A5015725420- Biomedical Text Mining and Ontologies
- Clinical practice guidelines implementation
- Mobile Health and mHealth Applications
- Machine Learning in Healthcare
- Cardiac, Anesthesia and Surgical Outcomes
- Business Process Modeling and Analysis
- Innovation in Digital Healthcare Systems
- Blood Pressure and Hypertension Studies
- IoT and Edge/Fog Computing
- Epilepsy research and treatment
- Digital Mental Health Interventions
- Cardiovascular Health and Risk Factors
- Intensive Care Unit Cognitive Disorders
- Genomics and Rare Diseases
- Colorectal Cancer Screening and Detection
- Topic Modeling
- Hip and Femur Fractures
- Sepsis Diagnosis and Treatment
- Neonatal and Maternal Infections
- Cardiac Health and Mental Health
- Health Literacy and Information Accessibility
- Electronic Health Records Systems
- Respiratory Support and Mechanisms
- Diet and metabolism studies
- Phonocardiography and Auscultation Techniques
Seoul National University Bundang Hospital
2015-2025
Seoul National University
2022-2023
eHealth Initiative
2022
Background The length of stay (LOS) is an important indicator the efficiency hospital management. Reduction in number inpatient days results decreased risk infection and medication side effects, improvement quality treatment, increased profit with more efficient bed purpose this study was to determine which factors are associated stay, based on electronic health records, order manage efficiently. Materials methods Research subjects were retrieved from a database patients admitted tertiary...
Recently, Electronic Health Records (EHR) are increasingly being converted to Common Data Models (CDMs), a database schema designed provide standardized vocabularies facilitate collaborative observational research. To date, however, rare attempts exist leverage CDM data for healthcare process mining, technique derive process-related knowledge (e.g., model) from event logs. This paper presents method extract, construct, and analyze logs the Observational Medical Outcomes Partnership (OMOP)...
Background Neonatal sepsis is associated with most cases of mortalities and morbidities in the neonatal intensive care unit (NICU). Many studies have developed prediction models for early diagnosis bloodstream infections newborns, but there are limitations to data collection management because these based on high-resolution waveform data. Objective The aim this study was examine feasibility a model by using noninvasive vital sign machine learning technology. Methods We used electronic...
Abstract Background De-identification of clinical notes is essential to utilize the rich information in unstructured text data medical research. However, only limited work has been done removing personal from Korea. Methods Our study utilized a comprehensive dataset stored Note table OMOP Common Data Model at Seoul National University Bundang Hospital. This includes 11,181,617 radiology and 9,282,477 various other departments (non-radiology reports). From this, 0.1% reports (11,182) were...
Atherosclerotic cardiovascular disease (ASCVD) is a leading cause of death and morbidity worldwide. This randomized controlled, single-center, open-label trial tested the impact mobile health (mHealth) service tool optimized for ASCVD patient care. Patients with clinical were enrolled randomly assigned to intervention or control group. Participants in group provided smartphone application named HEART4U, while dedicated interface integrated into electronic healthcare record system was...
As patient communication, engagement, personal health data tracking, and up-to-date information became more efficient through mobile (mHealth), cardiovascular diseases (CVD) other that require behavioral improvements in daily life are now capable of being managed prevented effectively. However, to increase engagement mHealth, it is important for the initial design consider functionality usability factors accurately assess user demands during developmental process so app can be used...
Background Common data models (CDMs) help standardize electronic health record and facilitate outcome analysis for observational longitudinal research. An of pathology reports is required to establish fundamental information infrastructure data-driven colon cancer The Observational Medical Outcomes Partnership (OMOP) CDM used in distributed research networks clinical data; however, it requires conversion free text–based into the CDM’s format. There are few use cases representing CDM....
Personal health records (PHRs) are web based tools that help people to access and manage their personalized medical information. Although needs for PHR increasing, current serviced PHRs unsatisfactory researches on them remain limited. The purpose of this study is show the process developing Seoul National University Bundang Hospital (SNUBH)'s own system analyze consumer's use pattern after providing service.Task force team was organized decide service range set program. They made available...
Older adults are at an increased risk of postoperative morbidity. Numerous stratification tools exist, but effort and manpower required.This study aimed to develop a predictive model adverse outcomes in older patients following general surgery with open-source, patient-level prediction from the Observational Health Data Sciences Informatics for internal external validation.We used Medical Outcomes Partnership common data machine learning algorithms. The primary outcome was composite 90-day...
A clinical pathway (CP) is a tool for effectively managing care process. There are several research efforts on developing pathways (CPs) in the process mining domain. However, nature of data affects analysis results, and patient variability makes it challenging to develop CPs. Thus, crucial determine candidate processes that can be standardized as CPs before applying techniques. This paper proposed method assessing CP feasibility regarding complexity using order logs from electronic health...
Objectives To successfully introduce an Internet of Things (IoT) system in the hospital environment, this study aimed to identify issues that should be considered while implementing IoT based on a user demand survey and practical experiences environment monitoring systems. Methods In field test, two types systems (on-premises cloud) were used Department Laboratory Medicine tested for approximately 10 months from June 16, 2016 April 30, 2017. Information was collected regarding arose during...
This study investigated the effectiveness of using standardized vocabularies to generate epilepsy patient cohorts with local medical codes, SNOMED Clinical Terms (SNOMED CT), and International Classification Diseases tenth revision (ICD-10)/Korean Diseases-7 (KCD-7). We compared granularity between CT ICD-10 for by counting number concepts mapped one code. Next, we created selecting all patients who had at least code included in concept sets defined each vocabulary. set generated codes as...
Abstract Background : De-identification of clinical notes is essential to utilize the rich information in unstructured text data medical research. However, only limited work has been done removing personal from Korea. Objective In this study, we aimed perform de-identifying radiology reports Seoul National University, Bundang Hospital, a tertiary university hospital South Methods We used two de-identification strategies improve performance with and few annotated data. First, rule-based...
<sec> <title>BACKGROUND</title> Common data models (CDMs) help standardize electronic health record and facilitate outcome analysis for observational longitudinal research. An of pathology reports is required to establish fundamental information infrastructure data-driven colon cancer The Observational Medical Outcomes Partnership (OMOP) CDM used in distributed research networks clinical data; however, it requires conversion free text–based into the CDM’s format. There are few use cases...
<sec> <title>BACKGROUND</title> Neonatal sepsis is associated with most cases of mortalities and morbidities in the neonatal intensive care unit (NICU). Many studies have developed prediction models for early diagnosis bloodstream infections newborns, but there are limitations to data collection management because these based on high-resolution waveform data. </sec> <title>OBJECTIVE</title> The aim this study was examine feasibility a model by using noninvasive vital sign machine learning...
Background: Older adults are at an increased risk of postoperative morbidity. Numerous stratification tools exist, but effort and manpower required. We developed a predictive model general surgery adverse outcomes in older patients with open-source patient-level prediction from the Observational Health Data Sciences Informatics for internal external validation.Methods: used Medical Outcomes Partnership Common Model (CDM) machine learning algorithms. The primary outcome was composite 90 days...