• Annual Conference of KIPS
• /
• 2019.05a
• /
• pp.346-347
• /
• 2019

의료영상은 환자의 질병을 진단하고 치료방침을 결정하는데 중요한 도구로 자리매김하고 있다. 최근 의료영상을 인공지능 연구가 국내외에서 활발하게 진행되고 있다. 특히 대규모의 의료영상들을 학습시켜 질병과 상태를 정밀 진단할 뿐만 아니라 예측하는 소프트웨어를 개발 하는 상황이다. 그러나 의료영상은 DICOM 표준에 따르고 있지만 태그정보의 사용은 의료기기와 의료기관마다 상이하다. 따라서 의료영상에 대한 메타 데이터의 표준화에 어려움이 있다. 본 논문은 이러한 의료영상 데이터를 표준화 할 수 있는 방법을 제안한다. 그리고 제안한 표준화 데이터로 변환할 수 있는 ETL 소프트웨어의 수행결과를 보이고, 조건에 따라 머신러닝 학습 데이터셋을 생성하는 결과를 제공한다. 향후 제안한 의료영상 표준화와 ETL 소프트웨어는 다양한 수요자 중심의 표준화된 데이터셋을 제공할 수 있는 플랫폼의 주요기능으로 활용 될 것으로 기대한다.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.318-321
• /
• 2022

최근 의료데이터 표준화에 대한 중요성이 보건의료 빅데이터 구축과 맞물려 보건의료데이터 표준화와 마이데이터 생태계 조성을 추진하고 있다. 그리고 개인들의 휴대용 기기 이용증가와 모바일 환경으로 전반적인 디지털헬스의 패러다임 변화에 따라 HL7 FHIR의 사용이 점차 확대될 것으로 예측된다. 본 논문에서는 의료정보 표준인 HL7 FHIR와 의료영상 표준인 DICOM으로 환자 정보를 전달하기 위한 다중 의료 정보 중재 플랫폼에 대해서 기술한다. 이를 구현하기 위해 HL7 FHIR의 Patient, Observation, DiagnosticReport, Bundle 리소스를 활용하여 환자 정보와 임상 리포트 정보를 전달하여 StudyList에서 보여줄 수 있도록 구현하였다. 현재 구현된 내용은 FHIR 기반의 임상데이터로 의료영상을 포함한 표준화된 정보로 제공하여 마이데이터 실증 플랫폼으로 활용될 것으로 기대된다.

• The Korean Journal of Health Service Management
• /
• v.1 no.1
• /
• pp.110-124
• /
• 2007

Currently, we have seen sudden increase of demand for emergency medical services by reason for high-speed economic development increase of traffics, etc. in this society. Consequently the government enforced & operated emergency medical system in 1991 as a link of more positive countermeasure against it, but many problem; still remain in reality. In particular, the regional emergency medical center designated by the Ministry of Health and Welfare falls short of our expectations for its essential role by the reason of such as insufficiency in professional manpower and institutions concerned in small-scale hospital, matters with transportation system, preference of large hospital, etc. Therefore, this study was conducted grasp for the actual conditions of emergency medical system based upon literature research & the preceding studies and interview research the motive of coming hospital, satisfaction & understand to the subject of 150 persons of patients and their guardians who used regional emergency center of Pusan National University Hospital, thereby examining & analysing the cause of emergency room overcrowding by non-emergency cases, one of the problem; enumerated from preceding studies. The main result of this study is as follows. First the actual condition of non-emergency patients coming hospital for examining overcrowding of emergency medical center showed that, of the patients who used the emergency medical center, non-emergency patients accounted for 49.3%, which acted as the main cause of delaying medical care for emergency cases, cases of which medical person or first-aid man decided to come hospital accounted for 36.1 %, thereby suggesting essential need for re-education & wide public information to even the professional manpower besides patients & guardians for their using emergency medical system. Second, as the result of researching patient acknowledgement with reference to their using emergency medical center, the rate of their giving right answer is no more than 60%, which means that non-emergency cases' using as such is due to the shortage of their knowledge of the said emergency medical center, which suggests us that wider P.R for emergency medical system to common people who may be one of the patients of it at any time is still in need. Third, the result of researching for finding out a future remedy of emergency medical system showed that the users who know well of the way of using emergency medical center had relatively high satisfaction of it, ones who have lesser knowledge of it lower satisfaction and users who feel in need of emergency specialized manpower feel the necessity of public information of emergency medical information center(1339) at the same time. The finding of examining the subject of study, in conclusion, showed that the degree of the patients & their guardians' understanding of emergency medical system is lower and the medical persons concerned also had no distinct difference in their understanding of it from the common people's, which suggests us for extensive enforcement of systematic education and public information in aspect of the government via various media for the purpose of effective operation of emergency medical center.

• Annual Conference of KIPS
• /
• 2010.04a
• /
• pp.352-352
• /
• 2010

It is difficult to settle the well-designed local terminology for imaging report in the hospital information system (HIS). One of the major reasons is the local terminology with poor contents have been used in the hospital. Thus, we mapped the locally used terms in nuclear medicine imaging report to the SNOMED-CT, which had been widely used in the electronic medical record system, for implementation of hospital information system. Preliminary construction of terminology dictionary was done by mapping of local terms to SNOMED-CT and LexCare Suite. Further study may be warranted.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.9
• /
• pp.4461-4475
• /
• 2017

Recently, introduction of the hospital information system has remarkably improved the efficiency of health care services within hospitals. Due to improvement of the hospital information system, the issue of integration of medical information has emerged, and attempts to achieve it have been made. However, as a preceding step for integration of medical information, the problem of searching the same patient should be solved first, and studies on patient identification algorithm are required. As a typical case, similarity can be calculated through MPI (Master Patient Index) module, by comparing various fields such as patient's basic information and treatment information, etc. but it has many problems including the language system not suitable to Korean, estimation of an optimal weight by field, etc. This paper proposes a method searching the same patient using MRI information besides patient's field information as a supplementary method to increase the accuracy of matching algorithm such as MPI, etc. Unlike existing methods only using image information, upon identifying a patient, a highest weight was given to physical information of medical image and set as an unchangeable unique value, and as a result a high accuracy was detected. We aim to use the similarity measurement result as secondary measures in identifying a patient in the future.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.428-429
• /
• 2018

본 논문에서는 다기관 의료영상 분석 방법 및 시스템을 제안한다. 다기관 연구에 참여하는 기관에게 분석 가이드 및 분석 프로그램을 제공하여 표준화된 영상분석 연구를 지원하고자 한다. 이를 위해 동일한 프로토콜로 표준화된 영상을 획득 및 분석하고 결과를 공유하는 분산형 연구방법을 제시한다. 제안하는 시스템은 개인정보보호법 및 보안문제가 강조되고 있는 의료현장에 적합한 시스템으로 다양한 다기관 의료 빅데이터 분석 연구에 활용될 것으로 기대된다.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2004.07a
• /
• pp.260-260
• /
• 2004

• Annual Conference of KIPS
• /
• 2019.05a
• /
• pp.348-349
• /
• 2019

의료영상은 진단을 목적으로 환자의 질환 부위별로 정밀하게 촬영하여 수집된 영상이다. 수집된 의료영상을 판독하여 진단하기 어려운 경우에는 조직검사를 통해 확진검사를 실시한다. 하지만 조직검사의 경우 환자에게 신체적, 경제적 부담을 줄 수 있다. 따라서 의료영상을 기반으로 정밀 진단하는데 도움을 줄 수 있는 정량분석 소프트웨어 개발이 중요하다. 현재 복부 간 질환의 경우 MELD(Model For End-Stage Liver Disease) 점수를 이용하여 간 질환의 중증도나 예후를 예측하는데 이용되고 있다. 하지만 MELD점수 산출에 있어서 의료영상 정보를 사용하지 않았기 때문에 질환 여부를 가늠하는데 에만 이용될 뿐 병변부위의 위치를 확인하는데 에는 어려움이 있다. 그러므로 본 논문은 다양한 의료영상장비에서 획득한 복부영상을 이용하여 복부질환의 중증도를 예측 및 분석함으로서 실제 임상진단에 도움을 줄 수 있는 시스템을 제안하고자 한다.

• Clinics in Shoulder and Elbow
• /
• v.26 no.1
• /
• pp.25-31
• /
• 2023

Background: This study analyzed questions searched by rotator cuff patients and determined types and quality of websites providing information. Methods: Three strings related to rotator cuff repair were explored by Google Search. Result pages were collected under the "People also ask" function for frequent questions and associated webpages. Questions were categorized using Rothwell classification and topical subcategorization. Webpages were evaluated by Journal of the American Medical Association (JAMA) benchmark criteria for source quality. Results: One hundred twenty "People also ask" questions were collected with associated webpages. Using the Rothwell classification of questions, queries were organized into fact (41.7%), value (31.7%), and policy (26.7%). The most common webpage categories were academic (28.3%) and medical practice (27.5%). The most common question subcategories were timeline of recovery (21.7%), indications/management (21.7%), and pain (18.3%). Average JAMA score for all 120 webpages was 1.50. Journal articles had the highest average JAMA score (3.77), while commercial websites had the lowest JAMA score (0.91). The most commonly suggested question for rotator cuff repair/surgery was, "Is rotator cuff surgery worth having?," while the most commonly suggested question for rotator cuff repair pain was, "What happens if a rotator cuff is not repaired?" Conclusions: The most commonly asked questions pertaining to rotator cuff repair evaluate management options and relate to timeline of recovery and pain management. Most information is provided by medical practice, academic, and medical information websites, which have highly variable reliability. By understanding questions their patients search online, surgeons can tailor preoperative education to patient concerns and improve postoperative outcomes.

• Journal of Preventive Medicine and Public Health
• /
• v.43 no.3
• /
• pp.257-264
• /
• 2010

Objectives: An accurate estimation of cancer patients is the basis of epidemiological studies and health services. However in Korea, cancer patients visiting out-patient clinics are usually ruled out of such studies and so these studies are suspected of underestimating the cancer patient population. The purpose of this study is to construct a more complete, hospital-based cancer patient registry using multiple sources of medical information. Methods: We constructed a cancer patient detection algorithm using records from various sources that were obtained from both the in-patients and out-patients seen at Asan Medical Center (AMC) for any reason. The medical data from the potentially incident cancer patients was reviewed four months after first being detected by the algorithm to determine whether these patients actually did or did not have cancer. Results: Besides the traditional practice of reviewing the charts of in-patients upon their discharge, five more sources of information were added for this algorithm, i.e., pathology reports, the national severe disease registry, the reason for treatment, prescriptions of chemotherapeutic agents and radiation therapy reports. The constructed algorithm was observed to have a PPV of 87.04%. Compared to the results of traditional practice, 36.8% of registry failures were avoided using the AMC algorithm. Conclusions: To minimize loss in the cancer registry, various data sources should be utilized, and the AMC algorithm can be a successful model for this. Further research will be required in order to apply novel and innovative technology to the electronic medical records system in order to generate new signals from data that has not been previously used.