http://repository.sungshin.ac.kr/handle/2025.oak/147 2026-05-19T14:07:27Z http://repository.sungshin.ac.kr/handle/2025.oak/8866 Title: Risk of anxiety disorders after epilepsy diagnosis: A nationwide retrospective cohort study Author(s): 정호현; 이승원; 강채윤; 최운비; 배영오 Abstract: Objective To evaluate the long-term psychiatric consequences of an epilepsy diagnosis on the incidence of anxiety disorders among patients in South Korea. Method This study utilized data from the Korean National Health Insurance Service spanning 2002–2013 to analyze longitudinal risks and contributing factors for anxiety disorders among 2109 patients with epilepsy compared to 21,090 matched controls. Results Patients with epilepsy demonstrated a significantly higher risk of developing anxiety disorders, with an incidence rate of 65.38 per 1000 person-years (95 % CI, 59.61–71.28) versus 33.13 per 1000 person-years (95 % CI, 31.89–34.38) for controls. The incidence rate ratio (IRR) was 1.97 (95 % CI, 1.79–2.18), indicating nearly double the risk relative to the control group. This risk was particularly pronounced in males and individuals under 60, underscoring age and male sex as key risk factors for anxiety post-epilepsy diagnosis. Conclusion The findings underscore the critical need for prompt psychological evaluations and interventions in the management of epilepsy. Addressing these psychological impacts early can significantly enhance outcomes and quality of life for patients, particularly among those at greater risk such as males under the age of 60. 2025-07-17T15:00:00Z http://repository.sungshin.ac.kr/handle/2025.oak/8856 Title: Machine Learning Driven Energy Transfer Prediction in Mn-Doped 2D Halide Perovskites Author(s): 조준상; Seonhong Min; Seyeon Park; Doyun Kim; Seong Wook Hwang Abstract: Manganese (Mn2+) metal doping into lead halide perovskite has offered an expanded palette to tailor the optoelectronic properties through engineering the energy transfer between the energy donor (exciton) and energy acceptor (Mn2+). However, it still remains a challenge to precisely modulate the two different emission centers to achieve the desired optoelectronic properties due to the presence of a complex interplay of competing excited-state dynamics, involving exciton recombination, forward and backward energy transfer, and dopant-state recombination. Here, we have developed a machine learning (ML) driven predictive framework to elucidate the complex energy transfer seen in Mn(x)-doped 2D PEA2Pb(Br1–yIy)4 perovskites. With integration of the ML predictive modelwith time-resolved spectroscopy, we identify and prioritize key important features (closely associated with excitonic properties such as bandgap, full width at half-maximum (FWHM), and wavelength), governing the degree of Mn sensitization. The ML-driven approach not only allows for precise prediction of desired optoelectronic properties but also uncovers nonlinear structure–function correlations that are often overlooked by conventional approaches. Our study, thus, provides new insight into the doped system, paving the way for the rational design of next 2025-07-16T15:00:00Z http://repository.sungshin.ac.kr/handle/2025.oak/8855 Title: Gene behaviors-based network enrichment analysis and its application to reveal immune disease pathways enriched with COVID-19 severity-specific gene networks Author(s): 박희원; Seiya Imoto; Satory Miyano Abstract: Motivation Gene network analysis is essential for understanding the complex mechanisms underlying diseases, which often involve disruptions in molecular networks rather than individual genes. Despite the availability of large-scale omics datasets and computational tools for gene network analysis, interpretation of the biological relevance of these extensive networks remains challenging.Results We propose a novel computational strategy, gene behaviors-based network enrichment analysis, which systematically identifies functional pathways enriched in phenotype-specific gene networks. Our novel method incorporates comprehensive network characteristics, i.e. gene expression levels, edge strengths, and structural patterns of edges, to rank genes based on activity and assess pathway enrichment, effectively identifying functional pathways enriched within these networks. Through simulation studies, our strategy demonstrated superior performance compared with that of existing methods in identifying enriched pathways. We applied this strategy to whole-blood RNA-seq data from 1102 COVID-19 samples provided by the Japan COVID-19 Task Force. The analysis revealed immune disease pathways enriched with COVID-19 severity-specific gene networks, including "Systemic lupus erythematosus" in asymptomatic and severe samples and "Infl 2025-06-30T15:00:00Z http://repository.sungshin.ac.kr/handle/2025.oak/8854 Title: CurriMAE: curriculum learning based masked autoencoders for multi-labeled pediatric thoracic disease classification Author(s): 강대성; 윤태영 Abstract: Masked autoencoders (MAE) have emerged as a powerful framework for self-supervised learning by reconstructing masked input data. However, determining the optimal masking ratio requires extensive experimentation, resulting in significant computational overhead. To address this challenge, we propose CurriMAE, a curriculum-based training approach that progressively increases the masking ratio during pretraining to balance task complexity and computational efficiency. In CurriMAE, the training process spans 800 epochs, with the masking ratio gradually increasing in four stages: 60% for the first 200 epochs, followed by 70%, 80%, and finally 90% in the last 200 epochs. This progressive masking approach, inspired by curriculum learning, allows the model to learn from simpler tasks before tackling more challenging ones. To ensure stable convergence, a cyclic cosine learning rate scheduler is employed, resetting every 200 epochs, effectively dividing the training process into four distinct stages. At the end of each stage, corresponding to one complete cycle of the learning rate schedule, a snapshot model is saved, resulting in four pretrained models. These snapshots are then fine-tuned to obtain the final classification results. We evaluate CurriMAE on multi-labeled pediatric thoracic disease classification, pretrainin 2025-07-08T15:00:00Z