Korea University · Medicine
Professor Jaehun Jung's research lab specializes in health informatics, temporal knowledge reasoning, and large language model optimization. The lab focuses on leveraging knowledge graphs—particularly temporal and dynamic variants—for improved medical decision-making and clinical reasoning, with applications in public health outcomes such as post-COVID complications and statin therapy in dialysis patients. A key research direction involves enhancing the reasoning capabilities of pre-trained language models through structured explanation generation and logical consistency, using techniques like Maieutic Prompting. The lab also explores efficient, context-aware knowledge retrieval in dialogue systems via attention-driven path traversal over knowledge graphs.
Figures are computed from collected data and may differ slightly.
This retrospective cohort study examines the incidence of acute myocardial infarction and ischemic stroke after COVID-19 infection among vaccinated vs unvaccinated adults in Korea.
Static knowledge graphs (KGs), despite their wide usage in relational reasoning and downstream tasks, fall short of realistic modeling of knowledge and facts that are only temporarily valid. Compared to static knowledge graphs, temporal knowledge graphs (TKGs) inherently reflect the transient nature of real-world knowledge. Naturally, automatic TKG completion has drawn much research interests for a more realistic modeling of relational reasoning. However, most of the existing models for TKG comp
The formation of herd immunity through vaccination is a key point in overcoming the coronavirus disease 2019 (COVID-19) pandemic. To acquire herd immunity, a high vaccination rate is required, which is necessary to instill confidence in the public regarding the effectiveness and safety of the vaccine. In the real-world setting, thorough preparation of components, such as priority setting, vaccine delivery, logistics, and side-effect monitoring is necessary to overcome vaccine hesitancy. Each cou
Pre-trained language models (LMs) struggle with consistent reasoning; recently, prompting LMs to generate explanations that self-guide the inference has emerged as a promising direction to amend this. However, these approaches are fundamentally bounded by the correctness of explanations, which themselves are often noisy and inconsistent. In this work, we develop Maieutic Prompting, which aims to infer a correct answer to a question even from the unreliable generations of LM. Maieutic Prompting i
Background Recommendations have not yet been established for statin therapy in patients on maintenance dialysis. In this study, we aimed to evaluate the effects of statin therapy on all-cause mortality in patients undergoing maintenance hemodialysis. Methods and Results This retrospective cohort study analyzed data from adults, aged ≥30 years, who were on maintenance hemodialysis for end-stage renal disease. Data on statin use, along with other clinical information between 2007 and 2017, were ex
<i>Background and Objectives:</i> Recently, the prevalence of metabolic syndrome in Korea has increased rapidly. Current knowledge reflects the importance of dietary control in relation to the metabolic syndrome. The objective of this study was to evaluate the influence of skipping breakfast on the metabolic syndrome. <i>Materials and Methods:</i> We conducted a cross-sectional study using data from the Korea National Health and Nutrition Examination Survey 2017 for the second year. A total of 3
Surveillance remains an important tool for timely outbreak detection and response. Many countries, including Korea, have established national infectious disease surveillance systems with clinical notification. We aimed to evaluate the National Health Insurance Claims-based Surveillance (NHICS) compared to conventional passive report-based National Infectious Diseases Surveillance (NIDS). Reported to claimed cases ratios (R/C ratio) were evaluated from monthly notifiable disease cases captured by
Infectious diseases spread rapidly, and epidemiological surveys are vital to detect high-risk transmitters and reduce transmission rates. To enhance efficiency and reduce the burden on epidemiologists, an automatic tool to assist with epidemiological surveys is necessary. This study aims to develop an automatic epidemiological survey to predict the influence of COVID-19-infected patients on future additional infections. To achieve this, the study utilized a dataset containing interaction informa
Our results showed that in patients with AF, resuming anticoagulants and antiplatelets after ICH improved the STEs and SHEs. Further, NOAC had additional benefits as compared to warfarin.
The number of reported varicella cases is continuously increasing in Korea; however, associated medical utilization is declining. The ratio between varicella insurance claims and reports of passive infectious disease surveillance has gradually increased to > 80% since the second half of 2017. The recent increase in reported varicella cases is influenced by improved reporting. We calculated the varicella incidence and cumulative incidence in each birth cohort according to age. The cumulative inci
A 14-day quarantine is implemented in many countries in response to the coronavirus disease pandemic. Korea implemented a mandatory quarantine for those who had close contact with infected patients and those returning from abroad. The present study explored the implications of mandatory coronavirus disease 2019 testing before releasing individuals from the 14-day quarantine in Incheon, Korea. From February 11 to July 5, 2020, 19,296 people were self-quarantined, and 56 (0.3%) of them were confir
Open papers in the app to read, cite, and organize with AI.
