성균관대학교 · 의학
Won Chul 교수의 연구실은 응급의료 시스템의 효율성 향상과 환자 안전 확보를 핵심 목표로 삼고 있습니다. 특히 응급실 혼잡도 분석, 실시간 모니터링 시스템 개발, 그리고 인공지능 기반 예측 모델링을 통해 응급실 운영의 정밀화와 임상적 활용가능성을 추구합니다. 고도화된 데이터 기반 의사결정 지원 시스템과 다학제적 간병 모델 개발을 통해 환자 중심의 응급의료 환경 조성에 기여하고 있습니다.
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
In this study, we evaluated national differences in emergency department (ED) crowding to identify factors significantly associated with crowding in institutes and communities across Korea. This was a cross-sectional nationwide observational study using data abstracted from the National Emergency Department Information System (NEDIS). We calculated mean occupancy rates to quantify ED crowding status and divided EDs into three groups according to their occupancy rates (cutoffs: 0.5 and 1.0). Fact
The ED crowding was associated with increased hazard for hospital mortality for pediatric patients in mixed EDs.
We developed a prediction model of cardiac arrest in the ED using machine learning and sequential characteristics. The model was validated for clinical usefulness by chronological visualization focused on clinical usability.
A multidisciplinary mobile care system for patients with advanced gastrointestinal cancer was developed with clinically oriented measures. A prospective study was performed for its evaluation, which showed favorable satisfaction.
We developed a real-time autonomous ED dashboard and successfully used it for 5 years with good evaluation from users.
After introduction of the ICP, ED LOS decreased without an increase in hospital capacity.
In this retrospective study, we described the alert override patterns with a medication CDSS in an academic emergency department. We found relatively low overrides and assessed their contributing factors, including physicians' designation and specialty, patients' severity and chief complaints, and alert and medication type.
We successfully developed and validated machine learning-based prediction models to predict UE in ICU patients using electronic health record data. The best AUROC was 0.787 and the sensitivity was 0.949, which was obtained using the RF algorithm. The RF model was well-calibrated, and the Brier score and ICI were 0.129 and 0.048, respectively. The proposed prediction model uses widely available variables to limit the additional workload on the clinician. Further, this evaluation suggests that the
