New York Institute of Technology

Blockchain technology has emerged as a transformative innovation, providing a transparent, immutable, and decentralized platform that underpins critical applications across industries such as cryptocurrencies, supply chain management, healthcare, and finance. Despite their promise of enhanced security and trust, the increasing sophistication of cyberattacks has exposed vulnerabilities within blockchain ecosystems, posing severe threats to their integrity, reliability, and adoption. This study presents a comprehensive and systematic review of blockchain vulnerabilities by categorizing and analyzing potential threats, including network-level attacks, consensus-based exploits, smart contract vulnerabilities, and user-centric risks. Furthermore, the research evaluates existing countermeasures and mitigation strategies by examining their effectiveness, scalability, and adaptability to diverse blockchain architectures and use cases. The study highlights the critical need for context-aware security solutions that address the unique requirements of various blockchain applications and proposes a framework for advancing proactive and resilient security designs. By bridging gaps in the existing literature, this research offers valuable insights for academics, industry practitioners, and policymakers, contributing to the ongoing development of robust and secure decentralized ecosystems.

The present study aims to examine how different geometrically shaped trays of similar volume and construction heat when a food model system is packaged inside them. Triangular, rectangular, oval, and round‐shaped trays were filled with a food simulant and processed in a water spray retort system at a rotational speed of 6 rpm. Each retortable tray was fitted with wired thermocouples to monitor temperature variation within the trays. After filling the simulant, the trays were hermetically sealed. Trays were processed in the center of a retort rack in the vessel to prevent any movement during retort processing. Heat penetration data were collected using a total of seven thermocouples placed in each tray. A thermocouple was positioned at the geometric center of the trays, while the remaining six were placed around the tray. All thermocouple ends were located along the same center plane of each tray. Data was collected via data collection software to evaluate the heating profile of each tray. The oval‐shaped (OS) tray reached internal equilibrium the fastest (p < 0.05) in 26 min. into processing, with the round‐shaped (RS) tray following 28 min. into processing. Modeling software was then used to illustrate heat penetration data. Heat maps for the central plane of each shaped tray were generated at 2, 6, 10, 14, 18, 22, 26, and 28 min. based on physical measurements obtained from thermocouples. These maps were created to assist in understanding the heating effects influenced by tray geometry.

Research Question/Issue Although scholars drawing on the classic agency framework predict a positive relationship between board independence and bank risk‐taking, scholars adopting a contingency approach predict that such a relationship may not be positive. This theoretical debate is reflected in the inconclusive empirical evidence. In this study, building on the new institutional economics tradition, we go beyond a universal approach to our focal relationship and further advance a contingency approach to the agency framework. Research Findings/Insights In a retrospective meta‐analysis combining 81 primary studies from 2002 to 2022 that cover 106 countries, we find that the board independence–bank risk‐taking relationship is not universal but contingent on the institutional conditions facing the bank. Specifically, we find that our focal relationship is positively moderated by the level of depositor monitoring and negatively moderated by regulatory stringency. Moreover, we find that the effect of depositor monitoring is negatively moderated by regulatory stringency. Theoretical/Academic Implications Our study simultaneously relaxes two central assumptions of classic agency theory and shows that formal institutions, and in particular those protecting stakeholders' interests, can directly affect independent directors' efficacy as shareholder agents. We also point out the ability and incentives of independent directors to monitor in the interest of shareholders as the relevant causal mechanisms. Lastly, we add to research adopting a polycentric view of institutions and complement a burgeoning research stream examining corporate governance bundles. Practice/Policy Implications Our study adds to the debate among practitioners and policymakers on what constitutes good bank governance. It also intersects the debate on the relative merits of market and regulatory discipline in banking.

We study a semilinear hyperbolic system of PDEs which arises as a continuum approximation of the discrete nonlinear dimer array model introduced by Hadad et al. (ACS Photonics 4:1974–1979, 2017). We classify the system’s traveling waves, and study their stability properties. We focus on traveling pulse solutions (“solitons”) on a nontrivial background and moving domain wall solutions (kinks); both arise as heteroclinic connections between spatially uniform equilibria of a reduced dynamical system. We present analytical results on: nonlinear stability and spectral stability of supersonic pulses, and spectral stability of moving domain walls. Our stability results are in terms of weighted $H^1$ norms of the perturbation, which capture the phenomenon of convective stabilization; as time advances, the traveling wave “outruns” the growing disturbance excited by an initial perturbation; the nontrivial spatially uniform equilibria are linearly exponentially unstable. We use our analytical results to interpret phenomena observed in numerical simulations.

The quantum state diffusion (QSD) equation technique has been used to effectively deal with the dynamics of the open quantum systems. Normally, the initial states of the baths are taken as vacuum states. In this paper, we use the squeezed vacuum states of the baths as the initial states. Then, the squeezing parameters are naturally introduced to the non-Markovian dynamics of the system. By using the QSD equation technique, a non-Markovian master equation in squeezed thermal baths has been derived under the weak system-bath coupling, high-temperature approximation. The dynamics of the systems can be numerically calculated by the master equation together with a group of closed $\overline{O}$($\overline{Q}$) operator equation. Taking a single and two-qubit coupled with the squeezed bath as examples, the dynamics of the spin state $\left\langle \sigma _{z}\right\rangle$ or correlation $\left\langle \sigma _{z}^{A}\sigma _{z}^{B}\right\rangle$ are numerically calculated. The effects of the squeezing and memory effects on the dynamics are analyzed. For both models, big p-quadrature squeezing or long memory time (strong non-Markovianity) of the baths corresponds to big values of $\left\langle \sigma _{z}\right\rangle$ or $\left\langle \sigma _{z}^{A}\sigma _{z}^{B}\right\rangle$. When the squeezing strength is zero, the correlation functions go back to the vacuum initial state cases. The developed technique in this paper provides an effective approach to analyze the impact of multiple parameters on the systems in squeezed thermal baths.

Background/Objectives: Electrocardiogram data are widely used to diagnose cardiovascular diseases, a leading cause of death globally. Traditional interpretation methods are manual, time-consuming, and prone to error. Machine learning offers a promising alternative for automating the classification of electrocardiogram abnormalities. This study explores the use of machine learning models to classify electrocardiogram abnormalities using a dataset that combines clinical features (e.g., age, weight, smoking status) with key electrocardiogram measurements, without relying on time-series data. Methods: The dataset included demographic and electrocardiogram-related biometric data. Preprocessing steps addressed class imbalance, outliers, feature scaling, and the encoding of categorical variables. Five machine learning models—Gaussian Naive Bayes, support vector machines, random forest trees, extremely randomized trees, gradient boosted trees, and an ensemble of top-performing classifiers—were trained and optimized using stratified k-fold cross-validation. Model performance was evaluated on a reserved testing set using metrics such as accuracy, precision, recall, and F1-score. Results: The extremely randomized trees model achieved the best performance, with a testing accuracy of 66.79%, recall of 66.79%, and F1-score of 62.93%. Ventricular rate, QRS duration, and QTC (Bezet) were identified as the most important features. Challenges in classifying borderline cases were noted due to class imbalance and overlapping features. Conclusions: This study demonstrates the potential of machine learning models, particularly extremely randomized trees, in classifying electrocardiogram abnormalities using demographic and biometric data. While promising, the absence of time-series data limits diagnostic accuracy. Future work incorporating time-series signals and advanced deep learning techniques could further improve performance and clinical relevance.

BACKGROUND Cerebrovascular malformations are a pivotal cause of hemorrhage and neurological disability alongside lacking effective medication. Thyroid hormones (THs), including thyroxine and triiodothyronine, are essential for vascular development, yet whether they participate in malformed cerebrovascular pathology remains elusive. METHODS Single-cell transcriptome analysis characterized human cerebral cavernous malformations and brain arteriovenous malformations, 2 typical cerebrovascular malformation diseases. Adeno-associated virus–mediated DIO2 (iodothyronine deiodinase 2; an enzyme that converts thyroxine to active triiodothyronine) overexpression/knockdown or triiodothyronine/methimazole (an antithyroid drug) treatment was applied to mouse models of cerebral cavernous malformations (endothelial-specific Pdcd10 knockout mice, Pdcd10 KO) and brain arteriovenous malformations (endothelial-specific Kras G12D mutant mice, Kras G12D ) to evaluate the involvement of DIO2 and TH signaling in cerebrovascular malformations. RESULTS TH signaling was markedly activated in fibroblasts of human cerebral cavernous malformation and arteriovenous malformation single-cell samples, accompanied by elevated DIO2 expression. Similar DIO2 upregulation was observed in cerebrovascular fibroblasts of Pdcd10 KO/ Kras G12D mice and patient brain sections. Exogenous DIO2 or triiodothyronine replenishment effectively reduced brain hemorrhage, excessive ECM (extracellular matrix) remodeling, and vascular leakage in juvenile and adult male and female Pdcd10 KO/ Kras G12D mice. In contrast, DIO2 silencing or TH inhibition deteriorated vascular anomalies. Mechanistically, transcription factor Foxk1 (forkhead box K1) was determined to interact with the DIO2 promoter region. The activation of fibroblast PI3K-Akt-mTOR signaling in Pdcd10 KO/ Kras G12D mice triggered Foxk1 nuclear translocation to promote DIO2 transcription. Triiodothyronine treatment mitigated inflammatory infiltration, normalized mitochondrial morphology, and restored mitochondrial biogenesis in malformed brain vessels by activating the Pgc1a (peroxisome proliferator-activated receptor gamma coactivator 1-alpha)-Sod2 (superoxide dismutase 2)/Prdx3 (peroxiredoxin 3)/Gpx1 (glutathione peroxidase 1) axis to reduce reactive oxygen species accumulation. We also determined that the vascular repair effects of triiodothyronine were Pgc1a-dependent. CONCLUSIONS We delineate a novel DIO2-mediated adaption in malformed cerebrovasculature and conclude that targeting TH signaling may represent a potential therapy for cerebrovascular disorders.

With the booming development of the digital economy, the dominant position of platform enterprises in various industries is gradually emerging. Many platform enterprises have formed market monopolies through mechanisms such as mergers and acquisitions, data control, and network effects. This monopoly phenomenon not only affects market competition and harms consumer interests, but also brings problems such as data privacy breaches. This article explores the mechanism of platform enterprise monopoly formation in the digital economy era, including factors such as increased market concentration, data barriers, and cross-border expansion. At the same time, it analyzes the harm of platform monopoly, such as damaged consumer welfare and limited market innovation. This article proposes effective regulatory strategies to address these issues, including improving anti-monopoly laws, strengthening data supervision, and promoting transparent platform operations. Finally, this article emphasizes the importance of international cooperation in regulating the digital economy and calls for the establishment of a unified regulatory framework worldwide.

Importance Treatment options for amyotrophic lateral sclerosis (ALS) remain suboptimal. Results from a phase 2 study of reldesemtiv in ALS suggested that it may slow disease progression. Objective To assess the effect of reldesemtiv vs placebo on functional outcomes in ALS. Design, Setting, and Participants A Study to Evaluate the Efficacy and Safety of Reldesemtiv in Patients With Amyotrophic Lateral Sclerosis (COURAGE-ALS) was a double-blind, placebo-controlled phase 3 randomized clinical trial conducted at 83 ALS centers in 16 countries from August 2021 to July 2023. The first 24-week period was placebo controlled vs reldesemtiv. All participants received reldesemtiv during the second 24-week period with a 4-week follow-up. Two interim analyses were planned, the first for futility and the second for futility and possible resizing. This was a hybrid decentralized trial with approximately half the trial visits performed remotely and the remaining visits in the clinic. Eligible participants met criteria for definite, probable, or possible ALS with lower motor neuron signs by modified El Escorial Criteria, ALS symptoms for 24 months or less, ALS Functional Rating Scale–Revised (ALSFRS-R) total score of 44 or less, and forced vital capacity of greater than or equal to 65% of predicted. Interventions Oral reldesemtiv, 300 mg, or placebo twice daily. Main Outcomes and Measures The primary end point was change in ALSFRS-R total score from baseline to week 24. Results Of the 696 participants screened, 207 were screen failures. A total of 486 participants (mean [SD] age, 59.4 [10.9] years; 309 male [63.6%]) were randomized to reldesemtiv (n = 325) or placebo (n = 161); 3 randomized patients were not dosed. The second interim analysis at 24 weeks after randomization included 256 participants. The data monitoring committee recommended that the trial should end due to futility, and the sponsor agreed. The mean (SE) group difference in the ALSFRS-R score from baseline to week 24 was −1.1 (0.53; 95% CI, −2.17 to −0.08; P = .04, favoring placebo). Given excess missing data from early termination, the combined assessment assumed greater importance; it, too, failed to show a benefit from treatment with reldesemtiv (win probability was 0.44 for reldesemtiv and 0.49 for placebo, with a win ratio of 0.91; 95% CI of win ratio, 0.77-1.10; P = .11). Conclusions and Relevance This randomized clinical trial failed to demonstrate efficacy for reldesemtiv in slowing functional decline in ALS. Trial Registration ClinicalTrials.gov Identifier: NCT04944784

Access to high‐quality outreach programs is crucial for preparing students for STEM careers, yet traditional classrooms often lack diverse, hands‐on learning opportunities, particularly in anatomy and evolutionary biology. We present "Are You Stronger Than a Lemur?"—an interactive STEM activity that introduces K‐12 students to fundamental concepts in anatomy, evolution, physics, and data analysis through real‐world applications. Participants formulate hypotheses, collect and analyze data, and engage with age‐tailored educational materials that support differentiated learning. We assessed the program's effectiveness through pre‐ and post‐program knowledge assessments across 1670 participants (1045 eligible responses) from the United States and Mongolia. Results showed a significant increase in knowledge acquisition in anatomy, evolution, physics, statistics, and zoology. After controlling for confounding variables, we also observed a significant increase in interest in STEM careers. "Are You Stronger Than a Lemur?" bridges gaps in STEM education, particularly in underrepresented fields like anatomy and evolutionary biology, by providing an adaptable program suited to different age groups, genders, and countries. Its success lies in connecting theoretical concepts to tangible data, fostering critical thinking, problem‐solving, and data interpretation skills. The program not only reinforces core STEM concepts but also offers students a unique, engaging experience that deepens their understanding and enhances their potential for future STEM careers.

Energy is a crosscutting concept in science, but college students often perceive a mismatch between how their biology and chemistry courses discuss the topic. The challenge of reconciling these disciplinary differences can promote faulty reasoning—for example, biology students often develop the incorrect idea that breaking bonds is exothermic and releases energy. We hypothesize that one source of this perceived mismatch is that biology and chemistry textbooks use different visual representations of bond breaking and formation. We analyzed figures of ATP hydrolysis from 12 college‐level introductory biology textbooks and coded each figure for its representation of energy, bond formation, and bond breaking. For comparison, we analyzed figures from six college‐level introductory chemistry textbooks. We found that the majority (70%) of biology textbook figures presented ATP hydrolysis in the form “one reactant → multiple products” and “more bonds in reactants → fewer bonds in products”. In contrast, chemistry textbook figures of the form “one reactant → multiple products” and “more bonds → fewer bonds” were predominantly endothermic reactions, which directly contradicts the exothermic nature of ATP hydrolysis. We hypothesize that these visual inconsistencies may be a contributing factor to student struggles in constructing a coherent mental model of energy and bonding.

Spending time with others affords numerous benefits. One way a person can spend time with others is through a self-invitation—asking to join the plans of others. We address the psychological processes involved with self-invitations to everyday social activities from both the self-inviter’s perspective and the perspective of those with the plans (“plan-holders”). Across eight studies (seven preregistered), we demonstrate that potential self-inviters fail to ask to join the plans of others as often as plan-holders would prefer, because potential self-inviters overestimate how irritated plan-holders would be by such self-invitations. Further, we show that these asymmetries are rooted in differing viewpoints about the mindsets of plan-holders when they originally made the plans. Namely, potential self-inviters exaggerate the likelihood that plan-holders had already considered inviting them but decided against it (vs. made plans without considering inviting them). We conclude by discussing the various implications of our findings.

Athletic success depends on several factors, including measurable factors such as training, sleep, and mental state. The women’s Basketball team at Sacred Heart University, USA, has been monitored over two consecutive seasons. The first season, 2021/22, was relatively unsuccessful, followed by a much-improved performance in the 2022/23 season, with a higher win percentage. Six metrics have been measured consistently: Training, sleep, mental state, game workload, jump analysis, and game performance. We compare those metrics over the two seasons, and our findings show the direct relationship between better training, better sleep, and mental health on the team's performance as a group. We analyze the performance of the players common to both seasons and note the improvement of this group's fitness over the two seasons (3.5% better sleep, 8% in recovery, 12% in stress, and 13% in jump height) even before the games started, and the effect of the new players on the team performance.

21st century urban design must configure densely occupied urban settlements that offset undesirable local outcomes through climate-sensitive urban design. To bridge between climate science and climate action, policymakers/stakeholders need tools and methods to identify, configure and evaluate evidence-based Urban Climate Factors at urban and local scales. The goal is to configure densely occupied urban settlements that offset undesirable local outcomes through climate-sensitive strategies. The challenge is configuring these districts to reduce the impact of increased urban heat island (UHI) and flooding due to the changing climate while enhancing an equitable, high-quality, low-carbon lifestyle. Enabling transformative climate action in cities requires expanding on the traditional influence and capabilities of urban planning and urban design. In response, we have developed a climate-driven planning process called the Urban Design Climate Workshop whose urban climate factors overlap with stakeholder-driven performance indicators.

Background: This study investigates the application of machine learning models to classify electrocardiogram signals, addressing challenges such as class imbalances and inter-class overlap. In this study, “normal” and “abnormal” refer to electrocardiogram findings that either align with or deviate from a standard electrocardiogram, warranting further evaluation. “Borderline” indicates an electrocardiogram that requires additional assessment to distinguish benign variations from pathology. Methods: A hierarchical framework reformulated the multi-class problem into two binary classification tasks—distinguishing “Abnormal” from “Non-Abnormal” and “Normal” from “Non-Normal”—to enhance performance and interpretability. Convolutional neural networks, deep neural networks, and tree-based models, including Gradient Boosting Classifier and Random Forest, were trained and evaluated using standard metrics (accuracy, precision, recall, and F1 score) and learning curve convergence analysis. Results: Results showed that convolutional neural networks achieved the best balance between generalization and performance, effectively adapting to unseen data and variations without overfitting. They exhibit strong convergence and robust feature importance rankings, with ventricular rate, QRS duration, and P-R interval identified as key predictors. Tree-based models, despite their high performance metrics, demonstrated poor convergence, raising concerns about their reliability on unseen data. Deep neural networks achieved high sensitivity but suffered from overfitting, limiting their generalizability. Conclusions: The hierarchical binary classification approach demonstrated clinical relevance, enabling nuanced diagnostic insights. Furthermore, the study emphasizes the critical role of learning curve analysis in evaluating model reliability, beyond performance metrics alone. Future work should focus on optimizing model convergence and exploring hybrid approaches to improve clinical applicability in electrocardiogram signal classification.

Knee osteoarthritis, a degenerative joint disease, results in the gradual deterioration and eventual loss of knee cartilage, causing pain, stiffness, and difficulties in movement. Initiating treatment based solely on symptoms can lead to irreversible joint changes, emphasizing the importance of early detection. This study applies various machine learning models to expedite and refine the detection of knee osteoarthritis using knee X-ray images. The study analyzed several commonly utilized models, including Convolutional Neural Networks, Spiking Neural Networks, Google Teachable Machine, Support Vector Machines, and a Convolutional Neural Network enhanced with a pre-trained VGG16. The models were trained using a collection of images, each representing a different stage of osteoarthritis according to the Kellgren-Lawrence scale. Among these models, the Google Teachable Machine demonstrated the greatest capability in analyzing knee X-ray images and categorizing them into different severity grades. While the accuracies of these models are not yet sufficient for clinical application, this approach shows potential as a clinical tool. With further improvement, it could enhance treatment effectiveness by enabling proactive management of knee osteoarthritis, even before symptoms appear.

The acuity of our vision and hearing depends on the accuracy of sensory receptors and the sensory processing capabilities of the brain. Although advancements have made it possible to restore impaired sensory receptors, methods that can rapidly improve sensory processing in the brain remain limited. Recent research with rodents discovered a new potential intervention that can immediately improve sensory processing in the brain by continuously activating the locus coeruleus-norepinephrine system directly or indirectly via vagus nerve stimulation. Here, we describe our recent pilot clinical study demonstrating that neuromodulation via transcutaneous cervical vagus nerve stimulation (tcVNS) can similarly enhance sensory performance in human adults. We conducted three sham-controlled experiments that assessed the impact of continuous tcVNS on auditory and visual perception. Participants engaged in auditory and visual psychophysics tasks while receiving continuous tcVNS or sham stimulation. Compared to sham, tcVNS evoked 37% better auditory performance (p = 0.00052) and 23% better visual performance (p = 0.038). Notably, participants with lower sensory performance during sham stimulation experienced larger tcVNS-evoked improvements (p = 0.0040), suggesting that this technology may be particularly beneficial for clinically impaired senses. These results establish the effectiveness of continuous tcVNS in humans, positioning it as a potential intervention that can alleviate symptoms of central sensory processing dysfunction on demand.

Cardiac remodeling is the process of adaptive or maladaptive growth of the heart in response to altered loading conditions or growth stimuli. A landmark review by Linzbach in 1960 and reports by Grant (1965) and Grossman (1975) brought attention to anatomical remodeling of the heart in cardiac hypertrophy and heart failure (HF). This was largely the age of cardiac physiology with many focusing on in vivo and in vitro studies in animal models of heart disease. The neurohormonal hypothesis became a major driving force with realization that plasma norepinephrine levels increase with progression to HF. This led drug companies to develop compounds aimed at these targets. Prior to the discovery of angiotensin converting enzyme (ACE) inhibitors, available drugs offered symptomatic relief but had little effect on mortality. Cardiac remodeling became a hot area of HF research in the late 1980s and early 1990s. This field took off when investigators observed that reductions in mortality by neurohormonal inhibition in HF were intimately linked to beneficial changes in cardiac anatomy. More recent work, highlights the critical role of thyroid hormones (THs) in maintaining myocyte shape and internal myocyte structures involved in calcium handling. This overview focuses on the role of myocyte remodeling related to chamber remodeling and wall stress. Another goal was to provide technical advice and information for researchers to improve critical analysis of data in this area of research. A comprehensive understanding of the molecular basis of myocyte remodeling is evolving and would require a separate communication to address.

Simultaneous contrast is the phenomenon in which the background influences perceived color. Existing color appearance models (CAMs) are inadequate in fully capturing this effect. This research aims to provide a deeper understanding of the simultaneous contrast effect and to extend the state‐of‐the‐art color appearance model, CIECAM16, to better account for this phenomenon. By conducting three experiments, we empirically assessed the effects of simultaneous contrast on chroma, hue, and lightness. The findings revealed that a stimulus's perceived color attributes changed significantly depending on the color properties of the background. Based on these results, we extended CIECAM16 to account for the simultaneous contrast effect. The results demonstrated that the extended CIECAM16 model significantly improved the ability to predict color appearance while accounting for simultaneous contrast and offered closer alignment with observed findings and visual mechanisms. This research enhances the understanding of simultaneous contrast and contributes to improving the CIECAM16 model.