허혈성 심장질환 진단을 위한 기계 학습 알고리즘 비교 연구 = A Comparative Study of Machine Learning Algorithms for Diagnosis of Ischemic Heart Disease

다국어 초록 (Multilingual Abstract)

In recent years, studies on artificial intelligence have been actively conducted, and artificial intelligence technology supports accurate and efficient decision-making for mankind. Also, the accumulation of medical knowledge and related data is accelerating, and studies on diagnosis of diseases through artificial intelligence technology are being carried out briskly. In this study, I chose a representative cardiovascular disease, specifically ischemic heart disease, as a research domain, and analyzed the available algorithms comparing effective approaches in the medical expert system for diagnosis of the disease. Concretely, the purpose of the study is to assist medical experts and physicians based on the initial patient record data, help them to explain the cause of ischemic heart disease, and minimize unnecessary related tests. In addition, the experimental data can be configured so that medical professionals can use them as learning models, thereby maximizing their experience and knowledge efficiently.

국문 초록 (Abstract)

최근, 인공지능에 대한 연구가 활발히 진행되고 있고, 인공지능 기술을 통한 정확하고 효율적인 의사결정이 가능해지고 있다. 또한, 점차 의료 지식 및 관련 데이터의 축적이 가속화되고 있으며, 인공지능 기술을 통한 질환 진단 및 처방에 대한 연구도 활발히 진행되고 있다. 본 연구에서는 대표적인 심혈관 질환인 허혈성 심장질환을 연구 도메인으로 설정하고, 해당 질환의 진단을 위한 의료 전문가 시스템 내에서 활용이 가능한 알고리즘과 효율적인 접근 방식을 비교 및 분석하여 제안한다. 본 연구의 궁극적 목표는 기존 환자의 초진기록 데이터를 바탕으로 의료 전문가 및 의사를 보조하는 것으로, 허혈성 심장질환에 대한 인과 관계 설명에 도움을 주고, 불필요한 관련 검사를 최소화한다는 데에 그 의미가 있다. 또한, 실험 데이터를 구성하여 의료 전문가 및 의사는 학습용 모델로 활용하면서, 이를 통해 경험과 지식을 효율적으로 극대화할 수 있다.

참고문헌 (Reference)

1 J. Y. Lee, "Statistics of Death Cause in Korea, 1st Ed." 2016

2 Noreen Kausar, "Review of Data Mining Approaches for Extraction and Classification of Clinical Data in Diagnosis of Coronary Artery Disease" 10 (10): 6679-6685, 2015

3 K. Y. Lee, "Present Condition and Assignment of Medical Artificial Intelligence" 219 (219): 7-11, 2016

4 S. L. Heo, "Policy Options to Tackle Unmet Health Needs, 1st Ed." 2009

5 Jim Adams, "Medical Service and Model, 1st Ed." 2015

6 Mikel Aickin, "Maximum Likelihood Estimation of Agreement in the Constant Predictive Probability Model, and Its Relation to Cohen's Kappa" 46 (46): 293-302, 1990

7 Korean Medical Association, "Main Pending Problem of 2015 Health and Medical Treatment, 1st Ed." 2016

8 Ankita R. Mokashi, "Heart Disease Prediction Using ANN and Improved K-Means" 4 (4): 214-224, 2016

9 Foster Provost, "Data Science for Business - What You Need to Know About Data Mining and Data-Analytic Thinking, 1st Ed." 2013

10 Aqueel Ahmed, "Data Mining Techniques to Find Out Heart Diseases: An Overview" 1 (1): 18-23, 2012

11 J. K. Lee, "Current Status and Future Policy Tasks of Health Promotion Project in Korea" 2005

12 M. G. Park., "Current Affairs Dictionary of pmg Knowledge Engine Laboratory"

13 Alladoumbaye Ngueilbaye, "Comparative Study of Data Mining Techniques on Heart Disease Prediction System: a case study for the "Republic of Chad" 5 (5): 1564-1571, 2016

14 P. W. Park, "Classification of Heart Disease Using K-Nearest Neighbor Imputation" 24 (24): 742-742, 2017

15 P. W. Park, "Classification of Heart Disease Comparing Imputation Algorithms with Classifiers" 3 (3): 18-19, 2017

16 Atul Kumar Pandey, "Classification Model for the Heart Disease Diagnosis" 14 (14): 9-14, 2014

17 Mudasir Manzoor Kirmani, "Cardiovascular Disease Prediction Using Data Mining Techniques: A Review" 10 (10): 520-528, 2017

18 Kantesh Kumar Oad, "A Fuzzy Rule based Approach to Predict Risk Level of Heart Disease" 14 (14): 17-22, 2014