IDEX Corporation (United States)

The online version contains supplementary material available at 10.1007/s13534-025-00533-0.

Managing metadata at scale in distributed Extract, Transform, Load (ETL) ecosystems present significant challenges including schema drift, source-target mapping inconsistencies, and error propagation across data pipelines. This paper introduces a novel reinforcement learning-based autonomous metadata management framework that dynamically adapts to schema evolution, optimizes source-target mapping configurations, and implements self-correcting mechanisms for data quality anomalies. The proposed system leverages deep Q-networks (DQN) and policy gradient methods to continuously learn from historical ingestion patterns, schema change events, and anomaly occurrences within modern cloud-native data platforms. Implementation utilizing Snowflake Data Cloud, Databricks Unified Analytics Platform, and Amazon Web Services (AWS) storage services demonstrates the framework's effectiveness across heterogeneous data environments. Experimental validation conducted on Truist Financial Corporation's enterprise data lakes shows a 67% reduction in manual metadata correction efforts and 40% improvement in data availability Service Level Agreements (SLAs), while maintaining 99.7% data accuracy across distributed data pipelines processing over 2.5 petabytes of financial data monthly.

Background: Mismatch negativity (MMN) is a component of the auditory event-related potential (ERP) that is elicited during a passive oddball paradigm where task-irrelevant infrequent deviants are presented in a stream of more frequent standard stimuli. MMN is believed to index a pre-attentive stage of auditory information processing closely linked to N-methyl-D-aspartate receptors (NMDAR). Ketamine is thought to act primarily as an NMDAR antagonist, has been used in clinical trials to model the symptoms of schizophrenia and is increasingly used in the clinic to treat depression. Various studies have reported that ketamine reduces MMN amplitude which, in turn, might reflect reduced function of NMDAR-mediated neurotransmission. Nonetheless, there is growing evidence showing MMN amplitude either having high variability or, paradoxically, moving in the opposite direction after ketamine in different individuals. Methods: In here, we analyzed results from three independent ERP studies to test the hypothesis of a cross-over interaction (″disordinal″ drug effect) between the duration-deviant MMN at baseline (without ketamine) and the direction and magnitude of the ketamine effect. To rule out regression to the mean (RTM), a statistical phenomenon that may also partially explain this cross-over interaction, we separately estimated RTM using a drug-free test-retest study. Results: Our results are the first to statistically demonstrate the existence of a disordinal drug response to ketamine, where the direction and magnitude of ketamine-induced changes in MMN amplitude can be predicted by baseline MMN amplitude. Conclusions: These new insights may contribute to novel precision medicine approaches to treatment of CNS disorders.

Event-based imagers are an emerging class of sensors with significant demonstrated advantages relative to traditional cameras. Because they operate asynchronously and only transmit data from pixels that have changed, they have been shown to produce over 100x less data in sparse scenes relative to traditional focal plane arrays (FPAs). This leads directly to 100x lower latency allowing for highly dynamic phenomenology to be captured. Raytheon Technologies (RTX) will present an overview of their recent work on Event Based Camera (EBC) technology both on the DARPA sponsored FENCE program, as well as under internal research and development. This will include discussion of the RTX camera hardware design, including the ROIC and Processor, fabrication of the ROIC and Processor at separate foundries, the camera build, and test status. We will also cover as modelling and simulation of EBC applications using the Digital Twin to simulate camera responses to different targets and background scenarios.

IntroductionAcne pathogenesis is multifactorial, involving systemic factors including gut dysbiosis, hormones, and chronic inflammation. Probiotics, myoinositol, and plant-derived molecules may modulate acne by targeting these factors. The objective is to compare a synbiotic containing herbs against a myoinositol-based herbal supplement on how they influence acne, the gut microbiome, short chain fatty acids (SCFAs), and hormonal profiles.MethodsThis was an 8-week, randomized study involving 36 male and female patients aged 12 to 45 years with non-cystic acne. Subjects received either a synbiotic or a myoinositol-based herbal supplement (MBHS). Acne lesions were counted, stool samples were collected for gut microbiome and SCFA analyses, and hormone collections were performed at baseline, 4, and 8 weeks.ResultsSeveral gut bacteria increased by at least threefold at both week 4 and 8 in the synbiotic (Erysipelatoclostridium merdavium, Blautia argi, Faecalibacterium prausnitzii, Prevotella copri, Streptococcus sp001556435, Blautia sp900541955) and MBHS group (Megamonas funiformis, Ligilactobacillus ruminis, Prevotella ssp015074785, Prevotella copri, Gca-900199835 sp900176495). Acne lesion counts decreased significantly in both groups at week 4 (p < 0.0001) and week 8 (synbiotic, p < 0.0001; MBHS, p < 0.0001). There were significant and trending increases in stool and plasma SCFAs in both cohorts at week 4 and 8. After 8 weeks of MBHS, 17-OHP and androstenedione significantly decreased from 27.3 to 11.3 pg/ml (p = 0.001) and 94.9 to 68.0 pg/ml (p = 0.04), respectively.ConclusionBoth the synbiotic and MBHS improved gut health, augmented SCFAs, and reduced lesion counts in those with non-cystic acne. The MBHS may act by reducing hormone levels of 17-OHP and androstenedione.Clinical Trial Registrationwww.clinicaltrials.gov (NCT05919810).

Effective staff retention strategies are essential in the workplace to address employee turnover, improve work-life balance, and promote employee commitment to the organization. This study examined the impact of work overload on staff retention in public elementary schools in the Philippines, focusing on 148 non-teaching personnel. Employing a descriptive-quantitative research design, data were gathered through a researcher-structured questionnaire formatted in Google Forms and distribution was made online via the personal email addresses of non-teaching personnel. The data were analyzed using frequency, percentage, weighted mean, standard deviation, and Pearson correlation coefficient as statistical tools. The findings revealed that work overload is prevalent, with task demands, time pressure, and role ambiguity identified as critical stressors. However, despite the workload experienced by the non-teaching personnel they are still committed to stay with the organization. Time pressure, work hours, and role ambiguity showed significant correlations with job satisfaction and organizational commitment, while resource availability and task demand exhibited no direct influence on staff retention. These findings suggest that external factors and personal resilience may mitigate the impact of work overload. Based on the findings of the study, a strategy roadmap for improving staff retention is proposed. This study is essential for an organization seeking to improve staff retention and the long-term success created through workload distribution and clarifying roles in the workplace.

Multi-tenant cloud infrastructures have become the backbone of modern enterprise software delivery, yet they introduce persistent challenges in safeguarding data privacy, regulatory compliance, and tenant isolation. Traditional CI/CD pipelines, designed for centralized control, are ill-suited to environments demanding computational sovereignty and cross-tenant privacy. This paper presents a Federated DevSecOps framework that integrates federated learning concepts, Zero Trust principles, and GitOps workflows to create privacy-preserving, auditable CI/CD pipelines for Kubernetes-based cloud environments. The proposed model leverages homomorphic encryption, differential privacy, and secure multi-party computation to enable collaborative yet isolated DevOps workflows. Implemented and validated on multi-account AWS EKS architectures, our framework demonstrates a 73% reduction in cross-tenant security incidents, 60% improvement in compliance readiness, and minimal performance overhead (&lt;8%) compared to traditional centralized models. The findings establish Federated DevSecOps as a viable paradigm for secure, compliant, and scalable software delivery in regulated, multi-tenant cloud ecosystems.

ObjectiveTo assess the factors associated with the use of active surveillance (AS) in NCCN favourable intermediate‐risk (FIR) prostate cancer (PCa) patients who received the 17‐gene Genomic Prostate Score (GPS) assay.Material and MethodsContemporary data were collected from academic and large community group practices across the United States. Eligible patients had localized PCa classified as FIR per NCCN guidelines and received a GPS report between May 2017 and April 2019. Higher GPS results (scale: 0–100) were associated with a higher risk of adverse outcomes. The proportion of patients selecting AS was calculated with 95% confidence intervals. Uni‐and multivariable logistic regression analyses were performed to determine the association between AS selection and relevant covariates.ResultsThere were 324 eligible patients (Gleason Score 3 + 4, 79%; PSA 10–20 ng/ml, 19%; clinical stage T2b‐T2c, 2%; median percent positive cores, 16.7%; median GPS result, 26). The distribution of GPS results was 0–19 (23%), 20–40 (60%), and 41–100 (16%). Overall, 31% (95% CI 26%, 36%) selected AS: 58% (46%, 69%) with GPS 0–19, 27% (21%, 33%) with GPS 20–40, and 6% (1%, 16%) with GPS 41–100. In univariable models, the Gleason score, percent positive cores, PSA, and GPS results were significantly associated with AS selection. In a multivariable model, the percent positive cores and the GPS result remained significantly associated with AS selection. AS persistence was 91% (82%, 95%) at 12 months.ConclusionsThe GPS result and percent positive cores appear associated with AS use after controlling for relevant clinical variables in NCCN FIR prostate cancer patients.