GIST Scholar
Not a member yet
30271 research outputs found
Sort by
Skin Cancer Detection by Laser-Induced Breakdown Spectroscopy Employing Deep Neural Network Based Classification Methods
No full textLaser-induced breakdown spectroscopy (LIBS) with deep neural network (DNN) was conducted to classify four types of skin tissue: melanoma, normal, melanocytic nevus and basal cell carcinoma (BCC). Total 1830 LIBS spectra were acquired for 39 lesions excised from 26 patients. LIBS measurements were performed using a 532 nm wavelength laser (τ = 3 ns) and a 6-channel charge-coupled device (CCD) detector. Linear discriminant analysis (LDA) was performed to classify the four samples into six categories, confirming that they could be classified with high accuracy. (Melanoma vs. Normal skin : 100%, Melanoma vs. Melanocytic nevus : 98.5%, Melanoma vs. BCC : 98%, Normal skin vs. melanocytic nevus : 97%, Normal vs. BCC : 89%, Melanocytic nevus vs. BCC : 88%) The elements classifying each sample were identified by Jensen-Shannon divergence (JSD). It was confirmed that melanoma has a higher Mg concentration than normal and melanocytic nevus, which are benign tumors. DNN was then used to classify the four types and to implement a more accurate model, noise from the CCD was added to the raw data and the model was built using 10 times the data through data augmentation. The Micro f1 score for the four classifications was confirmed to be 0.942 and for the three classifications except BCC it was confirmed to be 0.945. The results confirmed that the LIBS classification method can diagnose skin cancer faster and more accurately than the existing method.DoctorAbstract (English)
List of contents
List of Tables
List of Figures
I. Introduction
1.1 Research background
1.2 Motivation
1.3 Principle of Laser-induced breakdown spectroscopy
1.4 Previous LIBS studies on skin cancer
1.5 Research objective
II. Experiment
2.1 Experimental setup
2.2 Sample preparation
2.3 LIBS data collection
III. Results and discussion
3.1 Surface morphology before and after LIBS measurement
3.2 Peak analysis
3.3 Binary-classification
3.4 Multi-classification
3.4.1 Original data classification
3.4.2 Original + augmented data classification
IV. Conclusions
Reference
Seamless Tutorial: Contextual State Transition Generation Based on Player Internal Knowledge
No full textIn the domain of game artificial intelligence, tutorial systems have seen limited advancement despite their critical role in onboarding players. Traditional tutorials often neglect individual learning differences, leading to ineffective instruction. This study proposes a personalized in-game tutorial generation framework that leverages procedural content generation and student modeling. The system integrates a Monte Carlo tree search-based state transition generator and a player modeling module to dynamically adapt tutorial content based on inferred internal knowledge. The approach is validated through large-scale user testing (N = 88) in a commercial-style Match-3 puzzle game environment. Results show that the personalized generator improves learning by up to 44.4% within short sessions. The findings highlight the potential of adaptive tutorials in accelerating skill acquisition and enhancing game experience through seamless, personalized learning contexts.FALSEsciescopu
Regenerative Capacity of Alveolar Type 2 Cells Is Proportionally Reduced Following Disease Progression in Idiopathic Pulmonary Fibrosis-Derived Organoid Cultures
No full textBackground: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease that culminates in respiratory failure and death due to irreversible scarring of the distal lung. While initially considered a chronic inflammatory disorder, the aberrant function of the alveolar epithelium is now acknowledged as playing a central role in the pathophysiology of IPF. This study aimed to investigate the regenerative capacity of alveolar type 2 (AT2) cells using IPF-derived alveolar organoids and to examine the effects of disease progression on this capacity. Methods: Lung tissues from three pneumothorax patients and six IPF patients (early and advanced stages) were obtained through video-assisted thoracoscopic surgery and lung transplantation. HTII-280+ cells were isolated from CD31-CD45-epithelial cell adhesion molecule (EpCAM)+ cells in the distal lungs of IPF and pneumothorax patients using fluorescence-activated cell sorting (FACS) and resuspended in 48-well plates to establish IPF-derived alveolar organoids. Immunostaining was used to verify the presence of AT2 cells. Results: FACS sorting yielded approximately 1% of AT2 cells in early IPF tissue, and the number decreased as the disease progressed, in contrast to 2.7% in pneumothorax. Additionally, the cultured organoids in the IPF groups were smaller and less numerous compared to those from pneumothorax patients. The colony forming efficiency decreased as the disease advanced. Immunostaining results showed that the IPF organoids expressed less surfactant protein C (SFTPC) compared to the pneumothorax group and contained keratin 5+ (KRT5+) cells. Conclusion: This study confirmed that the regenerative capacity of AT2 cells in IPF decreases as the disease progresses, with IPF-derived AT2 cells inherently exhibiting functional abnormalities and altered differentiation plasticity. Copyright © 2025 The Korean Academy of Tuberculosis and Respiratory Diseases.TRUEscopuskc
Examining the effects of the aggregation function and alternative eco-friendly refrigerants using graph-based molecular machine learning
No full textMolecular structures that can be readily represented by graphs comprising constituent atoms (nodes) and their chemical bonds (edges) can also be used as input data for well-known machine learning (ML) models that process this data, such as graph neural networks (GNNs). GNNs have shown a reasonable performance in the predicting properties of chemical systems. In typical applications of GNNs to chemistry-related fields, the main objective is to create an optimal molecular representation by aggregating atomic features and pooling features in the graph. In this study, we investigated two different approaches that can possibly generate better molecular representations. First, we created intermolecular edges to predict the photochemical properties of chromophore molecules in the solution. These intermolecular edges were constructed using atomic partial charges, inspired from the fact that electrostatic interaction is the main component of solute-solvent interaction. In the second approach, we investigated the effect of the aggregation and pooling functions. The results showed that intermolecular electrostatic edges based on ground state charges prevent the GNN model from generating more effective molecular representations. On the contrary, the model demonstrated better performance when the averaging and adding operations were employed in a hybrid manner for the aggregation and pooling functions.
Finding refrigerants is challenging because they should be non-toxic, non-flammable, energy-saving and thermodynamically stable. In addition, they should have low global warming potentials to mitigate global warming. Traditionally, a large number of molecules could be studied by quantum chemical calculations or experimental measurements. However, these methods are often challenging to apply on a large scale due to their time-consuming and resource-intensive nature. In this aspect, we propose a fully data-driven screening with machine learning models to find promising refrigerant candidates. We used two prediction methods: a less accurate but explicable feedforward neural network and a more accurate graph neural network (GNN). Using GNN, non-toxic molecules were determined using positive-unlabeled learning from the molecular dataset, which partially contains toxicity information. Out of 33k molecules, 7 molecules were recommended as future refrigerants. Molecular properties of these molecules and their uncertainties were also reported using data-driven methods. These contributed to the identification of alternative refrigerants, which remains a challenging topic today. We believe that our approach can be used to search for non-toxic emerging molecules in the future.MasterI. INTRODUCTION 1
II. COMPUTATIONAL BACKGROUNDS 6
2. 1. Dataset 6
2. 2. Machine learning 9
2.2.1. Feedforward neural network 9
2.2.2. Graph neural network 10
2.2.3. Electrostatic edge linked graph neural network 14
2.2.4. Positive-unlabeled learning 20
2.2.5. Uncertainty estimation 21
2.2.6. Computational details 22
III. EFFECTS OF THE AGGREGATION FUNCTION 23
3. 1. Result of the first approach 23
3. 2. Result of the second approach 29
IV. ALTERNATIVE ECO-FRIENDLY REFRIGERANTS 33
4. 1. Result of the prediction 33
4. 2. Result of the screening 38
4. 3. Analysis of the result 41
V. CONCLUSION 4
Study on Photocatalytic Porous Metal Oxide Heterostructures Prepared from Urea Network (UN)-Based Nanoporous Membranes as a Template
No full textThis study investigates the use of Urea Network (UN)-based nanoporous membranes and their derivatives, PM-NCF and NHF, as templates for the synthesis of porous metal oxide heterostructures designed for photocatalytic applications. The unique thermal and chemical stability of these templates enables the formation of high-surface-area metal oxides with 3D interconnected pore structures and high crystallinity, overcoming the limitations of conventional hard templates that are prone to degradation under harsh synthesis conditions. The UN-based templates allow for multi-cycle precursor infiltration, facilitating the development of composite heterostructures with enhanced structural complexity and catalytic functionality. The NHF variant provides an efficient method for synthesizing core-shell structures through direct silica shell formation during template removal. Additionally, the UN sol serves as an effective thin film coating binder, producing adherent films suitable for catalytic substrates. The synthesized materials exhibit exceptional photocatalytic performance in the reduction of nitroarenes to amines, owing to their optimized structural properties and stability. This approach highlights the versatility of UN-based membranes as robust platforms for the design of advanced catalytic materials. MS/MS 20231129MasterAbstract i
Contents ii
List of Figures v
List of Scheme vii
List of Tables viii
1. INTRODUCTION 1
1.1 Metal Oxide Heterostructures 1
1.2 Porous Metal Oxide 2
1.3 Method for Synthesizing Porous Metal Oxides 3
1.3.1 Sol-Gel Method 3
1.3.2 Hydrothermal Method 3
1.3.3 Dealloying Method 3
1.3.4 Templating Method 3
1.4 Template Method for Synthesizing Porous Metal Oxides 5
1.4.1 Hard Template 5
1.4.2 Soft Template 5
1.5 3D-Bicontinuous Nanoporous Polymeric Membrane as a Template for Porous Metal Oxide Synthesis 6
1.5.1 Bicontinuous Nanoporous Membrane by Organic Sol-Gel (OSG) Method 6
1.5.1.1 Organic Sol-Gel Synthesis: Gelation of Organic Networks 6
1.5.1.2 Formation of 3-D Continuous Nanopores from Covalent Networks/Porogen Mixture 8
1.6 Photocatalytic Reduction of Nitroarenes to Amines 10
1.7 Objectives 11
2. EXPERIMENTAL 13
2.1 Materials 13
2.2 Synthesis of Tetraphenylmethane Derivatives 14
2.2.1. Synthesis of Tetraphenylmethane (TPM) 14
2.2.2 Synthesis of Tetrakis(4-nitrophenyl) methane (TPM-NO2) 15
2.2.3 Synthesis of Tetrakis(4-aminophenyl) methane (TAPM) 16
iii
2.3 Preparation of Bicontinuous Nanoporous Membranes as a Template. 17
2.3.1 Synthesis of Urea Network (UN) Sols 17
2.3.2 Preparation of UN/PMMA Blend Film (PM-NCF) 17
2.3.3 Preparation of UN/Silica Blend Film (NHF) 17
2.4 Preparation of Metal Oxide Precursor Infiltration into Template 18
2.4.1 Infiltration of CuCl2 Precursor 18
2.4.2 Infiltration of Titanium Isopropoxide (TIIP) Precursor 18
2.5 Preparation of Thin Film Fabrication on Glass Substrate Using UN Sol as a Binder 19
2.6 Photocatalytic Performance Test 20
2.7 Characterization Techniques 21
2.7.1 UV-Visible Spectroscopy 21
2.7.2 X-ray Photoelectron Spectroscopy (XPS) 21
2.7.3 Ultraviolet Photoelectron Spectroscopy (UPS) 21
2.7.4 Scanning Electronic Microscopy (SEM) Spectroscopy 21
2.7.5 Transmission Electronic Microscopy (TEM) Spectroscopy 21
2.7.6 Gas Sorption Experiment 21
2.7.7 Inductively Coupled Plasma Optical Emission Spectrometer (ICP-OES) 22
2.7.8 X-ray Diffraction (XRD) Analysis 22
3. RESULTS AND DISCUSSIONS 23
3.1 Fabrication of 3D-Continuous Porous Membrane as a Template 23
3.2 Fabrication of Porous Metal Oxide Heterostructure Using Template Method 24
3.3 Structural and Chemical Characterization of Cu and TiO2-Based Heterostructures Derived from
PMNCF and NHF Templates 25
3.4 Morphology, Heterostructure, and Copper Distribution 28
3.5 Specific Surface Area 31
3.6 Band Gap Alignment and p-n Heterojunction Scheme 33
3.7 Photoreduction of Nitroarenes to Corresponding Amine by NaBH4 34
3.7.1 Catalyst Performance 34
3.7.2 Catalyst Stability 37
3.8 Comparison of the Catalytic Studies of Various Catalysts Reported in the Literature for the
Photoreduction of Nitro compounds by NaBH4 39
iv
4. CONCLUSIONS 42
5. APPENDIX 43
5.1 XRD Pattern for Catalyst Film 43
5.2 XPS Spectra for N 1s 44
5.3 SEM Image SEM Image of the Catalyst Cast onto Glass Substrate 45
5.4 EDS TEM for Cu2O-TiO2 and SiO2-Cu2O-TiO2 46
5.5 BET (Brunauer, Emmett and Teller) Analysis 47
5.6 UV-Vis Absorbance of TiO2, Cu2O-TiO2, SiO2-TiO2 and SiO2-Cu2O-TiO2. 48
5.7 Band Gap (Eg) Value from Uv-Vis Spectra by Tauc Plot Method 49
5.8 Valence Band Edge (EV) Value of TiO2 and Cu2O from Work Function (ϕ) Value by UPS Spectrum 50
6. REFERENCES 51
ACKNOWLEDGEMENTS 58
Data-driven Precipitation Nowcasting Using Satellite Imagery
No full textAccurate precipitation forecasting is crucial for early warnings of disasters, such as floods and landslides. Traditional forecasts rely on ground-based radar systems, which are space-constrained and have high maintenance costs. Consequently, most developing countries depend on a global numerical model with low resolution, instead of operating their own radar systems. To mitigate this gap, we propose the Neural Precipitation Model (NPM), which uses global-scale geostationary satellite imagery. NPM predicts precipitation for up to six hours, with an update every hour. We take three key channels to discriminate rain clouds as input: infrared radiation (at a wavelength of 10.5 µm), upper- (6.3 µm), and lower-(7.3 µm) level water vapor channels. Additionally, NPM introduces positional encoders to capture seasonal and temporal patterns, accounting for variations in precipitation. Our experimental results demonstrate that NPM can predict rainfall in real-time with a resolution of 2 km. Copyright © 2025, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved
A Computational Method for Screening Low-GWP Fluorinated Gases in Semiconductor Manufacturing
No full textThe widespread use of fluorinated gases in semiconductor manufacturing has raised significant environmental concerns due to their high global warming potential (GWP). To address this issue, we developed a systematic methodology for screening and evaluating low-GWP F-gases based on density functional theory (DFT) calculations. The infrared absorption cross-section (ACS) spectra and radiative efficiencies (RE) of candidate gases were predicted and systematically corrected using an empirical scaling factor derived from the correlation between calculated and experimentally measured RE values. This correction significantly improved the accuracy of GWP predictions. The methodology was successfully validated against representative F-gases, yielding GWP1oo estimates that closely align with reported values for high-GWP gases (GWP1oo=2,240), mid-GWP gases (GWP1oo=87), and low-GWP gases (below GWP1oo=10). Notably, several candidate gases, such as COF2, CF3OCFCF2, CF3C(O)CF(CF3)2, and C6F6, exhibited estimated GWP1oo values of 1.79, 2.69, 3.03, and 7.56, respectively, which are consistent with reported values. They were re-confirmed by the proposed method as promising low-GWP alternatives to conventional high-GWP etching and cleaning gases. By adopting a practical, accessible DFT methodology, this approach delivers reliable comparisons of GWP values among candidate gases and supports rapid, on-site GWP assessments without requiring specialized expertise. © 1988-2012 IEEE.FALSEsciescopu
Sustainable Removal of Aqueous Hg(II) by Zeolitic Imidazolate Framework-Derived Co/NC Using Optimized Thermal Desorption
No full textThe effectiveness of materials in aqueous contaminant treatment technologies by sorptive removal relies on their ability to be reused and their removal efficiency. Thermal desorption stands out as a promising method to improve the reusability of these materials. In this study, Zeolitic Imidazolate Framework-67 (ZIF-67) derived Cobalt N-Doped Carbon (Co/NC) and metal-impregnated (Ru, Pt, and Pd) Co/NC nanoparticles have been synthesized and tested for the effective removal of aqueous Hg(II) and its reusability by thermal desorption. Reduced Co/NC efficiently removed Hg(II), adsorbing 99.9% of aqueous Hg(II) in 2.5 min through pyridinic-N adsorption sites and Co0 reducing Hg(0) on the surface. The testing of various metals (Ru, Pt, and Pd) on the surface of Co/NC showed that Pd(4%)-Co/NC achieved the highest reactivity with a maximum adsorption capacity of 49.93 mgg-1 using the Langmuir model. Pd(8%)-Co/NC showed the highest adsorbed Hg(0) (79.1%) and fastest removal kinetics (135.52 g mg-1 min-1). The Pd(4%)-Co/NC catalyst retained its durability and stability, eliminating 99.9% of the aqueous Hg species throughout 10 consecutive cycles. The 80.11% and 85.4% of adsorbed Hg were recovered by thermal desorption at 500 °C on Pd(4%)-Co/NC and Co/NC surfaces, respectively. Pd(4%)-Co/NC displayed notable promise as a sustainable catalyst for Hg(II) reductive removal in wastewater treatment technologies, emphasizing its enduring effectiveness and reuse potential for practical engineering applications. © 2025 The Authors. Published by American Chemical Society.TRUEscopu