https://scholar.korea.ac.kr/handle/2021.sw.korea/794 2026-04-05T15:57:39Z https://scholar.korea.ac.kr/handle/2021.sw.korea/270767 Title: Evaluating Patient Characteristics Influencing the Predictive Accuracy of Sleep Disorder Questionnaires Authors: Lee, Suin; Cawiding, Olive; Jo, Hyeontae; Kim, Jae Kyoung; Joo, Eun Yeon 2025-05-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/267713 Title: Allen-Cahn equation with matrix-valued anisotropic mobility in two-dimensional space Authors: Lee, Gyeonggyu; Lee, Seunggyu Abstract: The Allen-Cahn equation models antiphase domain coarsening in binary mixtures. To capture more complex and natural dynamics, we derive the anisotropic Allen-Cahn (aAC) equation, incorporating anisotropic mobility from an energy functional with a matrix-valued mobility. We present properties of the aAC equation, including energy dissipation and linear stability analysis, using this energy functional. The semi-discretized aAC equation is studied for properties like unique solvability, the maximum principle, and unconditional energy-gradient stability in a time-discrete manner. Additionally, numerical experiments demonstrate properties such as the maximum principle, unconditional energy stability, linear stability analysis, total area decreasing property, and transition layer profile. 2025-02-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/267958 Title: SymScore: Machine learning accuracy meets transparency in a symbolic regression-based clinical score generator Authors: Cawiding, Olive R.; Lee, Sieun; Jo, Hyeontae; Kim, Sungmoon; Suh, Sooyeon; Joo, Eun Yeon; Chung, Seockhoon; Kim, Jae Kyoung Abstract: Self-report questionnaires play a crucial role in healthcare for assessing disease risks, yet their extensive length can be burdensome for respondents, potentially compromising data quality. To address this, machine learning-based shortened questionnaires have been developed. While these questionnaires possess high levels of accuracy, their practical use in clinical settings is hindered by a lack of transparency and the need for specialized machine learning expertise. This makes their integration into clinical workflows challenging and also decreases trust among healthcare professionals who prefer interpretable tools for decision-making. To preserve both predictive accuracy and interpretability, this study introduces the Symbolic Regression-Based Clinical Score Generator (SymScore). SymScore produces score tables for shortened questionnaires, which enable clinicians to estimate the results that reflect those of the original questionnaires. SymScore generates the score tables by optimally grouping responses, assigning weights based on predictive importance, imposing necessary constraints, and fitting models via symbolic regression. We compared SymScore's performance with the machine learning-based shortened questionnaires MCQI-6 (n=310) and SLEEPS (n=4257), both renowned for their high accuracy in assessing sleep disorders. SymScore's questionnaire demonstrated comparable performance (MAE = 10.73, R2 = 0.77) to that of the MCQI-6 (MAE = 9.94, R2 = 0.82) and achieved AUROC values of 0.85-0.91 for various sleep disorders, closely matching those of SLEEPS (0.88-0.94). By generating accurate and interpretable score tables, SymScore ensures that healthcare professionals can easily explain and trust its results without specialized machine learning knowledge. Thus, SymScore advances explainable AI for healthcare by offering a user-friendly and resource-efficient alternative to machine learning-based questionnaires, supporting improved patient outcomes and workflow efficiency. © 2024 The Author(s) 2025-02-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/267901 Title: From homogeneity to heterogeneity: Refining stochastic simulations of gene regulation Authors: Chae, Seok Joo; Shin, Seolah; Lee, Kangmin; Lee, Seunggyu; Kim, Jae Kyoung Abstract: Cellular processes are intricately controlled through gene regulation, which is significantly influenced by intrinsic noise due to the small number of molecules involved. The Gillespie algorithm, a widely used stochastic simulation method, is pervasively employed to model these systems. However, this algorithm typically assumes that DNA is homogeneously distributed throughout the nucleus, which is not realistic. In this study, we evaluated whether stochastic simulations based on assumption of spatial homogeneity can accurately capture the dynamics of gene regulation. Our findings indicate that when transcription factors diffuse slowly, these simulations fail to accurately capture gene expression, highlighting the necessity to account for spatial heterogeneity. However, incorporating spatial heterogeneity considerably increases computational time. To address this, we explored various stochastic quasi-steady-state approximations (QSSAs) that simplify the model reduce simulation time. While both the stochastic total quasi-steady state approximation (stQSSA) and the stochastic low-state quasi-steady-state approximation (slQSSA) reduced simulation time, only the slQSSA provided an accurate model reduction. Our study underscores the importance of utilizing appropriate methods for efficient and accurate stochastic simulations of gene regulatory dynamics, especially when incorporating spatial heterogeneity. 2025-01-01T00:00:00Z