130553 research outputs found
Sort by
A Comparative Analysis of Deep Learning Architectures for Segmentation in Lung
No full textThis study explores the application of deep learning techniques to segment lung computed tomography (CT) scans, with a focus on cases involving COVID-19 and lung tumors. Utilizing a diverse dataset encompassing a wide range of CT scans, we conduct an extensive evaluation of various state-of-the-art deep neural network architectures. Our experimental results demonstrate the high efficiency and accuracy of deep learning models in performing image segmentation tasks, achieving impressive dice scores of 95.12% and 82.89% on COVID-19 and lung tumor data, respectively. These findings highlight the signif-icant potential of deep learning in medical imaging applications. Furthermore, we conduct thorough ablation studies, meticulously analyzing the performance of each network architecture. These studies provide valuable insights into the specific strengths and limitations of different deep learning approaches, facilitating the identification of the most effective methods for lung CT scan segmentation. This research not only underscores the promising capabilities of deep learning in medical image analysis but also offers a detailed understanding of how various models can be optimized to enhance performance in clinical applications
Deep Ensembling with Multimodal Image Fusion for Efficient Classification of Lung Cancer
No full textThis study focuses on the classification of cancerous and healthy slices from multimodal lung images. The data used in the research comprises Computed Tomography (CT) and Positron Emission Tomography (PET) images. The proposed strategy achieves the fusion of PET and CT images by utilizing Principal Component Analysis (PCA) and an Autoencoder. Subsequently, a new ensemble-based classifier developed, Deep Ensembled Multimodal Fusion (DEMF), employing majority voting to classify the sample images under examination. Gradient-weighted Class Activation Mapping (Grad-CAM) employed to visualize the classification accuracy of cancer-affected images. Given the limited sample size, a random image augmentation strategy employed during the training phase. The DEMF network helps mitigate the challenges of scarce data in computer-aided medical image analysis. The proposed network compared with state-of-the-art networks across three publicly available datasets. The network outperforms others based on the metrics - Accuracy, F1Score, Precision, and Recall. The investigation results highlight the effectiveness of the proposed network
Synergistic effect of dendritic fibrous nanosilica and In<sub>2</sub>O<sub>3</sub> photocatalysts for enhanced visible-light-driven hydrogen generation
No full textProducing green hydrogen from water using photocatalysts and solar energy is a pivotal strategy in combating climate change by adopting renewable energy sources. Herein we report the synthesis of a novel dendritic fibrous nanosilica and In2O3 composite (DFNS/In2O3) via a solvothermal method. Comprehensive characterizations of the crystal phase, morphology, and optical absorption properties of DFNS/In2O3 were conducted using powder X-ray diffraction, field emission scanning electron microscopy, transmission electron microscopy, 29Si cross-polarization magic angle spinning nuclear magnetic resonance, UV–Vis diffuse reflectance spectroscopy, Brunauer–Emmett–Teller analysis, and thermogravimetric analysis. The formation of an interface between In2O3 nanoparticles and the DFNS surface facilitates the charge separation, thereby improving the photocatalytic efficiency. The DFNS/In2O3 (30%) photocatalyst displayed a remarkable 23-fold increase in the hydrogen generation rate (1067 μmol h–1 g–1cat) compared to pristine In2O3 (45.79 μmol h–1 g–1cat). This enhancement is attributed to the superior light harvesting capability of DFNS owing to multiple light scattering events and the effective dispersion of In2O3 on the fibrous surface of DFNS, which improves water diffusion and interaction with active catalytic sites. This study presents a unique outlook on the development of new photocatalytic systems, combining In2O3 nanomaterials with an optimal band gap (2.8 eV) for photocatalytic water splitting and DFNS with its inherent high light harvesting capacity due to its fibrous nature and increased surface area
Enhanced efficiency in plastic waste upcycling: the role of mesoporosity and acidity in zeolites
No full textBy modulating zeolite confinement and improving pore diffusion properties, addressing a significant limitation in current plastic waste upcycling methodologies is essential. In this work, we have developed mesoporous zeolites that exhibit enhanced diffusion capabilities for long-chain polymers without compromising the crystalline structure. The mesopore volume doubled from 0.14 cm3 g−1 (CBV720) to 0.28 cm3 g−1 (M7203h) after zeolite modification. This has enabled to overcome the inefficiencies associated with polymer diffusion in conventional zeolites, significantly advancing the catalytic conversion of plastic waste into valuable products. Catalytic pyrolysis experiments on various polyethylenes underline the superior performance of mesoporous zeolites, especially for highly branched polymer structures where degradation temperatures are reduced by 29 °C compared to conventional zeolites, highlighting the importance of pore arrangement. Detailed analysis using NH3-TPD and in situ DRIFT spectroscopy reveals the crucial role of Brønsted acid sites in enhancing degradation efficiency. The optimized mesoporous zeolite catalyst, M720cit, showed excellent effectiveness in reducing degradation temperatures for a wide range of daily-use plastic waste. The T10 values were significantly reduced for various plastic wastes: food packaging dropped to 208 °C (from 354 °C), plastic bottles to 349 °C (from 381 °C), and milk packets to 277 °C (from 409 °C), among others. Moreover, the well-retained microstructure of the M720 catalyst yielded a very similar product distribution despite the introduction of mesoporosity. This study not only surmounts crucial obstacles in the modulation of zeolite confinement and the enhancement of pore diffusion properties but also augments the economic and environmental sustainability of plastic waste conversion processes
Confronting risks of mirror life
No full textAll known life is homochiral. DNA and RNA are made from “righthanded” nucleotides, and proteins are made from “left-handed” amino acids. Driven by curiosity and plausible applications, some researchers had begun work toward creating lifeforms composed entirely of mirror-image biological molecules. Such mirror organisms would constitute a radical departure from known life, and their creation warrants careful consideration. The capability to create mirror life is likely at least a decade away and would require large investments and major technical advances; we thus have an opportunity to consider and preempt risks before they are realized. Here, we draw on an indepth analysis of current technical barriers, how they might be eroded by technological progress, and what we deem to be unprecedented and largely overlooked risks (1). We call for broader discussion among the global research community, policy-makers, research funders, industry, civil society, and the public to chart an appropriate path forward
Modified Macdonald polynomials and the multispecies zero range process: II
No full textIn a previous part of this work, we gave a new tableau formula for the modified Macdonald polynomials H̃λ(X;q,t), using a weight on tableaux involving the queue inversion (quinv) statistic. In this paper we explicitly describe a connection between these combinatorial objects and a class of multispecies totally asymmetric zero range processes (mTAZRP) on a ring, with site-dependent jump-rates. We construct a Markov chain on the space of tableaux of a given shape, which projects to the mTAZRP, and whose stationary distribution can be expressed in terms of quinv-weighted tableaux. We deduce that the mTAZRP has a partition function given by the modified Macdonald polynomial
H̃λ(X;q,t). The novelty here in comparison to previous works relating the stationary distribution of integrable systems to symmetric functions is that the variables x1,..., xn are explicitly present as hopping rates in the mTAZRP. We also obtain interesting symmetry properties of the mTAZRP probabilities under permutation of the jump-rates between the sites. Finally, we explore a number of interesting special cases of the mTAZRP, and give explicit formulas for particle densities and correlations of the process purely in terms of modified Macdonald polynomials
Effect of particle stiffness on microgel self-assembly and suspension phase behavior over a broad temperature range
No full textWe synthesized thermoresponsive poly(N-isopropylacrylamide) (PNIPAM) colloidal microgel particles of different stiffnesses by controlling the concentration of a polar crosslinker in a precipitation polymerization synthesis method. When suspended in an aqueous medium, the particles collapsed by expelling water as the temperature was raised toward the volume phase transition temperature (VPTT) of ≈ 34°C. We noted that the sizes of the stiffer particles, synthesized with higher crosslinker concentration, collapsed less abruptly. Using Fourier transform infrared spectroscopy, we observed enhanced particle dehydration with increasing temperature and decreasing particle stiffness. Oscillatory rheology experiments on dense aqueous PNIPAM suspensions, prepared at a fixed particle effective volume fraction Φeff= 1.5 at 25°C, revealed that suspensions constituted by the stiffest particles are the most elastic over a broad temperature range. Above the VPTT, suspensions of particles of intermediate stiffnesses exhibited two-step yielding, a typical signature of fragile gel formation. Zeta potential measurements showed that PNIPAM particles of lower stiffnesses are rendered electrostatically unstable in aqueous suspension. Combining cryogenic scanning electron microscopy and rheology, we noted a glass–glass transition when the temperature of a dense suspension of stiff PNIPAM particles was raised across the VPTT. In contrast, suspensions of particles of the lowest stiffnesses showed a gel-viscoelastic liquid–gel transition during an identical temperature ramp experiment. Our study reveals that temperature-induced phase transformations in dense PNIPAM suspensions depend sensitively on the stiffness of the constituent particles and can be explained by considering amphiphilicity-driven morphological changes in the suspension microstructures
Uncovering the hidden structure: A study on the feasibility of induction thermography for fiber orientation analysis in CFRP composites using 2D-FFT
No full textThe induction heating process elicits a heating response in a carbon fiber reinforced polymer (CFRP) composite distinctly different from those in metal. Prior work in our lab and elsewhere has established that the intensity and spatial distribution of the heating patterns are governed by the fiber orientations in each layer and the degree of electrical contact between layers. Based on more extensive work on this non-conventional heating behavior, we show that the analysis of induction heating patterns enables the characterization of the fiber orientations and stacking order within the material. In the first step, 2D Fast Fourier Transform (2D-FFT) is used to extract the fiber orientations from the spatial characteristics of the heating pattern recorded with an infrared camera. In the next step, the extracted fiber orientation is used to design bandpass filters to carry out the inverse 2D-FFT and obtain the layer stacking order. Our findings demonstrate that this approach can accurately identify the layer orientations (maximum error of 6°) and the stacking sequence in quasi-isotropic CFRP laminates with up to 12 layers. We believe that the proposed approach has the potential as a valuable nondestructive, non-contact tool for large-area inspection and quality control in manufacturing fiber-reinforced composites
Towards Sustainable Delta Ecosystems: Pollution Mitigation for Achieving SDGs in Indian Delta Region
No full textBeing exceptionally fertile, the Indian delta regions contribute significantly to India’s food supply and serve as source of surface and ground water for millions of people. These areas contribute immensely to the livelihoods of the people through fishery, tourism, transportation, and commerce, improving country's economy and people’s wellbeing besides supporting rich biodiversity. Although these ecosystems have great ecological and economic importance, they face serious threat from climate change and human activities such as deforestation, industrial and domestic sewage discharge, increasing urbanization, and solid and agricultural waste disposal. Anthropogenic pollutants, primarily heavy metals such as As, Co, Cr, Pb, Ni, Mn, Fe, and Zn, macro- and micro-plastics, pesticides, polycyclic aromatic hydrocarbons, and polychlorinated biphenyl are being released in great quantity threatening the overall health of the ecosystems, and existence of several plant and animal species. Excessive groundwater removal has resulted in salt water intrusion into groundwater aquifers, rendering the groundwater unfit for human consumption and cultivation. The conservation of these areas needs effective pollution abatement measures taking a multidisciplinary approach. This article aims to examine anthropogenic pollutants’ ecological and economic impact on Indian delta regions and propose multidisciplinary mitigation measures that include sewage discharge, urbanization, garbage management, and pesticide usage. The goal is to integrate pollution-reduction efforts with SDGs, such as protecting livelihoods, boosting the economy, and maintaining biodiversity in the face of climate change and human activity
Towards Sustainable Delta Ecosystems: Pollution Mitigation for Achieving SDGs in Indian Delta Region
No full textBeing exceptionally fertile, the Indian delta regions contribute significantly to India’s food supply and serve as source of surface and ground water for millions of people. These areas contribute immensely to the livelihoods of the people through fishery, tourism, transportation, and commerce, improving country's economy and people’s wellbeing besides supporting rich biodiversity. Although these ecosystems have great ecological and economic importance, they face serious threat from climate change and human activities such as deforestation, industrial and domestic sewage discharge, increasing urbanization, and solid and agricultural waste disposal. Anthropogenic pollutants, primarily heavy metals such as As, Co, Cr, Pb, Ni, Mn, Fe, and Zn, macro- and micro-plastics, pesticides, polycyclic aromatic hydrocarbons, and polychlorinated biphenyl are being released in great quantity threatening the overall health of the ecosystems, and existence of several plant and animal species. Excessive groundwater removal has resulted in salt water intrusion into groundwater aquifers, rendering the groundwater unfit for human consumption and cultivation. The conservation of these areas needs effective pollution abatement measures taking a multidisciplinary approach. This article aims to examine anthropogenic pollutants’ ecological and economic impact on Indian delta regions and propose multidisciplinary mitigation measures that include sewage discharge, urbanization, garbage management, and pesticide usage. The goal is to integrate pollution-reduction efforts with SDGs, such as protecting livelihoods, boosting the economy, and maintaining biodiversity in the face of climate change and human activity