GIST Scholar
Not a member yet
30271 research outputs found
Sort by
Deep Learning-Based AI Model for Brain Tumor Segmentation in Digital Pathology and Terahertz Imaging
No full textThis study presents a deep learning-based AI model for brain tumor segmentation in digital pathology images. Using a transgenic mouse model and H&E-stained images, we developed and trained the model with DEEP:PHI, employing U-Net and attention U-Net architectures. The AI model facilitates accurate cancer detection, contributing to terahertz imaging-based diagnostics and enhancing real-time surgical decision-making with minimal pathologist intervention. © 2025 The Author(s
Dynamic Gyrator-based Transient Analysis of Inductive Power Transfer Systems
No full textThis thesis introduces a novel dynamic gyrator model designed for the transient analysis of inductive power transfer (IPT) systems. By applying the dynamic phasor method to magnetically coupled IPT circuits, a dynamic gyrator is derived in the phasor domain. This dynamic gyrator extends the functionality of the conventional static gyrator by incorporating an imaginary gain that is effective not only in steady-state conditions but also during transient operation. One of the significant contributions of this research is the simplification of IPT systems with two LC resonant tanks, traditionally characterized as fourth-order systems, into second-order systems using the proposed model. This reduction drastically eases the complexities associated with analyzing and dynamically controlling IPT systems, particularly those with high-order compensation circuits. The proposed dynamic gyrator model is robust and universally applicable, offering compatibility with various IPT system configurations, regardless of the type of compensation circuits employed. It eliminates the need for complex and cumbersome equations, providing a streamlined approach to the transient analysis and design of IPT systems. Additionally, the model facilitates a deeper understanding of system behavior during transient states, enabling more effective control strategies and optimized designs. The validity and practicality of the dynamic gyrator model are demonstrated through both simulations and experimental verification. A series-series compensated IPT system, 520 W tuned at 50.3 kHz, was analyzed to test the model's performance through simulations an experiments. Results confirm that the proposed model accurately captures the transient and steady-state behaviors of the system while significantly simplifying its analysis and design process. This work represents a major step forward in IPT system research, providing a powerful analytical tool on wireless power transfer technologies.MasterAbstract I
List of Figures IV
List of Tables V
Chapter 1. Introduction 1
Chapter 2. Theoretical and Mathematical Development 4
2.1 Dynamic Gyrator 4
2.2 Dynamic Phasor-Based Gyrator Model of IPT 6
2.3 Dynamic Phasor Circuit of an IPT Model. 7
2.3.1 Inductor Phasor Transform 7
2.3.2 Capacitor Phasor Transform 7
2.3.3 Resistor Phasor Transform 7
2.4 Dynamic Gyrator-based IPT Analysis 9
Chapter 3. Simulation Verifications 15
3.1 Circuit Parameters 15
3.2 Step Input Response 16
3.3 Transient Output Voltage 17
3.4 Combined Step Input Response and Transient Output Voltage Response 19
3.5 DC Voltage Gain 20
Chapter 4. Experimental Verifications 23
4.1 Preparation of Experimental Kit 23
4.2 Equipment Utilized 25
4.3 Measurement of Transient Output Voltage 26
4.4 Measurement of DC Voltage Gain 27
Chapter 5. Results and Discussions 30
Conclusion 34
Summary 35
Publications 36
References 37
Acknowledgments 4
Synthesis of bacterial cell penetrating peptoids
No full textThe delivery of therapeutic agents into bacterial cells remains a significant challenge due to the impermeability of bacterial membranes. This study investigated the potential of peptoids, stable mimetics of cell- penetrating peptides (CPPs), for application in proka yotic systems, focusing on mitochondria-targeting peptoids (MTPs) and RXRtoid, a peptoid mimetic of the RXR4 peptide commonly used in therapeutic delivery. MTPs were hypothesized to exhibit favorable bacterial-penetrating properties due to their inherent CPP activity and structural parallels with Gram-negative bacterial membranes. While MTPs demonstrated high uptake and mitochondrial colocalization in mammalian cells, confocal microscpy revealed that these properties often resulted in bacterial membrane disruption rather than successful cytoplasmic penetration. Co-staining with DAPI and FM 4-64 confirmed that two bacterial cell penetrating peptoid (BPP) candidates localized to the bacterial cytosol, with MJJ- 1-121, a peptoid exhibiting low mammalian permeability—achieving effective bacterial internalization, whereas other peptoids predominantly caused membrane damage. The inclusion of RXRtoid further highlighted the potential of peptoid design for modulating cellular uptake and internalization, offering insights into the optimization of peptoids as delivery agents for bacterial applications.MasterList of Contents
Abstract ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ⅰ
List of Contents ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ⅱ
List of Tables ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ⅳ
List of Figures ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ⅳ
List of Schemes ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ⅴ
Ⅰ. Introduction∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 1
1.1. Cell penetrating peptide ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 1
1.1.1. Classification of CPPs ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 2
1.1.2. Proposed mechanism of CPP internalization∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 3
1.1.3. Applications of CPPs in Bacteria∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 3
1.2. Peptoids∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 4
Ⅱ. Experimental Procedures∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 1. General method∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 1. 1. Abrreviations∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 1. 2. Materials and reagents∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 2. Protected submonomer synthesis∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 2. 1. Mono-Mmt protected 1,4-diaminobutane∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 6
2. 3. Solid phase peptoid synthesis ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 7
2. 4. Cell culture and cytotoxicity test using MTS ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 8
2. 5. Fluorescence spectroscopy screening∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 9
2. 6. Bacteria confocal microscopy ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 9
2. 7. Mammalian cell – FACS, confocal microscopy∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 9
Ⅲ. Results and Discussions∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 11
3. 1. Design and synthesis of Cell-penetrating peptoid∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 11
3. 2. Cytotoxicity test∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 17
3. 3. Mammalian cell – FACS and live confocal microscopy∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 18
3. 4. Bacteria – FS screening and confocal microscopy ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 21
Ⅳ. Conclusion ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 26
Ⅴ. References∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 28
Supporting materials∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 31
Acknowledgements∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ ∙ 3
Residue-Free Fabrication of van der Waals Heterostructures of Two-Dimensional Materials
No full textIn this work, a unique methodology is presented that utilizes van der Waals (vdW) interactions to fabricate residue-free single flakes of two-dimensional (2D) materials, which are subsequently assembled into intricate heterostructures. The approach focuses on ensuring that the flakes are free from any residue that could affect their properties and performance. Evidence from atomic force microscopy and Raman spectroscopy confirms that the transferred hexagonal boron nitride (h-BN) and molybdenum disulfide (MoS2) flakes exhibit excellent flatness and strain-free characteristics, which are crucial for various applications in electronics and optoelectronics. Furthermore, the pick-up and release processes of a target flake are demonstrated using a residue-free stamp, controlled by the type of applied force at the interface, whether normal or shear. By carefully directing the movement of the residue-free stamp, successful assembly of h-BN/MoS2/h-BN heterostructures is achieved through both bottom-up and top-down stacking processes. Moreover, pre-assembled heterostructures were modularly stacked to create a more complex heterostructure, showcasing the versatility of the proposed methodology. This work not only enables to achieve residue-free single flakes of 2D materials, but also paves the way for improved fabrication techniques in vdW heterostructures.FALSEsciescopu
Future interconnect materials for highly integrated semiconductor devices
No full textAs semiconductor-based electronic technologies continue downscaling, there is an urgent need to overcome the limitations of interconnect architectures and materials that are driving an unsustainable increase in energy consumption and jeopardizing performance. In this Review, we investigate the primary causes of prolonged signal delays in interconnect systems, providing an overview of the development of key interconnect components: metals, difusion barriers and intermetal dielectrics. We defne the essential requirements and technological hurdles for next- generation materials to be industrialized within damascene processes, including topological semi-metals such as molybdenum phosphide (MoP) and 2D materials such as graphene and amorphous boron nitride (a-BN). Integrating new materials into advanced device systems ofers opportunities for the advancement of interconnect technologies and highly integrated semiconductor devices.FALSEforeig
MFTrans: A Multi-Resolution Fusion Transformer for Robust Tumor Segmentation in Whole Slide Images
No full textAccurate tumor segmentation in whole slide image (WSI) is essential for histopathological diagnosis and research, but the traditional manual analysis is labor-intensive and prone to variability. Furthermore, many artificial models focus on specific magnification images, limiting the detailed information available for segmentation. To address these challenges, we propose MFTrans, a novel multi-resolution fusion transformer with a CNN-based architecture designed for efficient tumor segmentation in WSI. Inspired by the diagnostic procedures of expert pathologists, MFTrans integrates both high- and low-magnification images, capturing detailed local features and broader contextual relationships through a dual-branch architecture. The model employs a global token transformer and cross-attention mechanism to fuse hierarchical features from dual branches to improve segmentation performance. We evaluate MFTrans on three real-world WSI datasets: Camelyon16, PAIP2019, and Catholic Uijeongbu St. Mary's hospital dataset, demonstrating its superior segmentation performance over state-of-the-art methods in balanced and imbalanced setups. These results highlight MFTrans's effectiveness in medical image analysis and its generalizability across different datasets, making it a robust tool for automated cancer diagnostics. Our code is available at https://github.com/aimed-gist/MFTrans. © 2025 IEEE
Lighting Model-Guided Initialization for Gradient Descent-Based Projector Compensation
No full text프로젝션 매핑은 프로젝터 입력(이미지)을 물리적 표면에 매핑하여 사용자가 지정한 시각적 콘텐츠를 표시하는 데 널리 사용되어 왔다. 그러나 평평하고 흰색인 이상적은 표면을 사용할 수 없는 경우 투사된 콘텐츠가 왜곡될 수 있다. 따라서 투사된 이미지 가 사용자가 제공한 목표 이미지와 일치하도록 보장하기 위해 프로젝터 보정 기법을 사용하여 프로젝터 입력을 조정하는 것이 필요하다. 프로젝터 보정을 위한 다양한 기법 들이제안되었지만,최근의 접근법은 프로젝션 매핑을 가상공간에서 시뮬레이션 가능한 과정으로 모델링하며, 여기서 광 전송 알고리즘이 가상 매핑을 정의하는 방법이다. 이 러한 렌더링 프레임워크에서 프로젝터 보정은 일반적으로 임의의 값으로 설정된 초기 추정값에서 시작하여 경사 하강법 기반 최적화기를 통해 프로젝터 입력을 반복적으로 저정함으로써 달성된다. 본 논문에서는 초기값을 임의로 선택하는 대신 보다 효과적으로 설정하는 방법을 연구한다. 본 논문의 주요 기여점으로, 조명 모델을 기반으로 시작값을 결정하는 새로운 초기화 방식을 제안한다. 이후 이 조명 모델 기반 초기화를 경사 하강법 기반 최적화에 통합하고, 이를 통해 비평면 및 유색 표면에 대해 향상된 결과를 생성함을 보여준다. |Projection mapping, which maps a projector input (i.e., an image) onto a physical surface, has been widely used to display user-specified visual content. However, the projected content can become distorted when an ideal surface—such as a flat, white one—is unavailable. As a result, adjusting the projector input using a projector com- pensation technique becomes necessary to ensure that the projected image matches the user-provided target image. While various techniques have been proposed for projector compensation, a recent approach models projection mapping as a simulatable process in virtual space, where a light transport algorithm defines the virtual mapping. In such rendering frameworks, projector compensation is typically achieved by iteratively ad- justing the projector input via a gradient descent-based optimizer, starting from an initial guess set to arbitrary values. In this paper, we investigate how to set the initial values more effectively rather than choosing them arbitrarily. As the main contribution of the paper, we propose a new initialization scheme that determines the starting values based on a lighting model. We then integrate this lighting model-guided initialization into gradient descent-based optimization and demonstrate that it produces improved results for non-planar and colored surfaces.Master1 Introduction 1
2 Related works 4
3 Background and Problem Specification 7
3.1 Background 7
3.2 Our problem 10
4 Lighting Model-Guided Initialization for Gradient Descent-Based Projector Compensation 11
5 Results and Discussion 17
5.1 Comparisons 17
5.2 Convergence comparisons of PCDR with and without our initialization 20
5.3 Analysis of reconstruction kernels G 21
5.4 Analysis of sample counts and computational overheads 21
5.5 Analysis of different initialization schemes 23
5.6 Limitations and Future Work 25
6 Conclusion 27
References 28
Acknowledgements 34
– v
A Study of the Flexibility Region of Virtual Power Plant considering the Uncertainty of Other Virtual Power Plants
No full textIn recent years, the increased numbers of distributed energy resources (DERs) have reduced the stability of distribution networks. To address this issue, Virtual Power Plant (VPP) technology has emerged and research has been conducted on aggregating various DERs. However, existing studies assume that VPPs exist independently within distribution networks. In reality, though, multiple VPPs can exist within the same system. Therefore, this paper proposes a method for estimating the flexibility region of a VPP that considers the uncertainty of other VPPs in the grid. As historical data on other VPPs is private, it is impossible to create a probability distribution. This paper therefore models the uncertainty using multi-source distributionally robust chance constraints (MS-DRJCC), based on a theoretical approach that allows modelling of uncertainty in the absence of historical data. The results show that curtailment is reduced compared to stochastic programming (SP), which schedules based on the available data distribution. This indicates that modelling with MS-DRJCC is effective in the absence of historical data.MasterI. INTRODUCTION 1
II. Virtual Power Plant Flexibility Region 4
2. 1. Flexibility region of each DER 4
2. 1. 1. Types of flexibility region model 4
2. 2. Multi-source distributionally robust optimization (MS-DRO) 8
2. 2. 1. Wasserstein ambiguity set 8
2. 2. 2. p-Wasserstein barycenter 9
2. 2. 3. Worst-cast expectation (MS-DRO) 10
2. 3. Virtual Power Plant model (MS-DRJCC) 11
2. 3. 1. VPP model with network security constraints 11
2. 3. 2. MS-DRJCC (tractable form) 12
2. 4. Solution of the MS-DRJCC 14
2. 4. 1. Constraint elimination 14
2. 4. 2. Solving problem (find flexibility point) 15
III. CASE STUDIES 16
3. 1. Numerical settings 16
3. 2. Analysis of the result 18
3. 3. Analysis of the optimization parameter 22
3. 4. Results of the flexibility region 22
IV. CONCLUSION 23
References 2
Corrigendum to “Microbiome-derived indole-3-lactic acid reduces amyloidopathy through aryl-hydrocarbon receptor activation” [Brain Behav. Immun. 122 (2024) 568–582, (S0889159124005749), (10.1016/j.bbi.2024.08.051)]
No full textThe authors regret that in the original published version of the article, in the “2. Materials and Methods” section, specifically under “2.2. Preparation of the microbial strains and metabolites,” the sentence was inaccurately stated as follows: “Each lyophilized bacteria powder was suspended in 1 mL PBS at concentrations of 1 × 10^9 CFU prior to administration. Then, each mouse was given 200 μL of microbial strains daily by oral gavage.” This should be corrected to: “Each lyophilized bacterial powder was suspended in PBS to achieve final concentrations of 1 × 10^9 CFU per 200 μL during the screening phase and 1 × 10^12 CFU per 200 μL during the validation phase, which were then administered daily to each mouse by oral gavage.” The authors would like to apologise for any inconvenience caused. © 2025FALSEsciescopu
