17315 research outputs found
Sort by
In vitro evaluation of bortezomib-loaded superparamagnetic iron oxide nanoparticles for multiple myeloma cells under guide-targeted variable magnetic fields
No full textMultiple myeloma (MM) is a type of hematological cancer. Bortezomib is an effective proteasome inhibitor used in its treatment. However, Bortezomib causes many side effects such as peripheral neuropathy and thrombocytopenia, which affect the quality of life in patients. Prolonged use of Bortezomib can lead to drug resistance in MM patients. Superparamagnetic iron oxide nanoparticles (SPIONs) are oxide nanoparticles that exhibit the phenomenon of superparamagnetism. When an external magnetic field is applied, they become magnetized to saturation levels, but they do not show any magnetic interaction upon its removal. In this study, a drug delivery system is proposed that can precisely target cancer cells, persist in circulation, exhibit cytotoxic effects even at lower concentrations, and minimize side effects. For this, Bortezomib is successfully loaded into SPIONs. Controlled drug release is achieved solely in the acidic environment of the vacuoles. This approach results in higher growth inhibition compared to the equivalent dose (2.5 nM) of free Bortezomib in RPMI 8226 cells. The magnetic drive system that has been developed generates irregular magnetic fields with variable fluxes ranging from 2 to 60 mT on the rotary table. Moreover, the use of the developed variable magnetic field device significantly enhances the growth-inhibitory effect on the cells, with Bim/Mcl-1 shifts consistent with an apoptosis-associated response. Targeted therapy at lower doses becomes possible by using the synthesized Bortezomib-loaded SPION (2.5 nM–10 µg) in combination with a magnetic field device. The findings of this study not only reveal an effective drug delivery system for MM cells but also hold promise for potential treatment and clinical applications for other cancer types
Semantically guided gradient matching for open-set domain generalization
No full textNeural networks assume training and test data share the same distribution and label space, but real-world violations degrade performance when domain and category shifts occur simultaneously. Open Set Domain Generalization addresses recognizing unseen classes in unseen domains. Current approaches using one-vs-all classifiers suffer from biased decision boundaries due to class imbalance, while meta-learning methods ignore semantic relationships between classes. This paper proposes a semantically-guided gradient matching framework extending dualistic meta-learning with joint domain-class matching by incorporating a semantic encoder for modeling continuous class relationships. The key innovation is weighted gradient matching using semantic similarity to guide decision boundary formation. Experiments on PACS dataset show this approach outperforms previous methods in open set scenarios while maintaining competitive closed-set generalization, resulting in more balanced decision boundaries
Additive content and style disentanglement for domain generalized semantic segmentation of optical remote sensing images
No full textSupervised learning methods assume training and test data are independently and identically distributed, commonly used in land cover semantic segmentation. However, this assumption doesn't hold in real-world scenarios, as test data stems from unseen domains, leading to domain shifts that degrade model performance. Domain Generalization (DG) techniques address this by focusing on learning domain-invariant features. This paper investigates DG for land cover semantic segmentation in remote sensing optical images by progressively separating content and style features, emphasizing domain-invariant ones. Unlike state-of-the-art methods, it explores extracting only domain-invariant features. The approach is validated on the FLAIR dataset, achieving superior performance compared to existing methods
Non-Ohmic behavior in (Bi1-xSbx)2Te3 by Joule heating
No full textA prerequisite to using the net spin polarization generated by a source-drain bias in three-dimensional topological insulators for spintronic applications is understanding how such a bias alters the transport properties of these materials. At low temperatures, quantum corrections can dominate the temperature dependence of the resistance. Although a dc bias does not break time-reversal symmetry and is therefore not expected to suppress quantum corrections, an increase of the electron temperature due to Joule heating can cause a suppression. This suppression at finite bias can lead to a non-Ohmic differential resistance in the three-dimensional topological insulator (Bi1-xSbx)2Te3, consisting of a zero-bias resistance peak (from electron-electron interactions) and a high-bias background (from weak antilocalization). We show that the bias voltage dependence of quantum corrections can be mapped to the temperature dependence, while the heating effect on the lattice temperature remains small. When searching for non-Ohmic effects due to novel phenomena in three-dimensional topological insulators, Joule heating should not be overlooked
Practical aspects of declarative languages: 27th International Symposium, PADL 2025, Denver, CO, USA, January 20–21, 2025, Proceedings
No full textThis book constitutes the refereed proceedings of the 27th International Symposium on Practical Aspects of Declarative Languages, PADL 2025, held in Denver, CO, USA, during January 20–21, 2025.
The 15 full papers included in this book were carefully reviewed and selected from 26 submissions. The accepted papers span a range of topics related to functional and logic programming, including some novel applications of Answer Set Programming, language extensions, runtime monitoring, program transformations, type-checking, and applications of declarative programming techniques to artificial intelligence and machine learning, among others
APA: domain generalization using frequency based augmentation
No full textDomain shift remains a major challenge in computer vision, where models trained on source domain(s) perform poorly on unseen target domains. One way to address this issue is by using data augmentation techniques, which generate synthetic images of the source images to mimic potential characteristics of the target domain. In this work, we introduce Amplitude-Phase Augmentation (APA), a frequency-domain augmentation technique. APA generates cross-domain images by multiplying the amplitude spectra of image pairs while preserving their original phase information. This approach exposes models to a broader spectrum of frequency-based variations, simulating diverse domain styles. We demonstrate that on two different datasets and with two different backbones, APA outperforms the baseline model, which is trained with original images, and also achieves competitive results compared to other models trained on the same backbones and datasets. Code available at https://github.com/sina-nuel/APA
Colonial wars and trade restrictions: fighting for exclusive trading rights
No full textThis paper develops a model of colonial wars and trade restrictions,
in which two metropolises compete for control over a colony’s trade policy. In
equilibrium, the metropolis that gains control can improve its terms of trade
by restricting its rival’s access to colonial trade. However, implementing these
restrictions often affects trade between the two metropolises themselves. Three
equilibrium outcomes can arise: (i) peace with free trade, when both metropolises value trade with each other above colonial trade; (ii) war with exclusive colonial trade, when both prefer colonial trade to mutual trade; or (iii) war with either free or exclusive colonial trade, depending on whether the victor prioritizes trade with its rival over colonial trade. The model captures central features of the mercantilist era—rapid expansion of overseas commerce, colonial conflicts, and restrictive trade policies—and provides insights into the eventual decline of mercantilism
Turkey seeks a vision fit for a multipolar world
Get PDFFor many years, Turkey has viewed foreign policy through
a domestic lens. Will its ambitions for strategic autonomy
in a multipolar world force its leaders to articulate
a broader vision
Responsible robot integration for inclusive and sustainable work
Get PDFThe growing integration of robots into workplace environments, propelled by the advances of Industry 5.0, is reshaping not only operational efficiency but also the social fabric of organizations. While automation offers significant productivity gains, it also risks exacerbating alienation, exclusion, and inequity if not responsibly managed. This conceptual paper examines how robotization impacts marginalized groups within organizations, including women, ethnic minorities, LGBTQ individuals, and employees with disabilities. Adopting an intersectional framework, the study connects responsible robot integration to the broader objectives of the United Nations’ Sustainable Development Goals (SDGs), particularly SDG 12 (Responsible Consumption and Production) and SDG 8 (Decent Work and Economic Growth). Synthesizing insights from human-robot interaction, diversity management, and responsible innovation literature, the paper proposes a conceptual model outlining how robot design choices and organizational practices mediate inclusive or exclusive outcomes. It concludes by offering a future research agenda and practical implications for managers, designers, and policymakers committed to advancing socially sustainable digital transformations. Through this approach, the paper contributes to the emerging discourse on ensuring that technological progress aligns with principles of equity, justice, and long-term societal well-being
Characterizations of matrix-valued asymmetric truncated toeplitz operators
No full textWe characterize matrix-valued asymmetric truncated Toeplitz operators (which are compressions of multiplication operators acting between two possibly different model spaces) by using compressed shifts, modified compressed shifts and shift invariance