DELL (United States)

Understanding how underrepresented women with pelvic floor disorders use social media can guide clinicians in patient outreach and reducing care disparities. The objectives of this study were to characterize which social media platforms underrepresented populations use most and identify preferences of use when learning about their conditions. This cross-sectional survey of underrepresented women used a 21-item questionnaire assessing demographics, internet use, and social media preferences. Descriptive statistics and χ2 tests were performed. A total of 301 surveys were analyzed. Most participants identified as Hispanic/Latina (65.4%). The mean age was 56.8±14.2 years. Social media use did not differ by race. Participants younger than 65 years were more likely than those 65 years and older to use the internet or social media to learn about their conditions (67.2% vs. 45.4%, P <0.01). More participants younger than 65 years used TikTok (20.1% vs 5.2%, P <0.01) and Instagram (24.5% vs 8.2%, P=0.0011) compared with participants 65 years and older. Participants with ≥4 social media accounts used the internet more often than those without (81% vs 32%, P <0.01). Most participants used Facebook (46.5%) or YouTube (35.5%) to learn about their conditions and would be most interested in using YouTube (27.9%) and Facebook (26.2%) to learn more about their conditions. Most participants preferred physicians use Facebook (35.2%) or YouTube (22.6%). Our study population reported frequent social media use and interest in learning about their conditions via social media with platform preferences differing by age. Patient education and research recruitment strategies in underrepresented patient populations could be improved by incorporating these platforms.

The term Industrial Processes Industry 6.0 refers to a revolutionary approach to industrial processes that can be produced from the convergence of remote sensing technology and Artificial Intelligence (AI). This chapter explores the revolutionary possibility of AI-driven remote sensing to further sustainable economic development and respond to urgent environmental issues by systematically integrating advanced sensing capabilities with AI algorithms. These technologies work together synergistically, enabling efficient resource management, predictable maintenance, and proactive environmental protection measures. This research focuses on the recent advancements in AI-based remote sensing, its applications across the industrial sectors, and its application in encouraging sustainability. It dwells on issues of data privacy, technological constraints, and standardized practices in the implementation of these technologies. It also discusses possible mitigation activities to overcome these difficulties and maximize the benefits of AI-driven remote sensing in a commercial context. This chapter provides an overview of practical implications and prospects by analysing the case and experimental studies. The results elucidate the importance of AI-powered remote sensing for realizing sustainable industrial development and reducing ecological damage. This paper brings insight into the views of scholars, industrial strategists, and government policymakers in the quest for striking a balance between being good stewards of the environment and achieving technological advancement. This study contributes to the increasingly growing body of literature on Industry 6.0.

Psychiatric comorbidities associated with dermatological diseases, particularly in pediatric populations, are relatively understudied compared to psychiatric comorbidities of other chronic diseases. Research suggests that skin conditions in adolescence often co-occur with mental health difficulties, including heightened risk for suicide. Dermatologists play a vital role as early identifiers of suicidal thoughts and behaviors (STB) for their adolescent patients. Currently, there are no recommendations for screenings or assessments related to suicide risk in dermatologic settings. We provide evidence-based tools that can be helpful for STB assessment and highlight clinical pearls for increasing comfort when suicidal thoughts or behaviors occur in the pediatric dermatologist's office.

This is a companion review article to a narrative medicine story in this issue of the journal. “In The Ring” is a true story of a patient’s father’s near-death experience of Hell and the history of its follow up impact on the patient and her doctor. This article explores the medical perspective on NDE’s. How are NDE’s different from other altered mental states? What about deep anesthesia recollections? How do the NDE’s compare to psychedelic experiences and the new psylocibin treatments for psychiatric conditions? How are they different from our usual dreams and nightmares? What is the line between our imagination, dreams, and our conscious state? What determines that line? What does science now understand about consciousness itself? What is the difference between computer artificial intelligence (AI) and sentience? Can there be “machine consciousness”? A deep dive into these existential and highly relevant questions is beyond the scope of this manuscript, but this article is an attempt to succinctly address some of these questions in relation to this journal’s narrative medicine story of Hell, “In The Ring” and presents a similar additional distressing NDE case.

The figure drawing in this Art in Medicine article was created by the author as a first-year medical student attending a live model art class. We might call this work, “The Side Sleeper”. In deconstructing the anatomic axes and posturing of the model in this artwork, we get an inkling of how figure drawing might meld with improving a physician’s grasp of anatomy and kinesiology. This article will explore the developmental stages of figure drawing from childhood, moving to the development of adult artistic rendering of the human figure, and then discuss the potential impact of art education within medical education.

Deep Reinforcement Learning (DRL) has recently become a powerful tool for addressing complex optimization problems in mobile networks. Among DRL methods, Soft Actor-Critic (SAC), an off-policy algorithm, stands out due to its sample efficiency and ability to learn stable and resilient policies. This work explores the application of SAC to the resource allocation problem in Open Radio Access Networks (O-RAN) with stringent Quality of Service (QoS) requirements. The performance of SAC is benchmarked against Proximal Policy Optimization (PPO), a strong on-policy DRL algorithm, and a greedy baseline. Experimental results show that while both DRL methods outperform the greedy strategy, SAC consistently achieves faster convergence and better long-term stability compared to PPO, making it highly suitable for dynamic O-RAN environments.

As Artificial Intelligence (AI) adoption accelerates across industries, the need for robust real-time cybersecurity solutions becomes increasingly critical. Traditional security frameworks that rely on static rules, signature-based detection, and threshold alerts struggle to detect sophisticated identity-based attacks that exploit behavioral anomalies rather than explicit policy violations. These limitations, combined with the growing volume of telemetry data and the latency of conventional detection methods, leave organizations vulnerable to prolonged breaches and data exfiltration. This paper presents a use case with a scalable, AI-driven approach to sensitive data leak detection using NVIDIA Morpheus, a GPU-accelerated cybersecurity framework. We demonstrate how Morpheus, in conjunction with NVIDIA BlueField DPUs and DOCA, enables high-throughput, low-latency inspection of real-time network telemetry without impacting system performance. The solution leverages a pre-trained NLP-based Sensitive Information Detection model capable of identifying ten categories of sensitive data within packet payloads, including credentials, PII, and cryptographic keys. Network-captured PCAP data serialized in JSONlines format passes through a linear, feed-forward Morpheus pipeline, configured via CLI and powered by Triton Inference Server with dynamic batching for optimized inference. The final output provides binary classification results that indicate the presence or absence of sensitive information. This case study illustrates how AI-enhanced telemetry analysis can significantly reduce detection and response times, offering a proactive defense mechanism against modern cyber threats.

Cryptographically strong hashes are an integral part of deduplicated storage to identify unique chunks, so they are called fingerprints. Many systems were designed with MD5 or SHA-1 hashes, which seemed sufficiently strong at the time. With the ever-increasing computational power and ongoing security attacks, the security and reliability of the original fingerprints in deduplicated storage may be compromised, making them vulnerable to attacks, which can result in data loss or leakage. This may be a concern for deduplicated storage customers that have already purchased systems, who now are required to switch to a stronger hash function such as SHA-256, possibly due to future legal requirements. This reveals the need for upgrading original fingerprints to stronger hashes on existing duplicated storage systems, but there is currently a lack of research on how to perform such an upgrade effectively. Fingerprint upgrade entails not only adding a new fingerprint calculation function but also replacing original fingerprints in existing, large data structures with new fingerprints. One direct method is to first reconstruct all logical files and then deduplicate them again using new fingerprint calculations. However, such a method will repeatedly process duplicate chunks, resulting in significant space and time overhead. In this paper, we propose an upgrade method that decouples computationally and spatially expensive operations from repetitive operations, so that we can release the space of duplicate chunks and avoid excessive space consumption. We also propose leveraging the characteristics of deduplicated data and file similarities to optimize the cache line and data access pattern, respectively, which further reduces the time overhead. Experimental results demonstrate that our upgrade method is practical and efficient, achieving a throughput of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\sim 1.7 \text{GB} / \mathrm{s}$</tex>.