• Proceedings of the Korean Statistical Society Conference
• /
• 2005.05a
• /
• pp.291-298
• /
• 2005

지속성 외래 복막투석은 말기 신부전 환자들에게 널리 시행하는 신 대체 요법으로, 복막투석 환자에게서 주된 합병증으로 일어나는 단백질-열량 영양실조를 치료하기 위하여 아미노산을 복강 내로 주입하는 치료방법이다. 이현석 등(2004)의 연구에서는 아미노산 복막 투석액(IPAA)이 영양실조 환자들에게 실제로 영양상태에 미치는 영향을 평가하기 위하여 지속성 외래 복막투석 환자 43명을 12개월 동안 3개월 주기로 관측하여 얻어낸 반복측정자료를 바탕으로 IPAA의 효과 여부에 따라 반응군과 비반응군을 분류하였다. 본 논문에서는 이러한 두 그룹을 효과적으로 분류할 수 있는 분류기준변수들을 찾아내고 이 분류기준변수의 값을 바탕으로 새로운 환자에게 IPAA의 투여 여부를 진단할 수 있는 여러 분류방법들을 고찰하여 비교 연구하였다. 모수적인 방법으로 선형판별분석, 이차판별분석 및 로지스틱 판별분석을 소개하고 비모수적인 방법으로 support vector machine(SVM)을 소개하여 분류분석의 결과를 비교하여 두 그룹을 최소한의 오류로 분류하는 방법을 제안하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.314-316
• /
• 2009

본 논문에서는 사상체질분류검사 설문지를 이용하여 사상체질을 진단할 때, 진단의 정확도를 향상시키기 위한 사상체질 분류 함수를 개발하기 위하여 퍼지 분류기를 이용한다. 본 연구에서 사용하는 데이터는 9개 한의과대학의 10개 부속한방병원에서 치료를 받은 환자들 중 각 병원의 사상체질전문의로부터 체질진단을 받고 최소한 4주 이상 사상체질 처방을 사용한 후 주 증상이 전반적으로 호전되어 체질이 확인된 환자 1,914명을 대상으로 하고 있다. 본 연구는 사상체질의학의 광제설을 토대로 환자의 성별을 분리 하였을 뿐만 아니라, 비만도를 추가적으로 분류하였으며, 체형기상, 용모사기, 성질재간, 병증약리 중 체형기상을 토대로 분류하였으며, 사상체질을 판별할 수 있도록 설계되고 구현되었다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.9-17
• /
• 2021

In this study, we propose a part of the CDSS(Clinical Decision Support System) study, a system that can classify chemotherapy, one of the treatment methods for colorectal cancer patients. In the treatment of colorectal cancer, the selection of chemotherapy according to the patient's condition is very important because it is directly related to the patient's survival period. Therefore, in this study, chemotherapy was classified using a machine learning algorithm by creating a baseline model, a pathological model, and a combined model using both characteristics of the patient using the individual and pathological characteristics of colorectal cancer patients. As a result of comparing the prediction accuracy with Top-n Accuracy, ROC curve, and AUC, it was found that the combined model showed the best prediction accuracy, and that the LGBM algorithm had the best performance. In this study, a chemotherapy classification model suitable for the patient's condition was constructed by classifying the model by patient characteristics using a machine learning algorithm. Based on the results of this study in future studies, it will be helpful for CDSS research by creating a better performing chemotherapy classification model.

• The Journal of the Korea Contents Association
• /
• v.13 no.3
• /
• pp.287-298
• /
• 2013

To develop a technique classifying patients based on computerized clinical data followed by validity verification by comparing with nurse's examination. Class scores were determined by nurses for a day on 348 resident patients in 7 wards of a general hospital according to KPCS-1. The class scores were simultaneously evaluated by reviewing the computerized clinical data acquired from the hospital management information system. These two class scores were both significantly different among different departments as well as disease patterns. Intraclass correlation analysis resulted a very high correlation coefficient of 0.96(p<0.01) between the two scoring methods, but the clinical data scores were somewhat higher. An automated patient classification system seemed possible to be developed in future with further enhancement of the present results based on computerized clinical data without manual scoring, which can be applied for performance evaluation as well as workforce planning.

• Proceedings of the KAIS Fall Conference
• /
• 2010.11b
• /
• pp.759-762
• /
• 2010

본 연구는 당뇨인지환자들의 당뇨 조절에 관계되는 요인들을 포괄적으로 반영하는 집단으로 분류한 후 이를 기반으로 보다 효율적인 당뇨관리사업을 할 수 있는 기초자료를 제공하기 위해 수행되었다. 연구를 위해 2007년, 2008년도 국민건강영양조사를 통해 검진에 참여한 당뇨인지환자 666명의 자료를 수집하여 분석하였다. 당뇨인지환자의 관리행태에 대한 군집분류는 K-means 기법을 이용하였다. 당뇨인지환자의 군집은 건강행태사업 대상군, 중점관리사업 대상군, 합병증검사사업 대상군으로 분류되었다. 당뇨 조절율을 높이기 위해서는 각 군집의 특성에 따라 보다 특화된 당뇨관리 프로그램이 적용되어야 할 것이다.

• Journal of the Society of Disaster Information
• /
• v.19 no.1
• /
• pp.44-59
• /
• 2023

Purpose: In the event of mass casualties, triage must be done promptly and accurately so that as many patients as possible can be recovered and returned to the battlefield. However, medical personnel have received many tasks with less manpower, and the battlefield for classifying patients is too complex and uncertain. Therefore, we studied an artificial intelligence model that can assist and replace medical personnel on the battlefield. Method: The triage model is presented using reinforcement learning, a field of artificial intelligence. The learning of the model is conducted to find a policy that allows as many patients as possible to be treated, taking into account the condition of randomly set patients and the medical capability of the military hospital. Result: Whether the reinforcement learning model progressed well was confirmed through statistical graphs such as cumulative reward values. In addition, it was confirmed through the number of survivors whether the triage of the learned model was accurate. As a result of comparing the performance with the rule-based model, the reinforcement learning model was able to rescue 10% more patients than the rule-based model. Conclusion: Through this study, it was found that the triage model using reinforcement learning can be used as an alternative to assisting and replacing triage decision-making of medical personnel in the case of mass casualties.

• Journal of Digital Convergence
• /
• v.10 no.10
• /
• pp.347-352
• /
• 2012

As medical treatment is developing with technology, the men's average life expectancy is extended. Therefore, primary medical care becomes emphasized in order to reduce the medical expenses in the long term by satisfying individual's life being healthy. The date for this thesis was collected from January 2011 to June 2011. 889 patients who visited the university hospital emergency room and hospitalized in internal medicine, were picked as the research subjects and they were targeted to be recorded the distribution of chief complaint and principal diagnosis of the patients. Also, this record was used to apply to the standard Classification of Diseases(as known as ICD) and the method of detailed classification of the primary medical care(as known as ICPC) to compare each other. In order to analysis, frequency analysis was used to see vital statistics and the cross tabulations were used to see the distribution of chief complaint according to ICD and ICPC. Results of the research were Abdominal pain(17.7%), Dyspnea(13.5%), Fever (12.5%), and Haematemesis (9.8%), and those symptoms represented the 54.5% of overall chief complaints that is treated in primary care. Therefore, it is acceptable to use the classification of the primary medical care at doc-in-a-box. Also, in case of diagnosis of abdominal pain, it is classified to R10 in ICD and 116 patients(18.7%) belonged to it, but according to ICPC, it is subdivided to Epigastric(11.5%) and General(5.8%). ICPC classification, which is focused to primary medical care is more detailed than ICD classification. Because the data that is collected for this thesis is from only one hospital, it is hard to represent to all the cases, but ICPC in emergency medical care, it has more classification available and it can subdivide the patients effectively, so it is meaningful.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.295-297
• /
• 2001

최근 생물 유전자 정보를 효과적으로 분석하기 위한 적절한 도구의 필요성이 대두되고 있다. 본 논문에서는 백혈병 환자의 골수로부터 얻어낸 DNA Microarray 유전 정보를 분류하여 환자가 가지고 있는 암의 종류를 예측하기 위한 최적의 특징추출방법과 분류 방법을 찾고자 한다. 이를 위해 피어슨 상관관계, 유클리디안 거리, 코사인 계수, 스피어맨 상관관계, 정보 이득, 상호 정보, 신호 대잡음비의 7가지 특징 추출 방법을 사용하였으며, 역전과 신경망, 의사결정 트리, 구조 적응형 자기구성 지도, $textsc{k}$-최근접 이웃 등 가지의 기계학습 분류기를 이용하여 분류 실험을 하였다. 실험결과, 피어슨 상관관계와 역전파 신경망을 이용한 분류 방법이 97.1%의 인식률을 보임을 알 수 있었다.

• Journal of Digital Convergence
• /
• v.13 no.7
• /
• pp.269-282
• /
• 2015

This study is to define the hemodiafiltration treatment compliance indicators and discriminate standards for hemodiafiltration patients and development of hemodiafiltration treatment compliance measurement-convergent form. Date was collected from 300 on-line hemodiafiltration patients. To verify the hemodiafiltration treatment compliance indicators and discriminate standards, used construct validity and content validity by clinical professional group. Discriminant ability of 3 indicators-interdialysis weight gain rate(IWGR), serum phosphate level, rate of self change of total hemodiafiltration treatment time(SCR-HEFTT)- is 95.6%(wilks ramda=.256, p=.002). And hemodiafiltration treatment compliance measurement-convergent form has 91.7% discriminant accuracy. Hemodiafiltration treatment compliance is important that nurses can aware pre-stage of complication and give appropriate nursing intervention. Also this measurement can be used for foundation data of the nursing intervention development that prevent dialysis patient's complication.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.213-214
• /
• 2009

무선 센서들의 진보에 따라 환자의 상태를 모니터링 하거나 정보를 저장 후 원거리에 있는 의사들의 진단 제공이 가능하게 되었다. 하지만 환자의 데이터의 양에 비해 의사의 수가 적으므로 환자가 진단을 제공 받는데 시간적인 한계가 있다. 따라서 본 연구에서는 환자의 상태를 1 차적으로 자동 진단하는 시스템을 제안한다. 전체 데이터의 적용을 위해 Circadian rhythm에 기반한 데이터 직접방법을 제안하고 데이터를 효율적으로 분류하기 위해 PSO(Particle Swarm Optimization)을 기반으로 하는 분류화 알고리즘을 적용하여 시스템의 수행속도 향상을 도모하였다.