SKSoeun Kim

Soeun Kim

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Soeun Kim has a total of 112 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Multimodal Multi‐task Model Integrating Scalp Imaging, Lipidomics, and Microbiome Data for Scalp Condition and Social Stigma: An Algorithm Development and Validation Study Across Multi‐ethnic Cohorts

    by Jiyeon Oh, Soeun Kim, Taelim Choi, Su Min Son, Kyeongmo Lim, Kyong‐Oh Shin, So Min Kang, Kyung‐Eun Lee, Kyungho Park, Dasom Jeon, et al.

  • Open Access

    Machine Learning Prediction Model for Dyslipidemia and Its Association With Atherothrombotic Events in 3 Independent Cohorts From South Korea, Japan, and the United Kingdom: Algorithm Development and Validation Study

    by Tae Hyeon Kim, Soeun Kim, Yerim Kim, Hayeon Lee, Seung Ha Hwang, So Young Yang, Lee Smith, André Hajek, Selin Woo, Dong Keon Yon

    Background: Dyslipidemia is a multifactorial and complex condition that warrants investigation through advanced analytical approaches such as machine learning (ML). Few previous ML studies predicting dyslipidemia have been validated across multiple international populations. Objective: This study aimed to develop an ML model to predict the 5-year incidence of dyslipidemia using routinely collected health examination data. To ensure generalizability, the model was externally validated in populations from South Korea, Japan, and the United Kingdom. Furthermore, the clinical relevance of the model-derived risk was evaluated by examining its association with atherosclerotic outcomes, including acute myocardial infarction and cerebral infarction. Methods: This study was conducted using 3 independent, large-scale, population-based cohorts. The discovery cohort from South Korea (n=471,650) was used for model training and internal validation, while 2 validation cohorts from Japan (validation A; n=7,255,685) and the United Kingdom (validation B; n=408,725) were used for external validation. We evaluated various ML-based models using 23 features extracted from regular health screening data to predict the new onset of dyslipidemia within 5 years. Shapley Additive Explanations values were calculated to assess feature importance. To ensure the robustness of the proposed ML model, we evaluated the risk of atherothrombotic events (acute myocardial infarction or cerebral infarction) based on the model probability (tertiles; T1, T2, and T3) using a Cox proportional hazards model. Results: In the discovery cohort, soft-voting ensemble learning with Light Gradient Boosting Machine and categorical boosting exhibited performance metrics of area under the receiver operating characteristic curve (AUROC) of 0.783, precision of 37.9%, and area under the precision-recall curve of 0.469. The model showed moderate discriminatory performance in the external validation cohorts (cohort A: AUROC 0.744; precision 27.2%; and cohort B: AUROC 0.687; precision 5.07%). Shapley Additive Explanations value analysis identified smoking, alcohol intake, and physical activity as the most important features for predicting dyslipidemia. Finally, a higher model probability (T3 vs reference) was pronounced with an increased risk of acute myocardial infarction (adjusted hazard ratio 2.34, 95% CI 1.84‐2.97) and cerebral infarction (adjusted hazard ratio 2.43, 95% CI 2.19‐2.71). Conclusions: This multinational study developed and validated an ML-based model using routine health checkup data to predict the 5-year risk of new-onset dyslipidemia, which was also associated with atherosclerotic events.

  • Open Access

    National trends of unmet healthcare needs and risk factors by household income level, 2010 to 2022: A Nationwide cross-sectional study in South Korea

    by Hyunjee Kim, Jaeyu Park, Jinyoung Jeong, Saiah Kim, Hayeon Lee, Hyeon Jin Kim, Yejun Son, Soeun Kim, Sooji Lee, Kyeongmin Lee, et al.

    Socio-economic status plays a critical role in shaping unmet healthcare needs, and the COVID-19 pandemic has further intensified these disparities; however, research to date remains insufficient. Therefore, this study aims to analyze unmet healthcare needs by household income using large-scale longitudinal data (2010–2022) including pre- and post- pandemic differences. This large-scale study (n = 2,628,584) utilized nationwide data from the Korea Community Health Survey (KCHS) conducted between 2010 and 2022, administered by the Korea Disease Control and Prevention Agency. The analysis employed complex, weighted sampling to examine trends in unmet healthcare needs, with a specific focus on changes during the COVID-19 pandemic. Weighted logistic regression models were used to calculate odds ratios and β differences (β diff ) between the pre-pandemic (2010–2019) and pandemic (2020–2022) periods. In total, 2,628,584 individuals participated in the KCHS from 2010 to 2022, comprising 1,454,129 males (55.3%) and 1,174,455 females (44.7%). Before the pandemic, there was a consistent decline in the prevalence of unmet healthcare needs. However, following the onset of the pandemic, unmet healthcare needs increased (β diff, low-level of household income: 1.66 [95% CI, 1.41–1.92]; mid-level of household income: 0.88 [95% CI, 0.77–0.99]; high-level of household income: 0.71 [95% CI, 0.57–0.85]). Overall, households with lower incomes exhibited higher levels of unmet healthcare needs compared to those with higher incomes (low household income: 14.8 [95% CI, 13.91–14.24]; high household income: 8.45 [95% CI, 8.34–8.55]). Additionally, the disparity in healthcare access due to income differences was more pronounced among older individuals, those with lower educational attainment, and those with lower subjective health status. Our analysis found that older adults in low-income households consistently faced higher rates of unmet healthcare needs. The reversal of a pre-pandemic trend toward reducing healthcare gaps highlights the urgent need for targeted interventions to address socio-economic disparities.


  • Top Cited Papers
    Open Access

    Global Incidence and Prevalence of Trigeminal Neuralgia, 1945–2024: A Systematic Review and Meta-Regression Analysis

    by Yi Deun Jeong, Yeona Jo, Yejun Son, Dong Kyu Kim, Tae Hyeon Kim, Jaehyeong Cho, Soeun Kim, Jiseung Kang, Lee Smith, Ho Geol Woo, et al.

    Background and Purpose Although trigeminal neuralgia (TN) imposes a substantial burden on quality of life and mental health, epidemiological estimates of its incidence and prevalence remain limited. Thus, we aimed to comprehensively assess the epidemiological and clinical features of TN from 1945 to 2024. Methods A comprehensive literature search was performed using PubMed/MEDLINE, Embase, CINAHL, and Google Scholar databases up to January 22, 2025, with search terms related to "trigeminal neuralgia. " TN was defined according to the International Classification of Headache Disorders (ICHD)-1, Rushton and Olafson criteria, and International Classification of Diseases (ICD) codes prior to 2004, which primarily relied on clinical features for diagnosis, whereas the ICHD-2, ICHD-3, Read codes, and ICD codes after 2004 were grouped together, as these criteria incorporated imaging modalities into the diagnostic process. A random-effects model was applied to calculate pooled estimates with 95% confidence intervals (CIs) for the incidence, prevalence and lifetime prevalence of TN. In addition, meta-regression models were also fitted using inverse-variance weighting. Results A total of 17 eligible studies, comprising over 170 million participants and 109,070 patients with TN, were included in the analysis. Global pooled incidence of TN was estimated at 25.33 cases per 100,000 person-years (95% CI, 11.87-54.02), while the global pooled annual prevalence was 45.38 cases per 100,000 inhabitants (15.41-133.61), and the global pooled lifetime prevalence was 108.43 cases per 100,000 inhabitants (30.54-384.18). Incidence and prevalence estimates tended to be higher in females than in males, more often right-sided, and more frequent in the maxillary and mandibular divisions than in the ophthalmic division; however, the certainty of these subgroup differences was low. Lastly, the incorporation of imaging modalities into diagnostic criteria may have contributed to the increased prevalence of TN. Conclusions This study highlights substantial global variations in TN incidence and prevalence, with how the evolving diagnostic criteria affects the epidemiological features of TN.


  • Temporal trends in acute kidney injury-related mortality across 43 countries, 1996–2021, with projections up to 2050: a global time series analysis and modelling study

    by Hyeon Seok Hwang, Soeun Kim, Soo-Young Yoon, Hayeon Lee, Hyeon Jin Kim, Selin Woo, Lee Smith, Louis Jacob, Jinseok Lee, Dong Keon Yon

    Acute kidney injury (AKI) is a major global public health concern. However, a major challenge in addressing the AKI burden is the lack of global data on AKI-related mortality and its predictions, leaving significant limitations in understanding its trends over time. Therefore, we aimed to estimate AKI-related mortality rate trends and forecast future deaths. We evaluated the temporal trends in age-standardized AKI-related mortality from 1996 to 2021 across 43 countries using the World Health Organization Mortality Database, with future projections through to 2050. Temporal trends were assessed based on age-standardized mortality rates, and future projections up to 2050 were calculated using a predictive model that considered attributable risk factors from the Global Burden of Disease Study 2021. Age-standardized AKI-related mortality rate per 1,000,000 people remained stable from 1996 to 2021 (10.47 [95% confidence interval (CI), 8.84-12.11] to 9.94 [95% CI, 8.32-11.57]). Although age-standardized mortality rates were lower in high-income countries (HICs) compared to low- and middle-income countries (LMICs), HICs exhibited a modest but statistically significant increasing trend (from 5.83 per 1,000,000 people [95% CI, 4.21-7.46] to 7.30 [95% CI, 5.66-8.95]), whereas LMICs showed a declining trend (from 19.66 [95% CI, 16.78-22.53] to 15.33 [95% CI, 12.37-18.29]). Projections indicate that mortality will rise to 11.36 per 1,000,000 population (95% CI, 10.65-12.07) by 2050, primarily attributable to population aging. This global time-series modeling study highlights rising AKI-related mortality in HICs and/or aging populations. These findings underscore the need for targeted interventions to mitigate future AKI-related deaths.


  • Open Access

    National trends in the prevalence of hypertension and diabetes stratified by sleep duration in South Korea, 2009–2022: A nationwide cross-sectional study: Trends and association of sleep duration in hypertension and diabetes

    by Jinyoung Jeong, Hyunjee Kim, Jaeyu Park, Hyeon Jin Kim, Hyesu Jo, Sooji Lee, Kyeongmin Lee, Seoyoung Park, Yesol Yim, Ho Geol Woo, et al.

    The increase in sleep-related issues is associated with various health problems, particularly hypertension and diabetes. Therefore, this study aims to address these gaps in the literature by analyzing the relationship between sleep duration and the prevalence of hypertension and diabetes. This study analyzed the association between hypertension, diabetes, and sleep duration using the Korea Community Health Survey database from 2009 to 2022. The study used weighted linear regression models to identify trends in diseases and sleep duration. Additionally, this analysis employed weighted logistic regression to examine associations between diseases and covariates, presented as weighted odds ratio (wOR) and 95% confidence intervals (CIs). Our analysis included 2,903,887 individuals aged ≥19 years. The prevalence of individuals with both hypertension and diabetes increased before (β, 0.87 [95% CI, 0.83–0.90]) and during the COVID-19 pandemic (β, 0.64 [0.58–0.70]). This trend was particularly notable among individuals who sleep less than 6 hours, with an increase identified both before (β, 1.28 [1.16–1.40]) and during the pandemic (β, 1.21 [1.03–1.39]). The higher risk of having both conditions was found in individuals who slept 8 hours (wOR, 1.19 [1.17–1.22]) compared to those who slept 7 to 8 hours. Our study is the first to analyze the association between sleep duration and hypertension and diabetes by utilizing a large-scale dataset and conducting a comparative analysis between the pre-COVID-19 and COVID-19 periods. This study shows that inadequate sleep is associated with an increased risk of hypertension and diabetes.


  • Open Access

    Global Prevalence of Functional Abdominal Pain Disorder among Children and Adolescents According to the Rome III and IV Criteria: A Systematic Review and Meta-Analysis

    by Yi Deun Jeong, Soeun Kim, Yejun Son, Tae Hyeon Kim, Lee Smith, André Hajek, Jiseung Kang, Jong Woo Hahn, Dong Keon Yon

    Yi Deun Jeong, Soeun Kim, Yejun Son, Tae Hyeon Kim, Lee Smith, André Hajek, Jiseung Kang, Jong Woo Hahn, Dong Keon Yon. Gut and Liver -0001;0:. https://doi.org/10.5009/gnl250225


  • Top Cited Papers

    Burden of Cardiovascular Outcomes After SARS-CoV-2 Infection in South Korea and Japan: A Binational Population-Based Cohort Study

    by Sooji Lee, Seung Ha Hwang, Seoyoung Park, Yejun Son, Soeun Kim, Hyeon Jin Kim, Jaeyu Park, Hyesu Jo, Kyeongmin Lee, Jiyeon Oh, et al.

    BACKGROUND: Despite the significant global impact of the COVID-19 pandemic, limited studies have investigated the long-term cardiovascular sequelae of SARS-CoV-2 infection, particularly among Asian populations. This large-scale, population-based binational cohort study with long-term follow-up aimed to investigate the association between SARS-CoV-2 infection and the risk of cardiovascular events. METHODS: We used binational, large-scale, and population-based cohorts, including a Korean nationwide cohort (K-CONV-N [Korea Disease Control and Prevention Agency-COVID-19-National Health Insurance Service cohort]; discovery cohort; n=18 989 129) and a Japanese nationwide cohort (Japan Medical Data Center; validation cohort; n=12 218 680). Individuals aged 20 years or older were included from January 1, 2020, to December 31, 2022. We assessed the long-term risk of incident cardiovascular outcomes after SARS-CoV-2 infection. The primary outcome was the risk of cardiovascular diseases based on International Classification of Diseases, Tenth Revision code diagnosis. After propensity score–based overlap weighting, Cox proportional hazard models were used to estimate adjusted hazard ratios for cardiovascular outcomes. We assessed the time attenuation effect of cardiovascular outcomes after SARS-CoV-2 infection. Multiple subgroup analyses were conducted by 16 cardiovascular outcomes, COVID-19 severity, vaccination, and SARS-CoV-2 strain. RESULTS: In the overlap-weighted discovery cohort, 7 960 357 individuals were included (mean age, 48.52 years [SD, 9.33]; men, 4 283 878 [53.82%]). SARS-CoV-2 infection was associated with a long-term increased risk of overall cardiovascular outcomes (adjusted hazard ratio, 1.62 [95% CI, 1.60–1.64]), particularly ischemic heart disease (1.81 [95% CI, 1.77–1.84]), heart failure (1.79 [95% CI, 1.73–1.85]), cerebrovascular disorders (1.65 [95% CI, 1.60–1.69]), major adverse cardiovascular events (1.65 [95% CI, 1.60–1.70]), inflammatory heart diseases (1.53 [95% CI, 1.31–1.80]), dysrhythmia (1.44 [95% CI, 1.42–1.46]), and thrombotic disorders (1.42 [95% CI, 1.35–1.48]). The increased risk persisted up to 18 months, with the highest association observed for 1 to 6 months after infection. The risk of cardiovascular diseases was pronounced with COVID-19 severity; however, it decreased with the administration of complete vaccination and subsequent booster doses. A similar risk of cardiovascular outcomes existed across every SARS-CoV-2 era (pre-delta, delta, and omicron). Similar patterns were observed in the validation cohort. The absolute risk of cardiovascular disease events after SARS-CoV-2 infection remained remarkably low (2.12% versus 1.31% in the noninfected population), particularly stroke (0.24% versus 0.13%) and ischemic heart disease (0.73% versus 0.39%). CONCLUSIONS: This binational study observed associations between SARS-CoV-2 infection and cardiovascular events during extended follow-up across viral eras. Complete vaccination was linked to lower cardiovascular events. However, the absolute risk of cardiovascular disease events after SARS-CoV-2 infection remained remarkably low, particularly for stroke and ischemic heart disease. Although these findings suggest ongoing vigilance and preventive measures remain crucial, they should be interpreted within the context of these low absolute risks when considering long-term cardiovascular complications.


  • Top Cited Papers
    Open Access

    Global burden of asthma among children and adolescents with projections to 2050: a comprehensive review and forecasted modeling study

    by Tae Hyeon Kim, Hyunjee Kim, Jiyeon Oh, Soeun Kim, Michael Miligkos, Dong Keon Yon, Nikolaos G Papadopoulos

    Understanding pediatric asthma is crucial to its effective diagnosis and intervention, as it may alleviate the adulthood disease burden. This epidemiological review describes the prevalence of asthma among individuals under 20 years of age by categorizing them into 3 age groups: 1–4, 5–9, and 10–19 years. Estimates were obtained from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021, which covered the prevalence of asthma from 1990 to 2021 across 21 GBD regions with 95% uncertainty intervals (UIs). We also projected the prevalence of pediatric asthma in 2050 by using a logistic regression predictive model from the existing literature and incorporating body mass index as a covariate with fixed coefficients over time. Overall, a continuous decline in asthma prevalence rates among children and adolescents was observed from 1990 to 2021, with higher rates in males and a peak prevalence rate in the 5–9 years group. Central Europe showed significantly increased prevalence rates compared to those of other regions. Our projection suggests that the prevalence rate of pediatric asthma will decline to approximately 2,608.05 per 100,000 population by 2050 (95% UI, 1,632.94–3,868.26), representing a 39.5% decrease from the 2021 figures. Despite these trends, asthma remains a substantial health burden for children and adolescents that may persist into adulthood. Therefore, proactive diagnosis and intervention are essential to mitigating the associated disease burden.


  • Top Cited Papers

    Live zoster vaccination and cardiovascular outcomes: a nationwide, South Korean study

    by Sooji Lee, Kyeongmin Lee, Jiyeon Oh, Hyeon Jin Kim, Yejun Son, Soeun Kim, Jaeyu Park, Jiseung Kang, Damiano Pizzol, Jinseok Lee, et al.

    Background and Aims: Despite the potential association between herpes zoster infection and cardiovascular events, limited studies have investigated the relationship between live zoster vaccination and cardiovascular outcomes. This large-scale, population-based cohort study with a long-term follow-up aimed to investigate the association between live zoster vaccination and the risk of various cardiovascular events. Methods: Data on comprehensive information of individuals aged ≥50 years from South Korea (n = 2 207 784) were included from 1 January 2012, to 31 December 2021. National insurance information from the Korea Health Insurance Review and Assessment Service, the national health examination results from the Korean National Health Insurance Service, and the live zoster vaccination data from the Korea Disease Control and Prevention Agency were merged. The risk of incident cardiovascular outcomes after live zoster vaccination was assessed compared with unvaccinated individuals. The primary outcome was the risk of cardiovascular diseases based on International Classification of Diseases, Tenth Revision code diagnosis. In propensity score–based overlap weighted cohorts, Cox proportional hazard models were used to estimate hazard ratios (HRs) for overall and specific cardiovascular outcomes, while calculating restricted mean survival time (RMST) for each outcome. The observation period was from 1 January 2012, to 31 January 2024. Multiple stratification analyses were performed. Results: After applying propensity score–based overlap weighting, 1 271 922 individuals were included [mean age, 61.3 years (standard deviation, 3.4); 548 986 (43.2%) male; median follow-up time, 6.0 years] in overlap-weighted cohort. Live zoster vaccination was associated with lower risks of overall cardiovascular events [HR 0.77, 95% confidence interval (CI) 0.76–0.78], particularly major adverse cardiovascular events [0.74 (0.71–0.77)], heart failure [0.74 (0.70–0.77)], cerebrovascular disorders [0.76 (0.74–0.78)], ischaemic heart disease [0.78 (0.76–0.80)], thrombotic disorders [0.78 (0.74–0.83)], and dysrhythmia [0.79 (0.77–0.81)]. The RMST difference for overall cardiovascular events following live zoster vaccination was 95.14 days per decade (95% CI 94.99–95.30). The protective association persisted up to 8 years, with the greatest reduction observed 2–3 years post-vaccination. The decrease in cardiovascular disease risk was more pronounced among males, individuals aged <60 years, those with unhealthy lifestyle habits, and those from low-income households and rural residents. Conclusions: These findings suggest that live zoster vaccination may be beneficial as a public health strategy with potential implications for cardiovascular disease burden in the general population. This strategy may help address health disparities and mortality linked to cardiovascular complications.