23 research outputs found
Constructing adversarial samples against deep learning-based sensing system (part II)
No full textAs deep learning become more popular and have grown to become crucial components in the daily device we use. Despite their effectiveness, they are not invincible. Adversarial examples initially discovered and applied to computer vision systems are now becoming a noticeable issue in DeepSpeech processing classifier as well.
Adversarial examples are input samples that have very bad classification accuracy. They are generated by adding imperceptible perturbations by human. These adversarial examples result in misclassification of results.
In late 2017, an attack was shown to be quite effective against the Speech Commands classification model. Speech commands are used very frequently in many applications, such as Google Assistant, Amazon Alexa and Apple's Siri. Thus, adversarial examples produced by this attack could have real-world consequences.
While previous work in defending again these malicious attacks has investigated using gradient masking to hide information of the model and audio pre-processing to reduce or distort adversarial noise, this project explores the idea of simple pink noise injection at different loudness to detect adversarial examples. This technique of noise injection does not require retraining or modifying the model. It is also possible to transfer this technique from the model used in this project to other DeepSpeech models.Bachelor of Engineering (Computer Science
Navigating Gendered Networks: The Career Development of Women Engineers in China's Energy Sector
No full textThe career development of women engineers is a significant theme in the context of gender equality in STEM fields. This study aims to explore the impact of workplace relationships on the career development and well-being of women engineers in the Chinese energy industry, a traditionally male-dominated field. The research is grounded in social network theory and employs a qualitative methodology, including in-depth interviews with women engineers across various energy companies in China. The findings reveal that informal social networks, often male-dominated, play a crucial role in shaping career opportunities and challenges for women. These networks frequently exclude women, limiting their access to vital information and career advancement opportunities. Moreover, women in these industries face additional pressures from both male colleagues and female leaders, who may perpetuate existing gender biases. The study suggests that organizations can benefit from fostering more inclusive workplace cultures and support networks, which would promote gender equality and improve the career development prospects for women in engineering roles
Navigating Gendered Networks: The Career Development of Women Engineers in China's Energy Sector
No full textThe career development of women engineers is a significant theme in the context of gender equality in STEM fields. This study aims to explore the impact of workplace relationships on the career development and well-being of women engineers in the Chinese energy industry, a traditionally male-dominated field. The research is grounded in social network theory and employs a qualitative methodology, including in-depth interviews with women engineers across various energy companies in China. The findings reveal that informal social networks, often male-dominated, play a crucial role in shaping career opportunities and challenges for women. These networks frequently exclude women, limiting their access to vital information and career advancement opportunities. Moreover, women in these industries face additional pressures from both male colleagues and female leaders, who may perpetuate existing gender biases. The study suggests that organizations can benefit from fostering more inclusive workplace cultures and support networks, which would promote gender equality and improve the career development prospects for women in engineering roles
Correction to:Evolutionary game theoretic approach with deep learning for health decision-making in critical environment (Annals of Operations Research, (2024), 10.1007/s10479-024-06353-2)
No full textTypesetter during proofing mistakenly changed co-author Helen Huifen Cai’s affiliation and should remain as “Business School, Middlesex University, The Burroughs, Hendon, London NW4 4BT, United Kingdom.” Original article has been corrected.</p
Comparative Transcriptome Analysis Provides Insight into the Effect of 6-BA on Flower Development and Flowering in <i>Bougainvillea</i>
No full textBougainvillea spp. is a well-known ornamental plant that is widely applied in urban landscaping construction. The colorful bracts of Bougainvillea in full bloom become important for urban landscape during special festivals. Although flowering regulation measures of Bougainvillea attracted much attention, the underlying mechanism of flower bud differentiation and development remains poorly understood. Here, we induced flowering of Bougainvillea glabra ‘Sao Paulo’ under 6-BA treatment and conducted RNA sequencing data analysis to characterize the molecular regulatory mechanism of flower development in response to 6-BA. Transcriptome analysis indicated that a series of genes and transcription factors of cytokinin metabolism, flowering and floral development regulation, and photoperiod regulation were upregulated by the 6-BA treatment, including COL, AP2, FT, SOC1, LFY, SPL4, SPL9, and SPL13. Moreover, the expression of these important genes exhibited relatively high levels of thorns compared to apical buds, suggesting that flower bud differentiation probably starts with the thorns in Bougainvillea. This study confirms that 6-BA treatment at certain concentrations can promote flowering of Bougainvillea and provides insight into the regulatory mechanism of the growth regulator acting on early flowering of Bougainvillea.</i
Adverse drug events associated with tiotropium: a real-world pharmacovigilance study of FDA adverse event reporting system database
No full textIntroductionTiotropium, a long-acting muscarinic antagonist, is commonly employed for the maintenance treatment of chronic obstructive pulmonary disease (COPD) and asthma. While its efficacy has been validated through numerous randomized controlled trials, safety concerns in real-world post-marketing settings necessitate further evaluation.AimThis study aimed to analyze the adverse events (AEs) associated with tiotropium reported in the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) database to identify potential safety signals.MethodsA retrospective analysis was conducted on adverse reaction reports related to tiotropium in the FAERS database from the first quarter of 2004 to the fourth quarter of 2024. The AE names in the FAERS database were systematically classified using the Preferred Terms (PTs) and System Organ Classes (SOCs) provided by the latest version of the Medical Dictionary for Regulatory Activities (MedDRA 27.1). After deduplication, a combination of methods, including the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-item Gamma Poisson Shrinker (MGPS), was employed for disproportionality analysis.ResultsA total of 129,763 AE reports related to tiotropium were included in the analysis, affecting 65,045 patients. These reports encompassed 27 different SOC categories, identifying 264 AEs associated with tiotropium. After excluding certain AEs deemed clinically insignificant, the most common AEs reported were dyspnea (n = 8,600), cough (n = 2,440), and pneumonia (n = 2080). The AEs exhibiting the highest signal strength included aggravated dyspnea (ROR: 162.04), hoarseness (ROR: 43.42), and aggravated chronic obstructive airway disease (ROR: 43.17). Additionally, we identified potential risks not mentioned in the instructions (United States Prescribing Information and the Canadian Product Monograph), such as epiglottic cancer, halo vision, and malignant lung tumors.ConclusionThis study offers a more comprehensive understanding of tiotropium by uncovering previously unreported adverse reactions. Physicians should take these newly identified adverse reactions into account when prescribing this medication
Normal-state gap in the parent cuprate <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mi mathvariant="bold">Pr</mml:mi><mml:mn>2</mml:mn></mml:msub><mml:msub><mml:mi mathvariant="bold">CuO</mml:mi><mml:mrow><mml:mn>4</mml:mn><mml:mo>±</mml:mo><mml:mi>δ</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math>
No full textPharmacovigilance analysis of iodinated contrast media related respiratory adverse effects based on the FDA adverse event reporting system
No full textAimIodinated contrast media (ICM) serve as a cornerstone in diagnostic imaging; however, the risk profile of their respiratory system adverse reactions remains inadequately characterized. This study aims to quantify and compare the respiratory system adverse reaction signals associated with four widely utilized non-ionic ICMs—iohexol, iopamidol, iopromide, and ioversol—using the United States Food and Drug Administration’s Adverse Event Reporting System (FAERS) database.MethodsWe extracted data from the FAERS database spanning the first quarter of 2004 to the fourth quarter of 2024. Signal detection was performed employing the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-Item Gamma Poisson Shrinker (MGPS) methodologies to identify statistically significant drug-event associations.ResultsA total of 9,682 adverse event (AE) reports related to the respiratory system were analyzed, encompassing 53 distinct respiratory-related events. All four ICMs demonstrated a significant correlation with respiratory diseases. Notably, the analysis revealed that certain adverse reactions were not covered in the prescribing information. Moreover, sneezing and laryngeal edema exhibited unusually high signal intensities across all four ICMs.ConclusionThis study reveals heterogeneity in the respiratory system risk profiles among the four commonly used non-ionic ICMs. These findings suggest that clinical decision-making should incorporate the distinct risk profiles of specific ICMs, and individualized monitoring strategies should be implemented for high-risk patients
BAP regulates lateral bud outgrowth to promote tillering in Paphiopedilum callosum (Orchidaceae)
No full textAbstract Background Paphiopedilum orchids have a high ornamental value, and flower abundance is a key horticultural trait. Most Paphiopedilum plants exhibit weak tillering ability, with their tiller buds often entering a dormant state post-formation. Tiller production plays a crucial role in enhancing flower abundance and is potentially regulated by plant hormones. However, the effect of hormones on tillering in Paphiopedilum plants is still unclear. Results In this study, we investigated the promotion of tillering in P. callosum through exogenous root irrigation of benzylaminopurine (BAP). We observed a dose-dependent promotion of tiller production by BAP, with the strongest effect observed at a concentration of 400 mg/L. By comparing the expression of key genes in P. ‘SCBG Yingchun’ (with strong tiller ability) and P callosum (with weak tiller ability), we found that BAP promotes tillering by interacting with abscisic acid (ABA). This interaction involves down-regulation of the ABA degradation gene PcCYP707A, leading to a reduction in ABA content, and the subsequent down-regulation of dormancy-associated genes (PcDRMH1, PcSVP) to release bud dormancy. Additionally, BAP promotes sustained outgrowth of tiller buds by increasing the level of indole-3-acetic acid (IAA) through up-regulation of the IAA synthesis gene PcYUC2 and the transport gene PcPINIC. Conclusions Our results indicated that the application of BAP promotes lateral bud outgrowth and increases tiller production in P. callosum. Through transcriptome analysis, we found that the BAP-promotion of tillering involves not only changes in endogenous IAA, ABA, and CTKs content but is also associated with the regulation of metabolism-related genes and dormancy-associated genes. This study presents the first comprehensive report of BAP-promoted tillering in P. callosum, providing a foundational basis for further mechanistic studies on tiller development in Paphiopedilum species and other non-model plants
