31825 research outputs found
Sort by
Light-induced selective speed alteration of magnetically rolled semiconductor particles
No full textThis article was originally published in iScience. The version of record is available at: https://doi.org/10.1016/j.isci.2025.114484
User License: Creative Commons Attribution – Non Commercial – No Derivs (CC BY-NC-ND 4.0) | Elsevier's open access license policyMicrorobot teams or swarms are promising candidates for many applications, such as micromanipulation, microsurgery, or targeted drug delivery. However, attaining individual control of the microrobots, which is a critical component to many of their applications, remains a significant technical challenge. We introduce a method to control the magnetic rolling speed of hematite semiconductor particles using localized UV light, attributed to light-induced changes in particle-substrate friction. Simulations and theoretical models support our experimental observations, showing how particle-substrate separation influences speed. Additionally, we demonstrate fixed patterning of microparticles via selective UV illumination at lower pH, demonstrating selective immobilization of microrobots, a conceptual step toward applications such as targeted drug delivery or patterned cell stimulation in future studies. Therefore, this work provides a novel approach for independent control of microrobot systems by modulating particle-substrate interactions with light.H.S. acknowledges the support of the Natural Sciences and Engineering Research Council of Canada (funding reference number RGPIN-201804418). Cette recherche a e´ te´ finance´ e par le Conseil de recherches en sciences naturelles et en ge´ nie du Canada (CRSNG) (nume´ ro de re´ fe´ rence RGPIN-2018-04418). This work was supported by the National Science Foundation under grant GCR 2219101, CPS 309 2234869, and the National Health Institute under grant 1R35GM147451. This project was also supported by a grant from the National Institute of General Medical Sciences – NIGMS (5P20GM109021-07) from the National Institutes of Health and the State of Delaware. The authors thank Fatma Ceren Kirmizitas for culturing and providing cells used in the experiments
Robust decentralized federated learning and its application in intelligent transportation systems
No full textShen, Chien-ChungFederated learning (FL) has become a foundational paradigm for privacy-preserving, collaborative machine learning across decentralized data sources. By enabling distributed devices to train models without sharing raw data, FL addresses critical privacy and regulatory concerns in domains such as healthcare, finance, edge computing, and the Internet of Things. However, practical deployment of FL faces two fundamental challenges: ensuring robustness and trust in adversarial or untrusted environments, and extending FL to real-world, large-scale domains such as intelligent transportation systems, which generate vast, distributed, and dynamic data. ☐ This dissertation, titled Robust Decentralized Federated Learning and Its Application in Intelligent Transportation Systems, is structured around two complementary research directions that collectively advance the state of the art in decentralized FL. The first research direction focuses on blockchain-based robust federated learning. In this part, I develop and analyze two frameworks, VBFL and NBFL, that leverage blockchain technology to address the challenges of trust, security, and robustness in collaborative learning. Traditional FL frameworks often rely on a central coordinator and lack model validation mechanisms, which introduces a single point of failure and makes the system vulnerable to adversarial attacks such as model poisoning and free-riding. To overcome these limitations, VBFL and NBFL eliminate central coordination by employing decentralized consensus protocols based on blockchain, such as proof-of-stake and validator voting. These protocols probabilistically favor the aggregation of legitimate, high-quality model updates while reducing the influence of malicious or low-quality contributions. Both frameworks incorporate robust validation and incentive mechanisms that reward honest participation and penalize adversarial behaviors. NBFL also enables personalized and communication-efficient learning through structured model pruning inspired by the Lottery Ticket Hypothesis, while demonstrating enhanced generalization and robustness in Non-IID environments compared to baselines. The effectiveness and accountability of VBFL and NBFL are demonstrated through extensive experiments on benchmark datasets, showing resilience and adaptability in adversarial and heterogeneous environments. ☐ The second research direction investigates decentralized federated learning in intelligent transportation systems (ITS). In this context, I propose and evaluate two novel FL frameworks, BFRT and NeighborFL, that enable real-time, adaptive, and collaborative learning among distributed edge devices in large-scale ITS networks. BFRT is designed for blockchain-compatible federated real-time traffic prediction, supporting online model updates and multi-step forecasting using streaming traffic data. NeighborFL introduces individualized aggregation strategies, allowing each device to form dynamic, context-aware groups of collaborators based on spatial proximity and real-time prediction error reduction. These approaches enable improved local traffic prediction performance, and support real-time streaming data scenarios with online model updates. NeighborFL facilitates adaptive neighbor selection, which is crucial for responsive and robust traffic prediction. The effectiveness and scalability of BFRT and NeighborFL are validated through real-world ITS datasets and simulations, demonstrating their strength over centralized and naive federated baselines in terms of prediction accuracy. ☐ These two research directions converge on the goal of building robust, scalable, and trustworthy federated learning in decentralized settings. The blockchain-based frameworks, VBFL and NBFL, ensure trust, auditability, and resilience in adversarial environments. Meanwhile, the decentralized, blockchain-compatible FL protocols, BFRT and NeighborFL, tackle real-time challenges in large-scale ITS. Together, they establish a solid foundation for secure, efficient, and generalizable federated learning, with applications extending beyond transportation to domains like smart cities, energy systems, and personalized education.University of Delaware, Department of Computer and Information SciencesPh.D
Substrate birefringence as a source of artifacts in spintronic THz emission spectra
No full textThis article was originally published in Optics Express. The version of record is available at: https://doi.org/10.1364/OE.586977
©2026 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement https://doi.org/10.1364/OA_License_v2#VOR-OATerahertz (THz) emission from spintronic THz emitters (STEs) has been extensively studied, both for its potential application in THz technologies and as a tool for probing spin dynamics, conductivity, and crystal anisotropy via time-domain THz spectroscopy. However, substrate birefringence can modify the emitted THz polarization, distorting the measured spectra and leading to potential misinterpretation of the results. While some aspects of substrate birefringence on THz emission have been studied in the time domain, its effects on THz spectra have not been discussed in detail. In the THz emission spectrum from STEs, we observe distinct spectral signatures that depend on sample orientation. To understand these signatures, we use STEs grown on c-cut sapphire and systematically vary the orientation of a (100) rutile TiO2 window in both pump-through and THz-through geometries to identify the birefringence-induced effects. These findings demonstrate that substrate birefringence plays a critical role in the emitted THz spectrum and must be carefully considered in the analysis of time-domain THz spectroscopy data. We also show that artifacts from smaller birefringence can have a more misleading effect on the THz spectra.U.S. National Science Foundation (NSF) (DMR-2011824); United States Department of Energy (DOE) (DE-SC-0012509)
Generation of reference CHO cell lines and insights into media additive use for influencing mAb production and quality
No full textLee, Kelvin H.Chinese hamster ovary (CHO) cell lines with favorable attributes are used to commercially biomanufacture hundreds of therapeutic proteins and antibodies. Companies heavily invest in private Research and Development sectors, enabling the latest innovations of cell line development and custom media/feed blends to ensure reliable and highly productive cell platforms for mAb production. In comparison, academic researchers often face limitations in acquiring high producing cell lines, as they lack the resources and proprietary knowledge to generate these types of cells. This disparity is a challenge to the comparability between academic and industrial discoveries, as drastic variability can exist between low and high producers. As industrial CHO cell lines reach better performance characteristics, this uncertainty in the comparability of studies conducted using low producing cell lines continues to grow. To circumvent this issue, it is imperative that academic researchers have access to a high producing reference CHO cell line which will expedite and broaden understanding of a specific cell line through a communal approach of research while increasing the translation of academic break throughs to industry. ☐ To address these issues, we establish two monoclonal antibodies (mAbs) expressing, CHO “reference cell lines” from different lineages as part of a university- industrial consortium (Advanced Mammalian Biomanufacturing Innovation Center, AMBIC) funded project. During the development of these reference cell lines, we characterize key growth and mAb production attributes while confirming key performance outcomes and technology transferability across two academic laboratories. As part of these studies, we generate fed-batch cultivation data from shake flask and scaled-down bioreactor processes with data presented confirming titers over 2 g/L in commercial medias. Furthermore, a chemically defined media formulation was developed and evaluated in parallel to the commercial media. ☐ Next, we describe a developed cell line development (CLD) platform approach for creating clones with varying productivities using the reference cell line. We then describe a method for purifying the mAb using protein A chromatography, followed by a reliable, consistent glycosylation analysis using mass spectrometry. The proposed workflow can be applied for a robust CLD process optimization to generate robust clones, enhance product expression, and improve product quality attributes. Lastly, using our developed reference cell line clone platform, we evaluate the potential titer enhancing and glycosylation impact of the histone deacetylase inhibitor, sodium butyrate, as a productivity enhancer in relationship to mAb production levels. ☐ Finally, we performed an in-depth factorial design study using the reference cell line to evaluate the influence of glycan modifying media additives such as 2-F- peracetyl fucose (2FP), galactose, and N-acetylmannosamine (manNAc) supplemented at various stages of cell growth. The resulting growth, production, and glycan distribution data help improve our understanding of media supplementation as a strategic tool for achieving desirable quality glycan profiles while also enabling faster process development times. ☐ Collectively, this work provides a universal, industrially relevant CHO culture platform, consisting of two “reference cell lines” to accelerate biomanufacturing innovation amongst the academic community. Furthermore, this work serves as a foundational study on methods and strategies for improving mAb production and influencing glycan profiles for the academic community to build upon for these universal reference cell lines.University of Delaware, Department of Chemical and Biomolecular EngineeringPh.D
Engineering neuroimmune regulation: biomaterial and nanotechnology platforms for neuropathology diagnosis and targeted immunomodulation
No full textThis article was originally published in Frontiers in Immunology. The version of record is available at: https://doi.org/10.3389/fimmu.2025.1677612
COPYRIGHT © 2026 Jackson, Porter and Oakes. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). https://creativecommons.org/licenses/by/4.0/ The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.The immune system is a potent and interwoven regulator of neuropathology in the central nervous system (CNS). For example, gliomas, neurodegenerative diseases, and autoimmune neuroinflammation all have diverse etiology and pathogenesis. Yet, there are numerous similarities in immunological dysfunction between these neuropathological conditions. Synthesizing this knowledge could inform advanced diagnostics and therapeutic platforms. For instance, detection and precision management of neuroinflammation or blood-brain barrier integrity would be broadly applicable. Understanding the targetable or controllable immune pathways of these disorders and engineering local and systemic interventions are essential for improving disease prognosis. This review integrates and contrasts the biology, detection, and treatment of the aforementioned neuropathologies. First, immunological underpinnings and recent discoveries on the role of innate and adaptive immune cells are discussed. Second, diagnostics that span molecular and nanotechnology-based platforms are detailed for enhanced imaging and screening of neuroinflammation. Last, immunomodulatory therapeutics and the precision of biomaterial platforms for neuroimmune regulation are examined. Within this section on therapeutics, both clinical and experimental approaches are stratified by CNS-localized or systemic mechanism-of-action. Overall, the integration of immunotherapies, biomaterials, advanced drug delivery platforms, and precision medicine will be critical in overcoming current treatment limitations for neuropathology that have few or no therapeutic options.The author(s) declared financial support was received for this work and/or its publication. This publication was made possible by the Delaware IDeA Network of Biomedical Research Excellence (INBRE) program, supported by a grant from the National Institute of General Medical Sciences -NIGMS (P20 GM103446) from the National Institutes of Health and the State of Delaware. RSO was supported, in part, by the United States Department of Veterans Affairs (VA) Career Development Award (CDA-2) IK2BX005061
2026 KIDS COUNT in Delaware Legislative Calendar
No full textPage-a-day style calendar containing data on Delaware children and their families. Calendar counts down to the end of legislative session.Delaware Division of Libraries, the Annie E. Casey Foundation, the University of Delaware and the State of Delaware
