약물부작용 감시를 위한 공통데이터모델 기반 임상데이터웨어하우스 구축

노미정 ( Mi Jung Rho ) 한국병원경영학회 2023 병원경영학회지 Vol.28 No.3

Purposes: It is very important to establish a clinical data warehouse based on a common data model to offset the different data characteristics of each medical institution and for drug surveillance. This study attempted to establish a clinical data warehouse for Dankook university hospital for drug surveillance, and to derive the main items necessary for development. Methodology/Approach: This study extracted the electronic medical record data of Dankook university hospital tracked for 9 years from 2013 (2013.01.01. to 2021.12.31) to build a clinical data warehouse. The extracted data was converted into the Observational Medical Outcomes Partnership Common Data Model (Version 5.4). Data term mapping was performed using the electronic medical record data of Dankook university hospital and the standard term mapping guide. To verify the clinical data warehouse, the use of angiotensin receptor blockers and the incidence of liver toxicity were analyzed, and the results were compared with the analysis of hospital raw data. Findings: This study used a total of 670,933 data from electronic medical records for the Dankook university clinical data warehouse. Excluding the number of overlapping cases among the total number of cases, the target data was mapped into standard terms. Diagnosis (100% of total cases), drug (92.1%), and measurement (94.5%) were standardized. For treatment and surgery, the insurance EDI (electronic data interchange) code was used as it is. Extraction, conversion and loading were completed. R language-based conversion and loading software for the process was developed, and clinical data warehouse construction was completed through data verification. Practical Implications: In this study, a clinical data warehouse for Dankook university hospitals based on a common data model supporting drug surveillance research was established and verified. The results of this study provide guidelines for institutions that want to build a clinical data warehouse in the future by deriving key points necessary for building a clinical data warehouse.

Conversion and Data Quality Assessment of Electronic Health Record Data at a Korean Tertiary Teaching Hospital to a Common Data Model for Distributed Network Research

Yoon, Dukyong,Ahn, Eun Kyoung,Park, Man Young,Cho, Soo Yeon,Ryan, Patrick,Schuemie, Martijn J.,Shin, Dahye,Park, Hojun,Park, Rae Woong Korean Society of Medical Informatics 2016 Healthcare Informatics Research Vol.22 No.1

<P><B>Objectives</B></P><P>A distributed research network (DRN) has the advantages of improved statistical power, and it can reveal more significant relationships by increasing sample size. However, differences in data structure constitute a major barrier to integrating data among DRN partners. We describe our experience converting Electronic Health Records (EHR) to the Observational Health Data Sciences and Informatics (OHDSI) Common Data Model (CDM).</P><P><B>Methods</B></P><P>We transformed the EHR of a hospital into Observational Medical Outcomes Partnership (OMOP) CDM ver. 4.0 used in OHDSI. All EHR codes were mapped and converted into the standard vocabulary of the CDM. All data required by the CDM were extracted, transformed, and loaded (ETL) into the CDM structure. To validate and improve the quality of the transformed dataset, the open-source data characterization program ACHILLES was run on the converted data.</P><P><B>Results</B></P><P>Patient, drug, condition, procedure, and visit data from 2.07 million patients who visited the subject hospital from July 1994 to November 2014 were transformed into the CDM. The transformed dataset was named the AUSOM. ACHILLES revealed 36 errors and 13 warnings in the AUSOM. We reviewed and corrected 28 errors. The summarized results of the AUSOM processed with ACHILLES are available at http://ami.ajou.ac.kr:8080/.</P><P><B>Conclusions</B></P><P>We successfully converted our EHRs to a CDM and were able to participate as a data partner in an international DRN. Converting local records in this manner will provide various opportunities for researchers and data holders.</P>

프라이버시와 개인화를 위한 고품질 표준 데이터 기반 약물감시 시스템 연구

양세모,송인서,이강윤 사)한국빅데이터학회 2023 한국빅데이터학회 학회지 Vol.8 No.2

전 세계적으로 약물 부작용은 주요 사망원인의 상위를 차지하고 있다. 약물 부작용에 대해 효과적으로 대응하기 위해, 능동적인 실시간 분석 기반 약물감시 체계로의 전환과 함께 데이터의 표준화와 품질 향상이 필요하다. 이를 위해, 개별 기관의 데이터를 통합하고 대규모 데이터를 활용하여 약물 부작용 예측의 정확도를 높이는 것이 중요하다. 하지만, 각 기관 간의 데이터 공유는 프라이버시 문제를 야기시키고 각기 다른데이터 표준 구성도 다르다. 본 연구에서는 이 문제를 해결하기 위해, 개인정보보호 법규에 따라 데이터를직접 공유하지 않고 모델의 학습 결과를 공유하는 연합학습 방식을 채택한다. 각 기관마다 다른 데이터포맷을 Common Data Model(CDM)을 활용하여 데이터 표준화를 수행하고 데이터의 정확성과 일관성을 확립한다. 또한, 클라우드 기반의 연합학습 환경을 구성하여 보안 및 확장 관리에 효율성을 높이는 약물감시시스템을 제안한다. 이를 통해 기관 간 데이터의 프라이버시를 보호하면서도, 효과적인 의약품 부작용 모니터링과 예측이 가능하다. 약물 부작용으로 인한 사망률 감소와 의료비용 절감을 목표로 하며, 이를실현하기 위한 다양한 기술적 접근과 방법론을 탐구한다. Globally, drug side effects rank among the top causes of death. To effectively respond to these adverse drug reactions, a shift towards an active real-time monitoring system, along with the standardization and quality improvement of data, is necessary. Integrating individual institutional data and utilizing large-scale data to enhance the accuracy of drug side effect predictions is critical. However, data sharing between institutions poses privacy concerns and involves varying data standards. To address this issue, our research adopts a federated learning approach, where data is not shared directly in compliance with privacy regulations, but rather the results of the model’s learning are shared. We employ the Common Data Model (CDM) to standardize different data formats, ensuring accuracy and consistency of data. Additionally, we propose a drug monitoring system that enhances security and scalability management through a cloud-based federated learning environment. This system allows for effective monitoring and prediction of drug side effects while protecting the privacy of data shared between hospitals. The goal is to reduce mortality due to drug side effects and cut medical costs, exploring various technical approaches and methodologies to achieve this.

머신러닝과 공통데이터모델을 활용한 국가 간암 검진 대상자의 간암 예측 모델

이명철,최경선,서혜선 한국보건사회약료경영학회 2022 한국보건사회약료경영학회지 Vol.10 No.1

BACKGROUNDS To find early liver cancer, the ministry of health and welfare has conducted surveillance targeting high-risk patients. In 2017, the incidence rate of liver cancer in surveillance was 0.9%, suggesting that a broad patient group was included in surveillance. In this study, to reduce surveillance patients, a prediction model with zero-falsenegative was developed using a machine learning. METHODS To develop the model, we used 2016 Health Insurance Review & Assessment Service-National Patients Sample utilized to the Common Data Model (CDM). This study targeted patients who did not have a severe condition of liver cancer in surveillance. The number of the target was 13,703 cases. The covariates for the model were identified by a chi-square test conducted on gender, age group, condition between a case and control group. LASSO was performed to develop the model. RESULTS Gender, age group, forty diseases were selected as a covariate. The model has an AUC of 0.745, a negative rate of 4.0%, a specificity of 4.5%, and a PPV of 11.8% with zerofalse- negative. CONCLUSION It might be possible to refine surveillance and save the budget of the National Health Insurance Service, and governments.

항만 BIM 데이터의 상호운용성 확보를 위한 IFC 표준 개발

문현석,원지선,신재영 한국BIM학회 2020 KIBIM Magazine Vol.10 No.1

Recently, BIM has been extended to infrastructures such as roads and bridges, and the demand for BIM standard development for ports is increasing internationally. Due to the low level of utilization of classification system and drawing standards compared to other infrastructures, and the closed nature of national security facilities, ports have insufficient level of connection and sharing environment among external systems or users. In addition, since the standardization of data for port facilities is not made, it is still necessary to establish an independent DB for each system and to ensure interoperability of data between these systems since it does not have a shared environment among similar data. Therefore, the purpose of this study is to develop and verify IFC, the international standard for BIM, in order to cope with the BIM environment and to be commonly used in the design, construction, and maintenance of port facilities. To this end, we build a standard schema with port-specific Express Notation according to buildingSMART International's standard development methodology. First, domestic and international reference model standards were analyzed to derive components such as space and facilities of port facilities. Based on this, the components of the port facility were derived through the codification, categorization, and normalization process developed by the research team. This was extended based on the port BIM object classification system developed by the research team. Normalization results were verified by designers and associations. Then, IFC schema construction was based on Express-G data modeling based on IFC 4 * 2 Candidate, which is a bridge candidate standard based on IFC4 (ISO16739), and IFC 4 * 3 Draft, which is developed by buildingSMART International. The final schema was validated using the commercialized validation tool. In addition, in order to verify the structural verification of the port IFC schema, the transformation process was verified by converting the caisson model into a Part21 file. In the future, this result will not only be used as a delivery standard for port BIM products, but will also be applied as a linkage standard between systems and a common data format for port BIM platforms when BIM is used in the maintenance phase. In particular, it is expected to be used as a core standard for data exchange in the port maintenance stage.

OMOP CDM 구축 시 개인의료정보 보호를 위한 HIPAA PHI 적용 방법 연구

김학기,정은영,박동균 한국차세대컴퓨팅학회 2017 한국차세대컴퓨팅학회 논문지 Vol.13 No.6

In this study, we investigated how to protect personal healthcare information when constructing OMOP (Observational Medical Outcomes Partnership) CDM (Common Data Model). There are two proposed methods; to restrict data corresponding to HIPAA (Health Insurance Portability and Accountability Act) PHI (Protected Health Information) to be extracted to CDM or to disable identification of it. While processing sensitive information is restricted by Korean Personal Information Protection Act and medical law, there is no clear regulation about what is regarded as sensitive information. Therefore, it was difficult to select the sensitive information for protecting personal healthcare information. In order to solve this problem, we defined HIPAA PHI as restriction criterion of Article 23 of the Personal Information Protection Act and maps data corresponding to CDM data. Through this study, we expected that it will contribute to the spread of CDM construction in Korea as providing solutions to the problem of protection of personal healthcare information generated during CDM construction. 본 연구에서는 OMOP(Observational Medical Outcomes Partnership) CDM(Common Data Model) 구축시 개인의료정보를 보호하는 방법을 연구하였다. 제안된 방법은 HIPAA(The Health Insurance Portability and Accountability Act) PHI(Protected Health Information)에 대응되는 데이터가 CDM으로 추출 되는 것을 제한하거나 식별 불능 화 처리 하는 것이다. 하지만 한국의 개인정보보호법 및 의료법에는 민감 정보의 처리 제한에 관한 내용은 존재하나 그 민감 정보가 무엇인지에 관한 명확한 규정은 없어 개인의료정보 보호를 위한 민감 정보 선정에 어려움이 있다. 본 연구에서는 이러한 문제를 해결하기 위해 HIPAA PHI를 개인정보 보호법 제23조 민감 정보의 처리 제한 기준으로 정하고 CDM데이터와 매핑 하였다. 본 연구를 통해 CDM구축 시 발생되는 개인의료정보 보호문제에 대한 해결 방법을 제시함으로써 국내 CDM구축 확산에 기여할 것으로 예상된다.

한국형 EMS 시스템의 Baseline 계통 해석용 소프트웨어 개발을 위한 데이터 모델링

윤상윤(Sang-Yun Yun),조윤성(Yoon-Sung Cho),이욱화(Wook-Hwa Lee),이진(Jin Lee),손진만(Jin-Man Sohn) 대한전기학회 2009 전기학회논문지 Vol.58 No.10

This paper summarizes a data modeling for developing the baseline network analysis software of the Korean energy management system (EMS). The study is concentrated on the following aspects. First, the data for operating the each application software are extracted. Some of the EMS network application softwares are selected for basis model. Those are based on the logical functions of each software and are not considered the other softwares. Second, the common data are extracted for equipment model and topological structure of power system in Korea. We propose the application common model(ACM) that can be applied whole EMS network application softwares. The ACM model includes the hierarchy and non-hierarchy power system structure, and is connected each other using the direct and indirect link. Proposed database model is tested using the Korea Electric Power Corporation(KEPCO) system. The real time SCADA data are provided for the test. Through the test, we verified that the proposed database structure can be effectively used to accomplish the Korean EMS system.

Effects of Model-Data Misfit on IRT Equating for Polytomous Data

전경희(Kyong Hee Chon) 한국교육평가학회 2016 교육평가연구 Vol.29 No.4

이 연구에서는 모형-자료 부적합이 다분 문항반응반이론 모형을 이용한 동등화 결과에 미치는 영향에 대해 살펴보았다. 비등등집단 가교문항 동등화설계를 가정한 모의연구를 통해 모형-자료 간 부적합 정도와 가교문항 및 일반문항 세트 내에서의 부적합 문항의 위치 등의 조건을 달리하였을 때 다분문항반응 자료에 대한 동등화 결과가 어떻게 다르게 나타나는지 점수변환표 상에서의 일치도와 동등성 지표를 기준으로 분석하였다. 모의실험 결과, 모형-자료 간 부적합 문항이 포함될 경우 점수변환체계 상의 동등화 관계에 있어 왜곡이 나타났으며, 이러한 왜곡 현상은 부적합 문항이 가교문항 세트 보다 일반 문항세트에 포함될 경우 더 심각해지는 것으로 분석되었다. 동등성 지표를 이용한 결과비교에 있어서도 부적합 문항이 일반 문항세트 중에 배치될 때 부정적인 결과를 보였다. 또한, 모형-자료간 부적합 정도와 검사 내 부적합 문항 위치에 따른 영향을 비교할 때 부적합 문항의 위치에 의한 영향이 더 큰 것으로 나타났다. This study examines the effect of model-data misfit on polytomous IRT equating under the common-item nonequivalent groups design. A simulation study was conducted to assess equating results according to different impacts and aspects of model-data misfit. The study factors include the proportion and the location of misfitting items on uncommon and common item sets. Equating results from different misfit conditions were evaluated based on two criteria: consistency in score conversions and equity properties. This study found that the equating relationship was substantially altered when a test from involves misfitting items in comparison to the situation where no misfit exists. Distortion of the score conversions tended to be more severe when misfits occurred in uncommon items rather than common items. First- and second- equity properties were not well achieved when uncommon items were functioning as misfitting items. The location of misfit rather than proportion of misfit has a larger impact on the equating results.

Ontology Driven Meta-Modeling for NoSQL Databases: A Conceptual Perspective

Shreya Banerjee,Anirban Sarkar 보안공학연구지원센터 2016 International Journal of Software Engineering and Vol.10 No.12

In cloud environment, data intensive applications required to interact with various kinds of databases ranged from SQL (Structured Query Language) to NoSQL. NoSQL databases have wide-ranging physical level data models like key-value, document, column oriented, graph etc. Designing such databases are complex due to absence of well-accepted common meta-modeling concepts for varied physical level data model. Thus, research challenges are exist towards achieving common interpretation of data and their semantics at conceptual level over disparate databases. In this paper, an ontology driven meta-model, called Ontology Driven NoSQL Data Model (ODNSDM), is proposed to conceptualize data representation facets over heterogeneous kinds of databases. The novelty of the proposed approach is to provide common conceptual level abstraction based on semantically enriched formal vocabularies for both NoSQL and SQL databases. The formal vocabularies are further implemented using an ontology editorial tool Protégé based on OWL (Web Ontology Language). Several crucial properties of the proposed model are also prescribed in this paper. In addition, the proposed framework is illustrated using case studies in order to show its practical exhibition. Moreover, step wise algorithms are to map the proposed conceptualization towards SQL and NoSQL based databases.

우울증 환자에서 항우울제 단독요법과 항정신병제와 항우울제의 병용요법의 자살위험비교: 정신건강의학 공통 데이터 모델을 활용한 파일럿 연구

하재호,이은영,이동윤,조용혁,이혜림,박범희,손상준 대한신경정신의학회 2020 신경정신의학 Vol.59 No.3

Objectives This study examined the effects of reducing suicide attempts when taking antipsychotics in combination with antidepressants in depressive patients. Methods Using a common data model of electronic medical records at a university medical center in South Korea, the study populations were extracted if the depressive patients were treated either with antidepressants only or along with antipsychotics. The suicidal risks were compared with the Kaplan-Meier plot and log-rank test, and the risk factors were accessed using the Cox proportional hazard model. Results All demographic characteristics were similar in the monotherapy group taking only antidepressants and the combination therapy group taking antipsychotics with antidepressants, except for the smoking characteristic (p=0.023). The combination therapy group showed a lower suicidal risk [hazard ratio=0.58, 95% confidence interval (CI)=0.282–1.190] compared to the monotherapy group, which was not significant (p=0.138). Conclusion The combination therapy had no beneficial effects on reducing the suicidal risk in patients with depressive symptoms. This study is meaningful in that it is the first attempt to explore a psychiatric behavior/symptom using real-world data based on a common data model of general electronic medical records as well as narrative textual data.