University of Rijeka

About: University of Rijeka is a education organization based out in Rijeka, Croatia. It is known for research contribution in the topics: Population & Tourism. The organization has 3471 authors who have published 7993 publications receiving 110386 citations. The organization is also known as: Rijeka University & Sveučilište u Rijeci.

Deep Learning for Feature Extraction in Remote Sensing: A Case-Study of Aerial Scene Classification

Abstract: Scene classification relying on images is essential in many systems and applications related to remote sensing. The scientific interest in scene classification from remotely collected images is increasing, and many datasets and algorithms are being developed. The introduction of convolutional neural networks (CNN) and other deep learning techniques contributed to vast improvements in the accuracy of image scene classification in such systems. To classify the scene from areal images, we used a two-stream deep architecture. We performed the first part of the classification, the feature extraction, using pre-trained CNN that extracts deep features of aerial images from different network layers: the average pooling layer or some of the previous convolutional layers. Next, we applied feature concatenation on extracted features from various neural networks, after dimensionality reduction was performed on enormous feature vectors. We experimented extensively with different CNN architectures, to get optimal results. Finally, we used the Support Vector Machine (SVM) for the classification of the concatenated features. The competitiveness of the examined technique was evaluated on two real-world datasets: UC Merced and WHU-RS. The obtained classification accuracies demonstrate that the considered method has competitive results compared to other cutting-edge techniques.

Mechanically activated catalyst mixing for high-yield boron nitride nanotube growth

Abstract: Boron nitride nanotubes (BNNTs) have many fascinating properties and a wide range of applications. An improved ball milling method has been developed for high-yield BNNT synthesis, in which metal nitrate, such as Fe(NO3)3, and amorphous boron powder are milled together to prepare a more effective precursor. The heating of the precursor in nitrogen-containing gas produces a high density of BNNTs with controlled structures. The chemical bonding and structure of the synthesized BNNTs are precisely probed by near-edge X-ray absorption fine structure spectroscopy. The higher efficiency of the precursor containing milling-activated catalyst is revealed by thermogravimetric analyses. Detailed X-ray diffraction and X-ray photoelectron spectroscopy investigations disclose that during ball milling the Fe(NO3)3 decomposes to Fe which greatly accelerates the nitriding reaction and therefore increases the yield of BNNTs. This improved synthesis method brings the large-scale production and application of BNNTs one step closer.

Murine CMV-Induced Hearing Loss Is Associated with Inner Ear Inflammation and Loss of Spiral Ganglia Neurons

Abstract: Congenital human cytomegalovirus (HCMV) occurs in 0.5–1% of live births and approximately 10% of infected infants develop hearing loss. The mechanism(s) of hearing loss remain unknown. We developed a murine model of CMV induced hearing loss in which murine cytomegalovirus (MCMV) infection of newborn mice leads to hematogenous spread of virus to the inner ear, induction of inflammatory responses, and hearing loss. Characteristics of the hearing loss described in infants with congenital HCMV infection were observed including, delayed onset, progressive hearing loss, and unilateral hearing loss in this model and, these characteristics were viral inoculum dependent. Viral antigens were present in the inner ear as were CD3+ mononuclear cells in the spiral ganglion and stria vascularis. Spiral ganglion neuron density was decreased after infection, thus providing a mechanism for hearing loss. The lack of significant inner ear histopathology and persistence of inflammation in cochlea of mice with hearing loss raised the possibility that inflammation was a major component of the mechanism(s) of hearing loss in MCMV infected mice.

Progesterone induced blocking factor (PIBF) mediates progesterone induced suppression of decidual lymphocyte cytotoxicity.

Abstract: PROBLEM: Progesterone induced blocking factor (PIBF) is a mediator of progesterone that blocks peripheral blood lytic natural killer (NK) activity. Progesterone or PIBF stimulated decidual macrophages block up-regulation of perforin expression in decidual lymphocytes (DL). Therefore, we investigated whether progesterone regulates cytotoxicity of DL. METHOD OD STUDY: Decidual mononuclear cells were cultured with progesterone, PIBF, progesterone and anti-PIBF antibody or in the medium only. Cytolytic activity of non-adherent DL was measured by PKH-26 (red) 2 hr cytolytic assay and flow cytometry. Perforin positive DL were detected by immunofluorescency and PIBF-positive cells by immunohistology. RESULTS: Progesterone and PIBF, in a dose-dependent manner decreased cytotoxicity of DL against K-562 targets, and perforin egzocytosys was blocked. Anti-PIBF antibodies reversed the progesterone mediated reduction in cytolytic activity of DL. PIBF positive cells were found in first trimester pregnancy decidua. CONCLUSION: The results indicate possible role for PIBF, as a mediator of progesterone in regulation of DL cytolytic activity at the maternal-foetal (M-F) interface.