Search

Herbal medicines are used by patients with IBD despite limited evidence. We present a systematic review and meta-analysis of randomized controlled trials (RCTs) investigating treatment with herbal medicines in active ulcerative colitis (UC). A search query designed by a library informationist was used to identify potential articles for inclusion. Articles were screened and data were extracted by at least two investigators. Outcomes of interest included clinical response, clinical remission, endoscopic response, endoscopic remission, and safety. We identified 28 RCTs for 18 herbs. In pooled analyses, when compared with placebo, clinical response rates were significantly higher for Indigo naturalis (IN) (RR 3.70, 95% CI 1.97-6.95), but not for Curcuma longa (CL) (RR 1.60, 95% CI 0.99-2.58) or Andrographis paniculata (AP) (RR 0.95, 95% CI 0.71-1.26). There was a significantly higher rate of clinical remission for CL (RR 2.58, 95% CI 1.18-5.63), but not for AP (RR 1.31, 95% CI 0.86-2.01). Higher rates of endoscopic response (RR 1.56, 95% CI 1.08-2.26) and remission (RR 19.37, 95% CI 2.71-138.42) were significant for CL. CL has evidence supporting its use as an adjuvant therapy in active UC. Research with larger scale and well-designed RCTs, manufacturing regulations, and education are needed.

Background: The extended acquisition times required for MRI limit its availability in resource-constrained settings. Consequently, accelerating MRI by undersampling k-space data, which is necessary to reconstruct an image, has been a long-standing but important challenge. We aimed to develop a deep convolutional neural network (dCNN) optimisation method for MRI reconstruction and to reduce scan times and evaluate its effect on image quality and accuracy of oncological imaging biomarkers. Methods: In this multicentre, retrospective, cohort study, MRI data from patients with glioblastoma treated at Heidelberg University Hospital (775 patients and 775 examinations) and from the phase 2 CORE trial (260 patients, 1083 examinations, and 58 institutions) and the phase 3 CENTRIC trial (505 patients, 3147 examinations, and 139 institutions) were used to develop, train, and test dCNN for reconstructing MRI from highly undersampled single-coil k-space data with various acceleration rates (R=2, 4, 6, 8, 10, and 15). Independent testing was performed with MRIs from the phase 2/3 EORTC-26101 trial (528 patients with glioblastoma, 1974 examinations, and 32 institutions). The similarity between undersampled dCNN-reconstructed and original MRIs was quantified with various image quality metrics, including structural similarity index measure (SSIM) and the accuracy of undersampled dCNN-reconstructed MRI on downstream radiological assessment of imaging biomarkers in oncology (automated artificial intelligence-based quantification of tumour burden and treatment response) was performed in the EORTC-26101 test dataset. The public NYU Langone Health fastMRI brain test dataset (558 patients and 558 examinations) was used to validate the generalisability and robustness of the dCNN for reconstructing MRIs from available multi-coil (parallel imaging) k-space data. Findings: In the EORTC-26101 test dataset, the median SSIM of undersampled dCNN-reconstructed MRI ranged from 0·88 to 0·99 across d

BACKGROUND: The extended acquisition times required for MRI limit its availability in resource-constrained settings. Consequently, accelerating MRI by undersampling k-space data, which is necessary to reconstruct an image, has been a long-standing but important challenge. We aimed to develop a deep convolutional neural network (dCNN) optimisation method for MRI reconstruction and to reduce scan times and evaluate its effect on image quality and accuracy of oncological imaging biomarkers. METHODS: In this multicentre, retrospective, cohort study, MRI data from patients with glioblastoma treated at Heidelberg University Hospital (775 patients and 775 examinations) and from the phase 2 CORE trial (260 patients, 1083 examinations, and 58 institutions) and the phase 3 CENTRIC trial (505 patients, 3147 examinations, and 139 institutions) were used to develop, train, and test dCNN for reconstructing MRI from highly undersampled single-coil k-space data with various acceleration rates (R=2, 4, 6, 8, 10, and 15). Independent testing was performed with MRIs from the phase 2/3 EORTC-26101 trial (528 patients with glioblastoma, 1974 examinations, and 32 institutions). The similarity between undersampled dCNN-reconstructed and original MRIs was quantified with various image quality metrics, including structural similarity index measure (SSIM) and the accuracy of undersampled dCNN-reconstructed MRI on downstream radiological assessment of imaging biomarkers in oncology (automated artificial intelligence-based quantification of tumour burden and treatment response) was performed in the EORTC-26101 test dataset. The public NYU Langone Health fastMRI brain test dataset (558 patients and 558 examinations) was used to validate the generalisability and robustness of the dCNN for reconstructing MRIs from available multi-coil (parallel imaging) k-space data. FINDINGS: In the EORTC-26101 test dataset, the median SSIM of undersampled dCNN-reconstructed MRI ranged from 0·88 to 0·99 acros

Extracting hyper-relations is crucial for constructing comprehensive knowledge graphs, but there are limited supervised methods available for this task. To address this gap, we introduce a zero-shot prompt-based method using OpenAI's GPT-3.5 model for extracting hyper-relational knowledge from text. Comparing our model with a baseline, we achieved promising results, with a recall of 0.77. Although our precision is currently lower, a detailed analysis of the model outputs has uncovered potential pathways for future research in this area., Comment: 5 pages + references

In the dynamic landscape of open source software (OSS) development, understanding and addressing incivility within issue discussions is crucial for fostering healthy and productive collaborations. This paper presents a curated dataset of 404 locked GitHub issue discussion threads and 5961 individual comments, collected from 213 OSS projects. We annotated the comments with various categories of incivility using Tone Bearing Discussion Features (TBDFs), and, for each issue thread, we annotated the triggers, targets, and consequences of incivility. We observed that Bitter frustration, Impatience, and Mocking are the most prevalent TBDFs exhibited in our dataset. The most common triggers, targets, and consequences of incivility include Failed use of tool/code or error messages, People, and Discontinued further discussion, respectively. This dataset can serve as a valuable resource for analyzing incivility in OSS and improving automated tools to detect and mitigate such behavior.

Globally, there is a rising prevalence of infertility, and it has a negative consequence on the quality of life of women. The researchers aimed to understand the knowledge, attitude, and practice of women with infertility toward exercises. A cross-sectional study was conducted on 332 women aged between 18 and 45 years attending an infertility clinic in a tertiary hospital in Southern India. Participants filled a self-developed, content validated, pilot-tested questionnaire. Informants perceived stress and weight gain to be the major causes of infertility. Exercise was believed to improve fertilization by 55.4% of the participants, and walking and yoga were the preferred mode of exercises.

INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSIONS & CONCLUSIONS DISCLAIMER AKNOWLEDGEMENTS Exertional heat illness (EHI) is a serious threat to the operational readiness of Service Members (SM). EHI is a spectrum of conditions ranging from mild heat exhaustion to life threatening heat stroke (EHS). The occurrence of EHI continues to increase in number, many with unpredictable outcomes including recurrence, which may result in significant lost duty time and potential long-term adverse effects. Identification of risk factors and causes of EHI to differentiate between those at risk of recurrence and those who can promptly return to duty is a cornerstone issue in military medicine. Despite knowledge of various risk factors that might be associated with increased risk to EHI, inherent susceptibility to EHI remains poorly understood. The goal of this study was to identify genetic factors associated with EHI in SMs. This study was approved by the Institutional Review Board of Uniformed Services University of the Health Sciences. We retrospectively enrolled 38 SMs diagnosed with EHI as cases in this genetic study. Demographic data, family history and supplement use were obtained through self-reported questionnaire. Clinical data and clinical test results were obtained from electronic health records. Whole exome sequencing (WES) was performed on DNA extracted from whole blood. We conducted rare variant analysis with a frequency threshold of 0.1% from gnomAD database along with various in-silico prediction programs. Pathway analysis was conducted on genes with rare variants using Metascape (https://metascape.org). Genetic variants were interpreted according to the American College of Medical Genetics and Genomics (ACMG) guidelines. The majority of study cases were male patients with a history of one EHI episode, and mild increases in both core body temperature and creatine kinase (Table 1). About one third of the cases had symptoms of EHS and/or recurrent episodes of EHI, RITM0039907, Exertional heat illness (EHI) is a serious threat to the operational readiness of Service Members (SM). EHI is a spectrum of conditions ranging from mild heat exhaustion to life threatening heat stroke (EHS). The occurrence of EHI continues to increase in number, many with unpredictable outcomes including recurrence, which may result in significant lost duty time and potential long-term adverse effects. Identification of risk factors and causes of EHI to differentiate between those at risk of recurrence and those who can promptly return to duty is a cornerstone issue in military medicine. Despite knowledge of various risk factors that might be associated with increased risk to EHI, inherent susceptibility to EHI remains poorly understood. The goal of this study was to identify genetic factors associated with EHI in SMs.

Background: The majority of residents in long term care (LTC) facilities are older and frail, with multiple comorbidities and reduced cognitive capacity. Although the evidence suggests that advance care planning (ACP) improves the quality of end-of-life (EOL) care and promotes a good death for residents of LTC, such planning rarely occurs in these settings. Moreover, while nurses are in the ideal position to facilitate ACP, there is a paucity of empirical research examining their engagement in ACP. Purpose: The purpose of this qualitative study was to develop an inductively derived empirical model aimed at understanding the experiences of nurses working in LTC facilities, specifically with regard to their engagement in the ACP process. Design: A constructivist grounded theory (CGT) methodology was used to conduct this study. Symbolic interactionism (SI) and the socio-ecological model (SEM) served as sensitizing theoretical perspectives for this study. Purposive and theoretical sampling were used to recruit 25 registered nurses (RNs) from 18 proprietary and non-proprietary LTC facilities in Winnipeg, Manitoba who had worked a minimum of three months in LTC, were able to read/speak English, and were willing to provide consent to participate in the study. Methods: Data were collected using a demographic questionnaire; in-depth, semi-structured, audio-recorded, face-to-face/telephone interviews; field notes; and memos. Demographic data were analyzed with descriptive statistics. Verbatim transcriptions of the interviews were analyzed with specific CGT coding procedures. Findings: The basic social problem that emerged from the data was that of nurses trying to craft and implement an ACP level that they believed would optimize residents’ comfort in LTC. The empirically derived theoretical model that captured the experiences, processes, and strategies of nurses trying to address the identified social problem was orchestrating comfort: getting everyone on the same page. This

Background: The majority of residents in long term care (LTC) facilities are older and frail, with multiple comorbidities and reduced cognitive capacity. Although the evidence suggests that advance care planning (ACP) improves the quality of end-of-life (EOL) care and promotes a good death for residents of LTC, such planning rarely occurs in these settings. Moreover, while nurses are in the ideal position to facilitate ACP, there is a paucity of empirical research examining their engagement in ACP. Purpose: The purpose of this qualitative study was to develop an inductively derived empirical model aimed at understanding the experiences of nurses working in LTC facilities, specifically with regard to their engagement in the ACP process. Design: A constructivist grounded theory (CGT) methodology was used to conduct this study. Symbolic interactionism (SI) and the socio-ecological model (SEM) served as sensitizing theoretical perspectives for this study. Purposive and theoretical sampling were used to recruit 25 registered nurses (RNs) from 18 proprietary and non-proprietary LTC facilities in Winnipeg, Manitoba who had worked a minimum of three months in LTC, were able to read/speak English, and were willing to provide consent to participate in the study. Methods: Data were collected using a demographic questionnaire; in-depth, semi-structured, audio-recorded, face-to-face/telephone interviews; field notes; and memos. Demographic data were analyzed with descriptive statistics. Verbatim transcriptions of the interviews were analyzed with specific CGT coding procedures. Findings: The basic social problem that emerged from the data was that of nurses trying to craft and implement an ACP level that they believed would optimize residents’ comfort in LTC. The empirically derived theoretical model that captured the experiences, processes, and strategies of nurses trying to address the identified social problem was orchestrating comfort: getting everyone on the same page. This

This study focuses on designing, developing, and testing 3D-printed radar lenses with varying heights of Anti-Reflective (AR) structures (0.75mm, 1mm and 1.25mm). Radar lenses are essential to improve signal transmission and reception for radar systems. This study evaluates how effectively these lenses with AR structures perform in a 60 GHz radar system implemented on a printed circuit board to reduce losses and enhance accuracy, range, and resolution. The study also attempts to validate the reliability of the height-wavelength relationship theory. The lenses with various AR structures, heights and smooth surfaces were designed in SOLIDWORKS. The designed lenses were then imported to COMSOL software, and a simulation was performed. Then, the lens design was converted into a G-code with the printing instructions and printed in a 3D printer. The lenses were tested and measured alongside a radar source and reflector in the experimental setup. Tests and measurements in the Radar Fusion GUI software assessed the signal intensities of lenses with different heights, with caution to keep the experimental setup and the environment unaltered to get accurate results. The results of the COMSOL simulation and the experimental tests had similar outcomes, with the lens with an AR structure height of 1mm having the highest signal intensity. The findings also support the height and wavelength relationship theory, where the height of the AR structure and the wavelength depend on each other and emphasize the significance of lens design and height optimization to improve radar system performance and the efficiency of AR structures in boosting signal intensity.

This study focuses on designing, developing, and testing 3D-printed radar lenses with varying heights of Anti-Reflective (AR) structures (0.75mm, 1mm and 1.25mm). Radar lenses are essential to improve signal transmission and reception for radar systems. This study evaluates how effectively these lenses with AR structures perform in a 60 GHz radar system implemented on a printed circuit board to reduce losses and enhance accuracy, range, and resolution. The study also attempts to validate the reliability of the height-wavelength relationship theory. The lenses with various AR structures, heights and smooth surfaces were designed in SOLIDWORKS. The designed lenses were then imported to COMSOL software, and a simulation was performed. Then, the lens design was converted into a G-code with the printing instructions and printed in a 3D printer. The lenses were tested and measured alongside a radar source and reflector in the experimental setup. Tests and measurements in the Radar Fusion GUI software assessed the signal intensities of lenses with different heights, with caution to keep the experimental setup and the environment unaltered to get accurate results. The results of the COMSOL simulation and the experimental tests had similar outcomes, with the lens with an AR structure height of 1mm having the highest signal intensity. The findings also support the height and wavelength relationship theory, where the height of the AR structure and the wavelength depend on each other and emphasize the significance of lens design and height optimization to improve radar system performance and the efficiency of AR structures in boosting signal intensity.

Homelessness is a global issue exacerbated by poverty, forced migration, rapid urbanization, and lack of affordable housing. In India, around 1.8 million people were homeless in the 2011 Census, with many forced to sleep on the streets or in unofficial settlements. The link between homelessness, poverty, and mental illness is complex and uncertain. The increased risk of mental illness from poverty creates a greater risk of homelessness, with approximately 25% of homeless people experiencing serious mental health issues in India. The Poverty, Mental Illness & Homelessness (PMIH) nexus is particularly relevant in India due to its large homeless population, high resource inequalities, and high prevalence of mental illness. This thesis aims to understand the factors influencing homelessness in the population attending the clinic and find ways to leverage them in improving services. It focusses on the experiential knowledge of diverse stakeholders, including primarily women with severe mental illness, and identify pathways otherwise left unseen. By addressing the research gap and focusing on the experiences of women with severe mental illness, the Banyan better solutions may be found to address the complexities of poverty, mental illness, and homelessness in India. The interactive research approach and mixed methods are used to address the persistent problem of PMIH in outpatient clinics. The study focuses on understanding factors of homelessness among people with mental illness and the service needs and usage of those accessing outpatient facilities. It finds that 32.65% of women with mental illness accessing outpatient clinics reported experiencing homelessness, with low educational attainment and relational disruptions being associated with higher odds. Eliminating gender-based differences in education could be a crucial policy and practice direction to reduce homelessness risks among women with mental illness. The study examines the lived experiences of women wit

Background: The majority of residents in long term care (LTC) facilities are older and frail, with multiple comorbidities and reduced cognitive capacity. Although the evidence suggests that advance care planning (ACP) improves the quality of end-of-life (EOL) care and promotes a good death for residents of LTC, such planning rarely occurs in these settings. Moreover, while nurses are in the ideal position to facilitate ACP, there is a paucity of empirical research examining their engagement in ACP. Purpose: The purpose of this qualitative study was to develop an inductively derived empirical model aimed at understanding the experiences of nurses working in LTC facilities, specifically with regard to their engagement in the ACP process. Design: A constructivist grounded theory (CGT) methodology was used to conduct this study. Symbolic interactionism (SI) and the socio-ecological model (SEM) served as sensitizing theoretical perspectives for this study. Purposive and theoretical sampling were used to recruit 25 registered nurses (RNs) from 18 proprietary and non-proprietary LTC facilities in Winnipeg, Manitoba who had worked a minimum of three months in LTC, were able to read/speak English, and were willing to provide consent to participate in the study. Methods: Data were collected using a demographic questionnaire; in-depth, semi-structured, audio-recorded, face-to-face/telephone interviews; field notes; and memos. Demographic data were analyzed with descriptive statistics. Verbatim transcriptions of the interviews were analyzed with specific CGT coding procedures. Findings: The basic social problem that emerged from the data was that of nurses trying to craft and implement an ACP level that they believed would optimize residents’ comfort in LTC. The empirically derived theoretical model that captured the experiences, processes, and strategies of nurses trying to address the identified social problem was orchestrating comfort: getting everyone on the same page. This

Background: The majority of residents in long term care (LTC) facilities are older and frail, with multiple comorbidities and reduced cognitive capacity. Although the evidence suggests that advance care planning (ACP) improves the quality of end-of-life (EOL) care and promotes a good death for residents of LTC, such planning rarely occurs in these settings. Moreover, while nurses are in the ideal position to facilitate ACP, there is a paucity of empirical research examining their engagement in ACP. Purpose: The purpose of this qualitative study was to develop an inductively derived empirical model aimed at understanding the experiences of nurses working in LTC facilities, specifically with regard to their engagement in the ACP process. Design: A constructivist grounded theory (CGT) methodology was used to conduct this study. Symbolic interactionism (SI) and the socio-ecological model (SEM) served as sensitizing theoretical perspectives for this study. Purposive and theoretical sampling were used to recruit 25 registered nurses (RNs) from 18 proprietary and non-proprietary LTC facilities in Winnipeg, Manitoba who had worked a minimum of three months in LTC, were able to read/speak English, and were willing to provide consent to participate in the study. Methods: Data were collected using a demographic questionnaire; in-depth, semi-structured, audio-recorded, face-to-face/telephone interviews; field notes; and memos. Demographic data were analyzed with descriptive statistics. Verbatim transcriptions of the interviews were analyzed with specific CGT coding procedures. Findings: The basic social problem that emerged from the data was that of nurses trying to craft and implement an ACP level that they believed would optimize residents’ comfort in LTC. The empirically derived theoretical model that captured the experiences, processes, and strategies of nurses trying to address the identified social problem was orchestrating comfort: getting everyone on the same page. This

Fostering a collaborative and inclusive environment is crucial for the sustained progress of open source development. However, the prevalence of negative discourse, often manifested as toxic comments, poses significant challenges to developer well-being and productivity. To identify such negativity in project communications, especially within large projects, automated toxicity detection models are necessary. To train these models effectively, we need large software engineering-specific toxicity datasets. However, such datasets are limited in availability and often exhibit imbalance (e.g., only 6 in 1000 GitHub issues are toxic), posing challenges for training effective toxicity detection models. To address this problem, we explore a zero-shot LLM (ChatGPT) that is pre-trained on massive datasets but without being fine-tuned specifically for the task of detecting toxicity in software-related text. Our preliminary evaluation indicates that ChatGPT shows promise in detecting toxicity in GitHub, and warrants further investigation. We experimented with various prompts, including those designed for justifying model outputs, thereby enhancing model interpretability and paving the way for potential integration of ChatGPT-enabled toxicity detection into developer communication channels.

Use of figurative language, such as metaphors and idioms, is common in our daily-life communications, and it can also be found in Software Engineering (SE) channels, such as comments on GitHub. Automatically interpreting figurative language is a challenging task, even with modern Large Language Models (LLMs), as it often involves subtle nuances. This is particularly true in the SE domain, where figurative language is frequently used to convey technical concepts, often bearing developer affect (e.g., `spaghetti code'). Surprisingly, there is a lack of studies on how figurative language in SE communications impacts the performance of automatic tools that focus on understanding developer communications, e.g., bug prioritization, incivility detection. Furthermore, it is an open question to what extent state-of-the-art LLMs interpret figurative expressions in domain-specific communication such as software engineering. To address this gap, we study the prevalence and impact of figurative language in SE communication channels. This study contributes to understanding the role of figurative language in SE, the potential of LLMs in interpreting them, and its impact on automated SE communication analysis. Our results demonstrate the effectiveness of fine-tuning LLMs with figurative language in SE and its potential impact on automated tasks that involve affect. We found that, among three state-of-the-art LLMs, the best improved fine-tuned versions have an average improvement of 6.66% on a GitHub emotion classification dataset, 7.07% on a GitHub incivility classification dataset, and 3.71% on a Bugzilla bug report prioritization dataset.

Understanding and identifying the causes behind developers' emotions (e.g., Frustration caused by `delays in merging pull requests') can be crucial towards finding solutions to problems and fostering collaboration in open-source communities. Effectively identifying such information in the high volume of communications across the different project channels, such as chats, emails, and issue comments, requires automated recognition of emotions and their causes. To enable this automation, large-scale software engineering-specific datasets that can be used to train accurate machine learning models are required. However, such datasets are expensive to create with the variety and informal nature of software projects' communication channels. In this paper, we explore zero-shot LLMs that are pre-trained on massive datasets but without being fine-tuned specifically for the task of detecting emotion causes in software engineering: ChatGPT, GPT-4, and flan-alpaca. Our evaluation indicates that these recently available models can identify emotion categories when given detailed emotions, although they perform worse than the top-rated models. For emotion cause identification, our results indicate that zero-shot LLMs are effective at recognizing the correct emotion cause with a BLEU-2 score of 0.598. To highlight the potential use of these techniques, we conduct a case study of the causes of Frustration in the last year of development of a popular open-source project, revealing several interesting insights.

Emotions play a significant role in teamwork and collaborative activities like software development. While researchers have analyzed developer emotions in various software artifacts (e.g., issues, pull requests), few studies have focused on understanding the broad spectrum of emotions expressed in chats. As one of the most widely used means of communication, chats contain valuable information in the form of informal conversations, such as negative perspectives about adopting a tool. In this paper, we present a dataset of developer chat messages manually annotated with a wide range of emotion labels (and sub-labels), and analyze the type of information present in those messages. We also investigate the unique signals of emotions specific to chats and distinguish them from other forms of software communication. Our findings suggest that chats have fewer expressions of Approval and Fear but more expressions of Curiosity compared to GitHub comments. We also notice that Confusion is frequently observed when discussing programming-related information such as unexpected software behavior. Overall, our study highlights the potential of mining emotions in developer chats for supporting software maintenance and evolution tools.

Interpersonal trust plays a crucial role in facilitating collaborative tasks, such as software development. While previous research recognizes the significance of trust in an organizational setting, there is a lack of understanding in how trust is exhibited in OSS distributed teams, where there is an absence of direct, in-person communications. To foster trust and collaboration in OSS teams, we need to understand what trust is and how it is exhibited in written developer communications (e.g., pull requests, chats). In this paper, we first investigate various dimensions of trust to identify the ways trusting behavior can be observed in OSS. Next, we sample a set of 100 GitHub pull requests from Apache Software Foundation (ASF) projects, to analyze and demonstrate how each dimension of trust can be exhibited. Our findings provide preliminary insights into cues that might be helpful to automatically assess team dynamics and establish interpersonal trust in OSS teams, leading to successful and sustainable OSS.

Continual learning (CL) has remained a persistent challenge for deep neural networks due to catastrophic forgetting (CF) of previously learned tasks. Several techniques such as weight regularization, experience rehearsal, and parameter isolation have been proposed to alleviate CF. Despite their relative success, these research directions have predominantly remained orthogonal and suffer from several shortcomings, while missing out on the advantages of competing strategies. On the contrary, the brain continually learns, accommodates, and transfers knowledge across tasks by simultaneously leveraging several neurophysiological processes, including neurogenesis, active forgetting, neuromodulation, metaplasticity, experience rehearsal, and context-dependent gating, rarely resulting in CF. Inspired by how the brain exploits multiple mechanisms concurrently, we propose TriRE, a novel CL paradigm that encompasses retaining the most prominent neurons for each task, revising and solidifying the extracted knowledge of current and past tasks, and actively promoting less active neurons for subsequent tasks through rewinding and relearning. Across CL settings, TriRE significantly reduces task interference and surpasses different CL approaches considered in isolation., Comment: Accepted at 37th Conference on Neural Information Processing Systems (NeurIPS 2023)

To foster collaboration and inclusivity in Open Source Software (OSS) projects, it is crucial to understand and detect patterns of toxic language that may drive contributors away, especially those from underrepresented communities. Although machine learning-based toxicity detection tools trained on domain-specific data have shown promise, their design lacks an understanding of the unique nature and triggers of toxicity in OSS discussions, highlighting the need for further investigation. In this study, we employ Moral Foundations Theory to examine the relationship between moral principles and toxicity in OSS. Specifically, we analyze toxic communications in GitHub issue threads to identify and understand five types of moral principles exhibited in text, and explore their potential association with toxic behavior. Our preliminary findings suggest a possible link between moral principles and toxic comments in OSS communications, with each moral principle associated with at least one type of toxicity. The potential of MFT in toxicity detection warrants further investigation.

The media's representation of illicit substance use can lead to harmful stereotypes and stigmatization for individuals struggling with addiction, ultimately influencing public perception, policy, and public health outcomes. To explore how the discourse and coverage of illicit drug use changed over time, this study analyzes 157,476 articles published in the Philadelphia Inquirer over a decade. Specifically, the study focuses on articles that mentioned at least one commonly abused substance, resulting in a sample of 3,903 articles. Our analysis shows that cannabis and narcotics are the most frequently discussed classes of drugs. Hallucinogenic drugs are portrayed more positively than other categories, whereas narcotics are portrayed the most negatively. Our research aims to highlight the need for accurate and inclusive portrayals of substance use and addiction in the media.

In the recent past, real-time video processing using state-of-the-art deep neural networks (DNN) has achieved human-like accuracy but at the cost of high energy consumption, making them infeasible for edge device deployment. The energy consumed by running DNNs on hardware accelerators is dominated by the number of memory read/writes and multiplyaccumulate (MAC) operations required. As a potential solution, this work explores the role of activation sparsity in efficient DNN inference. As the predominant operation in DNNs is matrix-vector multiplication of weights with activations, skipping operations and memory fetches where (at least) one of them is zero can make inference more energy efficient. Although spatial sparsification of activations is researched extensively, introducing and exploiting temporal sparsity is much less explored in DNN literature. This work presents a new DNN layer (called temporal delta layer) whose primary objective is to induce temporal activation sparsity during training. The temporal delta layer promotes activation sparsity by performing delta operation facilitated by activation quantization and l1 norm based penalty to the cost function. During inference, the resulting model acts as a conventional quantized DNN with high temporal activation sparsity. The new layer was incorporated as a part of the standard ResNet50 architecture to be trained and tested on the popular human action recognition dataset (UCF101). The method caused 2x improvement in activation sparsity, with 5% accuracy loss., Electrical Engineering

In the recent past, real-time video processing using state-of-the-art deep neural networks (DNN) has achieved human-like accuracy but at the cost of high energy consumption, making them infeasible for edge device deployment. The energy consumed by running DNNs on hardware accelerators is dominated by the number of memory read/writes and multiplyaccumulate (MAC) operations required. As a potential solution, this work explores the role of activation sparsity in efficient DNN inference. As the predominant operation in DNNs is matrix-vector multiplication of weights with activations, skipping operations and memory fetches where (at least) one of them is zero can make inference more energy efficient. Although spatial sparsification of activations is researched extensively, introducing and exploiting temporal sparsity is much less explored in DNN literature. This work presents a new DNN layer (called temporal delta layer) whose primary objective is to induce temporal activation sparsity during training. The temporal delta layer promotes activation sparsity by performing delta operation facilitated by activation quantization and l1 norm based penalty to the cost function. During inference, the resulting model acts as a conventional quantized DNN with high temporal activation sparsity. The new layer was incorporated as a part of the standard ResNet50 architecture to be trained and tested on the popular human action recognition dataset (UCF101). The method caused 2x improvement in activation sparsity, with 5% accuracy loss., Electrical Engineering

In this dissertation, we present an information hiding approach incorporating anonymity that builds on existing classical steganographic models. Current security definitions are not sufficient to analyze the proposed information hiding approach as steganography offers data privacy by hiding the existence of data, a property that is distinct from confidentiality (data existence is known but access is restricted) and authenticity (data existence is known but manipulation is restricted). Combinations of the latter two properties are common in analyses, such as Authenticated Encryption with Associated Data (AEAD), yet there is a lack of research on combinations with steganography. This dissertation also introduces the security definition of Authenticated Stegotext with Associated Data (ASAD), which captures steganographic properties even when there is contextual information provided alongside the hidden data. We develop a hierarchical framework of ASAD variants, corresponding to different channel demands. We present a real-world steganographic embedding scheme, Authenticated SteGotex with Associated tRansaction Data (ASGARD), that leverages a blockchain-based application as a medium for sending hidden data. We analyze ASGARD in our framework and show that it meets Level-4 ASAD security. Finally, we implement ASGARD on the Ethereum platform as a proof-of-concept and analyze some of the ways an adversary might detect our embedding activity by analyzing historical Ethereum data., Lieutenant, United States Navy, Approved for public release. Distribution is unlimited.

In recent years, the Navy has shown interest in an integrated satellite-terrestrial networking (ISTN) architecture for unmanned systems. With the development of satellite networks and growing numbers of unmanned system networks being connected, security and privacy are major concerns in an ISTN. In this thesis, we develop a network intrusion detection system (NIDS) specifically for an ISTN. We identify the critical location of the NIDS within the ISTN architecture and use the decision tree machine learning algorithm to perform cyber-attack detection against various threat vectors, including distributed denial of service. The decision tree algorithm is used to classify and segregate attack traffic from benign traffic. We use an open source ISTN data set available in the literature to train our algorithm. The decision tree is implemented using different split criteria, varying number of splits, and the use of principal component analysis (PCA). We manipulate the size of the training data and the number of data features to achieve reasonable false positive rates. We show that our NIDS framework based on decision tree learning can effectively detect and segregate different attack data classes.

Internet Banking is a course of action of organizations given by a gathering of sorted out bank workplaces. Bank customers may get to their assets from any of the part branch or working environments by means of web. The main problem in Internet Banking is the realness of the client. We proposed framework having the character for each individual note and proficient viable client verification conspire utilizing use diverse cryptographic natives, for example, encryption and pixel distinguishing proof and clients have extra pixel recognizable proof framework. In proposed framework implies that for every last cash in our application surrendered by the client we will produce the interesting id for each money, when the sum is exchanged from source to goal not just the sum and check of the money will be taken not withstanding that one-of-a-kind id will likewise be exchanged with the goal that we can track the way of the cash going around. The Text based password uses username and password. So, recalling of password is necessary which may be a difficult one. Images are generally easier to be remembered than text and in Graphical password; user can set images as their password. We implemented Cued click point (CCP) graphical password which uses circular tolerance.

Science and expertise based on wireless-sensing node technologies have developed in today's health-care environment. Patients are in danger of dying unexpectedly owing to a specific health problem, which is caused by a lack of good medical care provided to patients at the appropriate time. This suggested system uses sensor technologies and the Internet of Things (IoT) to monitor patient health issues and rapidly communicate patient status to clinicians. Temperature sensor, Pulse sensor, Pulse Oximeter, Pressure sensor, ECG, Gyroscope, and Vibration sensor all play essential roles in this suggested system for tracking patient health parameters. All of the sensors are connected to the Arduino Nano, which is used to track the patient's health. The data is delivered immediately to the IoT cloud and stored on an SD card for offline storage utilizing Wi-Fi technology. The Thingspeak software can be used to view the graphical depiction.

Science and expertise based on wireless-sensing node technologies have developed in today's health-care environment. Patients are in danger of dying unexpectedly owing to a specific health problem, which is caused by a lack of good medical care provided to patients at the appropriate time. This suggested system uses sensor technologies and the Internet of Things (IoT) to monitor patient health issues and rapidly communicate patient status to clinicians. Temperature sensor, Pulse sensor, Pulse Oximeter, Pressure sensor, ECG, Gyroscope, and Vibration sensor all play essential roles in this suggested system for tracking patient health parameters. All of the sensors are connected to the Arduino Nano, which is used to track the patient's health. The data is delivered immediately to the IoT cloud and stored on an SD card for offline storage utilizing Wi-Fi technology. The Thingspeak software can be used to view the graphical depiction.

Internet Banking is a course of action of organizations given by a gathering of sorted out bank workplaces. Bank customers may get to their assets from any of the part branch or working environments by means of web. The main problem in Internet Banking is the realness of the client. We proposed framework having the character for each individual note and proficient viable client verification conspire utilizing use diverse cryptographic natives, for example, encryption and pixel distinguishing proof and clients have extra pixel recognizable proof framework. In proposed framework implies that for every last cash in our application surrendered by the client we will produce the interesting id for each money, when the sum is exchanged from source to goal not just the sum and check of the money will be taken not withstanding that one-of-a-kind id will likewise be exchanged with the goal that we can track the way of the cash going around. The Text based password uses username and password. So, recalling of password is necessary which may be a difficult one. Images are generally easier to be remembered than text and in Graphical password; user can set images as their password. We implemented Cued click point (CCP) graphical password which uses circular tolerance.

With the battlespace rapidly shifting to the cyber domain, it is vital to have secure, robust routing protocols for unmanned systems. Furthermore, the development of nano drones is gaining traction, providing new covert capabilities for operators at sea or on land. Deploying a flying ad hoc network (FANET) of nano drones on the battlefield comes with specific performance and security issues. This thesis provides a novel approach to address the performance and security concerns faced by FANET routing protocols, and, in our case, is specifically tailored to improve the Ad Hoc On-Demand Distance Vector (AODV) routing protocol. The proposed routing protocol, Lightweight Secure Ad Hoc On-Demand Distance Vector (LS-AODV), uses a lightweight stream cipher, Trivium, to encrypt routing control packets, providing confidentiality. The scheme also uses Chaskey-12-based message authentication codes (MACs) to guarantee the authenticity and integrity of control packets. We use a network simulator, NS-3, to compare LS-AODV against two benchmark routing protocols, AODV and the Optimized Link State Routing (OLSR) protocol, in order to gauge network performance and security benefits. The simulation results indicate that when the FANET is not under attack from black-hole nodes, LS-AODV generally outperforms OLSR but performs slightly worse than AODV. On the other hand, LS-AODV emerges as the protocol of choice when a FANET is subject to a black-hole attack.

In 2019, the Naval Facilities Engineering Command (NAVFAC) deployed its first smart grid infrastructure in Norfolk, VA, enabling shore commands to meet energy goals set by the secretary of the Navy. However, with increased functionality and control comes increased vulnerability to malicious cyber activity. This research aims to address anomaly detection using an autoencoder neural network as an intrusion detection mechanism on the NAVFAC smart grid. We built and experimented with multiple autoencoder structures to identify an optimal model that provides the best results in terms of precision, recall, and accuracy for the data sets used. We trained our autoencoder on NAVFAC-provided advanced metering infrastructure (AMI) data. We used the NAVFAC smart grid data set to simulate 14 different false data injection attacks (FDIA). Our experiments, performed with Python and TensorFlow, showed that an autoencoder is an effective instruction detection system (IDS) when the threshold is tuned correctly. Moreover, our results show that the activation function and optimizer used may affect performance. Thus, the “best” autoencoder depends on the customer’s needs and the threat environment.

As the Department of Defense (DOD) deploys renewable distributed energy resources (DERs) to reduce fossil fuel consumption, microgrids are being evaluated as one way to generate and deliver reliable electric power to stationary and mobile military units. Commercial off the Shelf (COTS) microgrid components are a viable cost-effective option to setup stand-alone microgrid systems to support mobile military units and help drive the transition to a more sustainable yet energy-resilient military. Reliability and resilience are key parameters in determining the effectiveness of microgrids in supporting military missions. Although in the past few years many researchers have presented reliability and resilience models of various complexity, experimental measurements and model validation are not available in the literature for mobile COTS microgrids. The goal of this thesis research is to experimentally assess the reliability and resilience of stand-alone, mobile microgrids that can be carried by one or two individuals and can be easily assembled in the field in support of operations in locations where utility power is not available. Utilizing COTS DERs including batteries, PV arrays and power converters, three different standalone microgrid architectures were designed, analyzed, and tested in the laboratory. Reliability block diagrams, and system fault trees were created per MIL-HDBK-338B, to compare the reliability of the three microgrid configurations., Major, United States Marine Corps, Approved for public release. Distribution is unlimited.

The research microgrid currently deployed at Marine Corps Air Station, Miramar, is leveraging Verizon’s Non-Standalone (NSA) 5G communications network to provide connectivity between dispersed energy assets and the energy and water operations center (EWOC). Due to its anchor to the Verizon 4G/LTE core, the NSA network does not provide technological avenues for cyber anomaly detection. In this research, we developed a traffic anomaly detection model using supervised machine learning for the energy communication infrastructure at Miramar. We developed a preliminary cyber anomaly detection platform using a convolutional neural network (CNN). We experimentally generated a benign 5G data set using the AT&T 5G cellular tower at the NPS SLAMR facility. We injected synthetic anomalies within the data set to test the CNN and its effectiveness at classifying packets as anomalous or benign. Data sets with varying amounts of anomalous data, ranging from 10% to 50%, were created. Accuracy, precision, and recall were used as performance metrics. Our experiments, conducted with Python and TensorFlow, showed that while the CNN did not perform its best on the data sets generated, it has the potential to work well with a more balanced data set that is large enough to host more anomalous traffic., ONR, Lieutenant, United States Navy, Approved for public release. Distribution is unlimited.

With a rising global emphasis on the use of renewable energy and the reduction of fossil fuels, the Department of Defense is incorporating microgrid technology into energy management systems at forward-deployed and domestic installations. Understanding the nature of the connection between a microgrid and the local utility grid is critical in determining if sensitive loads will be adequately supported. IEEE Standard 1547-2018 applies to this interconnection and specifies the technical requirements and limitations for normal and abnormal operation and expectations for interrupting events. A commercial-off-the-shelf (COTS) microgrid is tested for its compliance to this standard and also for characteristics, such as efficiency, that help define its operational capability. The microgrid can operate in a grid-connected state or an islanded state, disconnected from the utility grid, if the grid is unavailable or if the power quality is unable to support military operations. A microgrid testbed was set up that included the individual COTS components, a centralized control system and several measurement instruments connected that read voltages and currents continuously throughout the experiments. By observing the microgrid functionality, this thesis created an objective characterization template that can be used to assess the capabilities of other COTS microgrids to determine whether they are capable of supporting sensitive loads on forward-deployed or domestic military installations., Captain, United States Marine Corps, Approved for public release. Distribution is unlimited.

In 2019, the Naval Facilities Engineering Command (NAVFAC) deployed its first smart grid infrastructure in Norfolk, VA, enabling shore commands to meet energy goals set by the secretary of the Navy. However, with increased functionality and control comes increased vulnerability to malicious cyber activity. This research aims to address anomaly detection using an autoencoder neural network as an intrusion detection mechanism on the NAVFAC smart grid. We built and experimented with multiple autoencoder structures to identify an optimal model that provides the best results in terms of precision, recall, and accuracy for the data sets used. We trained our autoencoder on NAVFAC-provided advanced metering infrastructure (AMI) data. We used the NAVFAC smart grid data set to simulate 14 different false data injection attacks (FDIA). Our experiments, performed with Python and TensorFlow, showed that an autoencoder is an effective instruction detection system (IDS) when the threshold is tuned correctly. Moreover, our results show that the activation function and optimizer used may affect performance. Thus, the “best” autoencoder depends on the customer’s needs and the threat environment., ONR-ESTEP (Arlington, VA 22203), Lieutenant, United States Navy, Approved for public release. Distribution is unlimited.

The research microgrid currently deployed at Marine Corps Air Station, Miramar, is leveraging Verizon’s Non-Standalone (NSA) 5G communications network to provide connectivity between dispersed energy assets and the energy and water operations center (EWOC). Due to its anchor to the Verizon 4G/LTE core, the NSA network does not provide technological avenues for cyber anomaly detection. In this research, we developed a traffic anomaly detection model using supervised machine learning for the energy communication infrastructure at Miramar. We developed a preliminary cyber anomaly detection platform using a convolutional neural network (CNN). We experimentally generated a benign 5G data set using the AT&T 5G cellular tower at the NPS SLAMR facility. We injected synthetic anomalies within the data set to test the CNN and its effectiveness at classifying packets as anomalous or benign. Data sets with varying amounts of anomalous data, ranging from 10% to 50%, were created. Accuracy, precision, and recall were used as performance metrics. Our experiments, conducted with Python and TensorFlow, showed that while the CNN did not perform its best on the data sets generated, it has the potential to work well with a more balanced data set that is large enough to host more anomalous traffic.

With the battlespace rapidly shifting to the cyber domain, it is vital to have secure, robust routing protocols for unmanned systems. Furthermore, the development of nano drones is gaining traction, providing new covert capabilities for operators at sea or on land. Deploying a flying ad hoc network (FANET) of nano drones on the battlefield comes with specific performance and security issues. This thesis provides a novel approach to address the performance and security concerns faced by FANET routing protocols, and, in our case, is specifically tailored to improve the Ad Hoc On-Demand Distance Vector (AODV) routing protocol. The proposed routing protocol, Lightweight Secure Ad Hoc On-Demand Distance Vector (LS-AODV), uses a lightweight stream cipher, Trivium, to encrypt routing control packets, providing confidentiality. The scheme also uses Chaskey-12-based message authentication codes (MACs) to guarantee the authenticity and integrity of control packets. We use a network simulator, NS-3, to compare LS-AODV against two benchmark routing protocols, AODV and the Optimized Link State Routing (OLSR) protocol, in order to gauge network performance and security benefits. The simulation results indicate that when the FANET is not under attack from black-hole nodes, LS-AODV generally outperforms OLSR but performs slightly worse than AODV. On the other hand, LS-AODV emerges as the protocol of choice when a FANET is subject to a black-hole attack., ONR, Outstanding Thesis, Lieutenant, United States Navy, Approved for public release. Distribution is unlimited.

In recent years, the Navy has shown interest in an integrated satellite-terrestrial networking (ISTN) architecture for unmanned systems. With the development of satellite networks and growing numbers of unmanned system networks being connected, security and privacy are major concerns in an ISTN. In this thesis, we develop a network intrusion detection system (NIDS) specifically for an ISTN. We identify the critical location of the NIDS within the ISTN architecture and use the decision tree machine learning algorithm to perform cyber-attack detection against various threat vectors, including distributed denial of service. The decision tree algorithm is used to classify and segregate attack traffic from benign traffic. We use an open source ISTN data set available in the literature to train our algorithm. The decision tree is implemented using different split criteria, varying number of splits, and the use of principal component analysis (PCA). We manipulate the size of the training data and the number of data features to achieve reasonable false positive rates. We show that our NIDS framework based on decision tree learning can effectively detect and segregate different attack data classes., Civilian, DSO National Labs, Singapore, Approved for public release. Distribution is unlimited.

In this dissertation, we present an information hiding approach incorporating anonymity that builds on existing classical steganographic models. Current security definitions are not sufficient to analyze the proposed information hiding approach as steganography offers data privacy by hiding the existence of data, a property that is distinct from confidentiality (data existence is known but access is restricted) and authenticity (data existence is known but manipulation is restricted). Combinations of the latter two properties are common in analyses, such as Authenticated Encryption with Associated Data (AEAD), yet there is a lack of research on combinations with steganography. This dissertation also introduces the security definition of Authenticated Stegotext with Associated Data (ASAD), which captures steganographic properties even when there is contextual information provided alongside the hidden data. We develop a hierarchical framework of ASAD variants, corresponding to different channel demands. We present a real-world steganographic embedding scheme, Authenticated SteGotex with Associated tRansaction Data (ASGARD), that leverages a blockchain-based application as a medium for sending hidden data. We analyze ASGARD in our framework and show that it meets Level-4 ASAD security. Finally, we implement ASGARD on the Ethereum platform as a proof-of-concept and analyze some of the ways an adversary might detect our embedding activity by analyzing historical Ethereum data.

Internet Banking is a course of action of organizations given by a gathering of sorted out bank workplaces. Bank customers may get to their assets from any of the part branch or working environments by means of web. The main problem in Internet Banking is the realness of the client. We proposed framework having the character for each individual note and proficient viable client verification conspire utilizing use diverse cryptographic natives, for example, encryption and pixel distinguishing proof and clients have extra pixel recognizable proof framework. In proposed framework implies that for every last cash in our application surrendered by the client we will produce the interesting id for each money, when the sum is exchanged from source to goal not just the sum and check of the money will be taken not withstanding that one-of-a-kind id will likewise be exchanged with the goal that we can track the way of the cash going around. The Text based password uses username and password. So, recalling of password is necessary which may be a difficult one. Images are generally easier to be remembered than text and in Graphical password; user can set images as their password. We implemented Cued click point (CCP) graphical password which uses circular tolerance.

Science and expertise based on wireless-sensing node technologies have developed in today's health-care environment. Patients are in danger of dying unexpectedly owing to a specific health problem, which is caused by a lack of good medical care provided to patients at the appropriate time. This suggested system uses sensor technologies and the Internet of Things (IoT) to monitor patient health issues and rapidly communicate patient status to clinicians. Temperature sensor, Pulse sensor, Pulse Oximeter, Pressure sensor, ECG, Gyroscope, and Vibration sensor all play essential roles in this suggested system for tracking patient health parameters. All of the sensors are connected to the Arduino Nano, which is used to track the patient's health. The data is delivered immediately to the IoT cloud and stored on an SD card for offline storage utilizing Wi-Fi technology. The Thingspeak software can be used to view the graphical depiction.

Internet Banking is a course of action of organizations given by a gathering of sorted out bank workplaces. Bank customers may get to their assets from any of the part branch or working environments by means of web. The main problem in Internet Banking is the realness of the client. We proposed framework having the character for each individual note and proficient viable client verification conspire utilizing use diverse cryptographic natives, for example, encryption and pixel distinguishing proof and clients have extra pixel recognizable proof framework. In proposed framework implies that for every last cash in our application surrendered by the client we will produce the interesting id for each money, when the sum is exchanged from source to goal not just the sum and check of the money will be taken not withstanding that one-of-a-kind id will likewise be exchanged with the goal that we can track the way of the cash going around. The Text based password uses username and password. So, recalling of password is necessary which may be a difficult one. Images are generally easier to be remembered than text and in Graphical password; user can set images as their password. We implemented Cued click point (CCP) graphical password which uses circular tolerance.