics (in relatively weak electric field), have been investigated for monoclinic TlInS_2xSe_2(1?x) crystals, which are perspective materials for IR applications. From the temperature dependence's of conductivity in the direction perpendicular to c- axis the band gap E^g = 2.22 eV was determined for β--TlInS₂ crystals. The impurity centres were determined located at 0.43, 0.73 eV and 0.35, 0.48, 1.12 eV for the direction of current i//c and i ⊥ c, respectively. The concentration of the centres located at 0.48 and 1.12 eV were calculated to be N_A ? N_D = 4.8 · 10⁹ cm^?3 and 1.9 · 10¹¹ cm^?3, respectively. It was found that in the solid solutions TlInS_2xSe_2(1?x) for 0.3 ≤ x ≤ 1, the conductivity follows the dependence σ (v) = σ₀·υ^s in the temperature range between 100 to 600 K. In the temperature range of 80-400 K charge bounce plays an important role in the conductivity mechanism. Occurrence of the deep and low-levels impurity centres and a “tail” of the density of energy states in TlInS_2xSe_2(1?x) crystals make them perspective for practical applications: switching and memory effects, N-type current-voltage characteristics, induced conductivity etc.     

Photoconductivity in the Ternary Layered Semiconductors

Spectral dependences of the photoconductivity, near-band-edge absorption were studiet at room and low temperatures for a number of TlInS2 and TlInSl2 type layered ternary chalcogenides. Some peculiarities in the temperature dependences of the spectral properties are assigned as due to a low temperature phase transition to an incommensurate and ferroelectric state. Temperature coefficients of the band gap for TlInSe2 are determined. Spectral properties of TlInS2 doped with Nd and Yb are also presented and discussed. It is shown that the long-wave lenght absorption edge of TlInS2 crystal obeys the Urbach's rule. High photosensitivity in the visible and near IR range of spectra makes these materials attractive for optoelectornics application.     

IR-Detectors Based on IN2O3-Anode Oxide-CDxHg1-xTe

High sensitive IR detectors were developed by using noncompensated Cd_xHg_1-xTe (x=0.25÷0.3) with the field-effect electrode from In_xO₃, which were received by magnetron spattering onto the anodized surface of the single crystal. Photoelectric properties (spectral distribution of the photoconductivity, time-response, the uniformity of the spatial distribution of sensitivity) were investigated under different potential applied to the field-effect electrode. For high sensitive material the detectivity D* = 8·10¹¹ cm Hz^1/2 W^-1was obtained at the wavelength of 5 μm in the aperture angle of 30°. It was shown that the semitransparent field-effect electrode made from In₂O₃ allows to rule the output parameters of the detector in a wide range.     

Predicting bank financial failures using neural networks,support vector machines and multivariate statistical methods: A comparative analysis in the sample of savings deposit insurance fund (SDIF) transferred banks in Turkey

Bank failures threaten the economic system as a whole. Therefore, predicting bank financial failures is crucial to prevent and/or lessen the incoming negative effects on the economic system. This is originally a classification problem to categorize banks as healthy or non-healthy ones. This study aims to apply various neural network techniques, support vector machines and multivariate statistical methods to the bank failure prediction problem in a Turkish case, and to present a comprehensive computational comparison of the classification performances of the techniques tested. Twenty financial ratios with six feature groups including capital adequacy, asset quality, management quality, earnings, liquidity and sensitivity to market risk (CAMELS) are selected as predictor variables in the study. Four different data sets with different characteristics are developed using official financial data to improve the prediction performance. Each data set is also divided into training and validation sets. In the category of neural networks, four different architectures namely multi-layer perceptron, competitive learning, self-organizing map and learning vector quantization are employed. The multivariate statistical methods; multivariate discriminant analysis, k-means cluster analysis and logistic regression analysis are tested. Experimental results are evaluated with respect to the correct accuracy performance of techniques. Results show that multi-layer perceptron and learning vector quantization can be considered as the most successful models in predicting the financial failure of banks.