224 research outputs found
Correction: Guidelines for the prevention, detection and management of the renal complications of COVID-19 in Africa
No full textThe authors of the article ‘Guidelines for the prevention, detection and management of the renal complications of COVID-19 in Africa’ [1] wish to acknowledge the contribution of Professor Hussein El Fishawy. Our guidelines drew on various sources, including the Egyptian Ministry of Health guidelines, portions of which were adapted and reproduced with permission from the Egyptian Ministry of Health. Two of the authors of those guidelines, Professors Elsayed and Zaki, are also coauthors of our paper. Professor El Fishawy was the third author of the Egyptian guidelines and we would like to acknowledge his contribution to our review through this source, especially with respect to the treatment algorithms for patients with kidney transplants and those with acute kidney injury. Reference1. Elsayed HM, Wadee S, Zaki MS, Were AJO, Ashuntantang GE, Bamgboye EL, et al. Guidelines for the prevention, detection and management of the renal complications of COVID-19 in Africa. Afr J Nephrol. 2020; 23(1):109-126
Représentations littéraires du sacré dans le roman maghrébin de langue française
No full textThis interdisciplinary study explores how Driss Chraïbi’s L’Homme du Livre (1995), Assia Djebar’s Loin de Médine (1991), and Anissa Boumediène’s La fin d’un monde (1991) present accounts of particular historical moments in early Islam. This study explores the role of the imagination as well as freedom of invention when reconstructing historical events. It engages the novels through a study of the interplay between the literary text and the sources and traditions that impact and shape the text narrative. Gaining direct access to the original sources in Arabic serves to analyze how religious and early historical materials are considered in and reflected by the fictional texts. Because the sources tend to differ in both content and approach, this study examines their preoccupations in order to determine the criteria of selection applied by each novelist.Ph.D.Includes bibliographical referencesIncludes vitaby Hanan Elsaye
Intrathecal Drug Delivery Systems Survey: Trends in Utilization in Pain Practice [Corrigendum]
No full textAbd-Sayed A, Fiala K, Weisbein J, et al. J Pain Res. 2022;15:1305–1314.
The authors have advised there is an error in the author list on page 1305. The author name “Alaa Abd-Sayed” should read “Alaa Abd-Elsayed”.
The authors apologize for this error
Investigating the Role of the Nucleosome Remodeling Factor INO80 in Development and NURF in Anti-Tumor Immunity
No full textUnderstanding how an epigenetic regulator such as ATP-dependent chromatin-remodeling complexes modulate processes such as development and/or immune response is essential for our comprehension of cell biology. Deletion of the ATP-dependent chromatin remodeling factor INO80 is known to be embryonically lethal, however, the mechanism is not known. To identify roles for INO80 in mouse early development we generated Ino80 KO mice. Ino80 KO ESCs (Embryonic stem cells) were viable when maintained at ground state pluripotency but fail to differentiate in vitro and in vivo. Gene expression analysis of Ino80 KO early embryos by in situ hybridization showed elevated Bmp4 expression and reduced expression of DVE (distal visceral endoderm) markers Cer1, Hex, and Lefty1. BMP4 is a known negative regulator of DVE differentiation in the early embryo. Molecular studies in Ino80 KO ESCs demonstrated that INO80 is bound to the Bmp4 promoter, and regulates its chromatin structure, to suppress the positive regulator SP1 from stimulating its transcription. These results, suggest that INO80 directly regulates the chromatin structure of the Bmp4 promoter with consequences to mouse embryo development. These results are significant because they demonstrate a specific role of INO80 in establishing P-D embryonic axis.
NURF (Nucleosome remodeling factor) is another ATP-dependent chromatin remodeling complex that is overexpressed in many cancer types including breast cancer. To demonstrate the roles of NURF in breast cancer biology, we knocked-down the NURF essential subunit BPTF (bromodomain PHD finger transcription factor) in
mouse breast cancer cell lines. Transplantation of these cell lines into immune-competent mice revealed that BPTF KD enhances NK cell antitumor activity. BPTF KD enhanced NK-92 cytotoxic activity toward BPTF KD cells by NKp30 activation in vitro. NK-92 activity is reduced by the addition of heparin to the culture medium, further indicating the involvement of NKp30 (in human) and NCR1 (in mice) in killing of tumor cells. We found that BPTF controls the abundance of NKp30/NCR1 ligands (heparin sulfate proteoglycans (HSPGs) by regulation of heparanase expression (endoglycosidase that degrades HSPGs). In addition, BPTF depletion in established mouse breast tumors enhanced anti-tumor immunity, without affecting NK or T cell cytotoxic activity, providing a novel immunotherapy target
Semi-supervised learning and bidirectional decoding for effective grammar correction in low-resource scenarios
No full textThe correction of grammatical errors in natural language processing is a crucial task as it aims to enhance the accuracy and intelligibility of written language. However, developing a grammatical error correction (GEC) framework for low-resource languages presents significant challenges due to the lack of available training data. This article proposes a novel GEC framework for low-resource languages, using Arabic as a case study. To generate more training data, we propose a semi-supervised confusion method called the equal distribution of synthetic errors (EDSE), which generates a wide range of parallel training data. Additionally, this article addresses two limitations of the classical seq2seq GEC model, which are unbalanced outputs due to the unidirectional decoder and exposure bias during inference. To overcome these limitations, we apply a knowledge distillation technique from neural machine translation. This method utilizes two decoders, a forward decoder right-to-left and a backward decoder left-to-right, and measures their agreement using Kullback-Leibler divergence as a regularization term. The experimental results on two benchmarks demonstrate that our proposed framework outperforms the Transformer baseline and two widely used bidirectional decoding techniques, namely asynchronous and synchronous bidirectional decoding. Furthermore, the proposed framework reported the highest F1 score, and generating synthetic data using the equal distribution technique for syntactic errors resulted in a significant improvement in performance. These findings demonstrate the effectiveness of the proposed framework for improving grammatical error correction for low-resource languages, particularly for the Arabic language
Is fertility-preservation safe for adult non-metastatic gynecologic rhabdomyosarcoma patients? Systematic review and pooled survival analysis of 137 patients
No full textOptimizing the impact of data augmentation for low-resource grammatical error correction
No full textGrammatical Error Correction (GEC) refers to the automatic identification and amendment of grammatical, spelling, punctuation, and word-positioning errors in monolingual texts. Neural Machine Translation (NMT) is nowadays one of the most valuable techniques used for GEC but it may suffer from scarcity of training data and domain shift, depending on the addressed language. However, current techniques (e.g., tuning pre-trained language models or developing spell-confusion methods without focusing on language diversity) tackling the data sparsity problem associated with NMT create mismatched data distributions. This paper proposes new aggressive transformation approaches to augment data during training that extend the distribution of authentic data. In particular, it uses augmented data as auxiliary tasks to provide new contexts when the target prefix is not helpful for the next word prediction. This enhances the encoder and steadily increases its contribution by forcing the GEC model to pay more attention to the text representations of the encoder during decoding. The impact of these approaches was investigated using the Transformer-based for low-resource GEC task, and Arabic GEC was used as a case study. GEC models trained with our data tend more to source information, are more domain shift robustness, and have less hallucinations with tiny training datasets and domain shift. Experimental results showed that the proposed approaches outperformed the baseline, the most common data augmentation methods, and classical synthetic data approaches. In addition, a combination of the three best approaches Misspelling, Swap, and Reverse achieved the best F1 score in two benchmarks and outperformed previous Arabic GEC approaches
Degradation modeling of ink fading and diffusion of printed images
No full textColor printing plays an important role in the modern society. It is known that the color of printed images degrades gradually due to the fading and diffusion of the inks. Color degradation leads to a distortion or loss of the original information in printed images. Therefore, it is desirable to understand how the color of printed images changes over time. In this dissertation, we present degradation models to predict the characteristics of the ink fading and diffusion of printed images.
We begin by modeling the ink degradation from a physics-based perspective. Color images are printed by projecting small ink dots on medium, usually paper. This technique is called halftone printing. Halftone printing of color images results in a variety of ink mixtures and subsequently their potential catalytic fading. For the most commonly used Cyan-Magenta-Yellow-Black (CMYK) ink set, sixteen possible ink mixtures are generated during printing. A state transition diagram is then proposed for the ink fading in this multi-ink printing scenario. The ink area coverage is used as the performance indicator. Assuming constant fading and diffusion rates, we develop an ink fading model based on the differential equations according to the state transition diagram and an autoregressive ink diffusion model by discretizing the two-dimensional diffusion equation. The two models are then integrated into a single degradation model.
Further examination of the developed degradation models reveals that the fading or diffusion rate is equivalent to the hazard rate in reliability engineering. It is known that the hazard rate of the exponential failure time distribution is constant. Hence, the developed degradation model with constant fading and diffusion rates is equivalent to the multistate Markov process model with exponential transition time distribution. By using non-exponential transition time distributions, the fading and diffusion rates become time-varying and a more general semi-Markov process degradation model is developed accordingly.
Moreover, stochastic process models are investigated to provide stochastic area coverage prediction for the ink degradation. We first model the ink fading using the Hull-White/Vasicek (HWV) stochastic process. The HWV ink fading model considers that the variance of the ink area coverage shrinks as it approaches zero. Besides, spatial convolution is used to model ink diffusion. The two models are integrated into a spatio-temporal stochastic degradation model for the ink fading and diffusion of printed images. The cases of recurrent and non-recurrent time-varying fading and diffusion rates are investigated.
Inks on the paper degrade, so does the paper. The degradation of paper condition may in turn affect the degradation of the inks. Therefore, the investigation of the degradation modeling of ink fading and ink diffusion with ink-paper interactions is needed. Two aspects of the ink-paper interactions are considered, i.e., the effect of paper aging such as depolymerization and yellowing, and the fiber orientation of the paper.
The degradation process of printed images usually takes a very long time. An accelerated degradation model and the optimal design of accelerated degradation test planning is developed for accurate degradation prediction of printed images. The effects of three constant environmental stresses: temperature, humidity, and illumination (intensity), are investigated, and experimental data are used to validate the proposed model. The results show strong agreements between the proposed ink fading and ink diffusion prediction model and the actual experimental data.Ph.D.Includes bibliographical reference
Reliability estimation of balanced systems with multi-dimensional distributed units
No full textBalanced systems with multi-dimensional distributed units are emerging in a diverse range of industries. This includes Unmanned Aerial Vehicles (UAV) with multi-level of rotary wings, Spherical Unmanned Vehicles (SUV), Spherical Phased Array Antenna (SPAA), etc. In this dissertation, we present the reliability estimation for such systems. In particular, we consider two configurations: 1) balanced systems with units distributed circularly on multi-level and 2) balanced systems with units distributed spherically.
First, balanced systems with units distributed circularly on multi-level are generalized as (k₁, k₂)-out-of-(n, m) pairs: G balanced systems. We consider two scenarios: 1) all units perform the same function and 2) adjacent pairs perform complementary functions. For both scenarios, unbalanced system is considered as failed. When units fail and cause the system imbalance, we explore two approaches to rebalance the system: 1) forcing down units on other locations and 2) resuming units that are previously forced down (if any). When units in a system perform the same function, operational states are defined as balanced states with at least k₁ operating pairs and each operating pair has at least k₂ units on each side. The system reliability is obtained by enumerating all of the operational states and summing the probabilities of those states. For (k₁, k₂)-out-of-(n, m) pairs: G balanced systems with adjacent pairs performing complementary functions, in addition to maintaining system balance, the adjacent operating pairs are required to perform complementary functions. Thus, if a pair fails, one of the adjacent pairs is forced down. Similarly, the system reliability is obtained by enumerating all of the operational states. It becomes computational expensive when the number of units in each pair and/or the number of pairs are large. In that case, efficient algorithms are developed to obtain the reliability for such systems.
The balanced system with units distributed spherically is generalized as a spherical k-n-i: G balanced system. We consider two balancing requirements: 1) rotational balance is maintained so that the system is not rotating w.r.t. roll, yaw and pitch axes and 2) symmetrical balance is essential in improving the systems’ stability. We present mathematical approaches to determine the balance status of a system. Similarly, the unbalanced system is rebalanced by 1) forcing down units on other locations and 2) resuming previously forced-down units. The system reliability is obtained by the enumeration of operational states and calculation of operational states’ probabilities. We develop an efficient algorithm for reliability estimation when the number of units in the system is large.
Degradation models are developed for the (k₁, k₂)-out-of-(n, m) pairs: G balanced systems to further investigate the system reliability when degradation data are available. The degradation processes of units in the system are either stationary (inverse Gaussian process) or non-stationary (improved inverse Gaussian process). We propose a degradation balance mechanism in which the ‘most’ degraded units are forced down temporarily during the degradation process so that the system is less possible to fail due to imbalance. A closed-form lower bound reliability is presented when the balance mechanism is not applied. When it is applied, reliability is obtained by Monte Carlo simulation.
From the reliability study of the both configurations, it is observed that the reliability of a balanced system with multi-dimensional distributed units depends not only on the system’s total number of units and the least number of operating units, but also on the system configurations and balance requirements. Systems with more units do not necessarily provide a higher reliability since they are more likely to fail due to imbalance. Thus, optimal system design is key to maximize the system reliability which is investigated through numerical examples in this dissertation.Ph.D.Includes bibliographical reference
- …
