시간흐름을 고려한 특징 추출과 군집 분석을 이용한 헬스 리스크 관리
Health Risk Management using Feature Extraction and Cluster Analysis considering Time Flow
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 논문에서는 시간 흐름을 고려한 특징추출과 군집분석을 이용한 헬스 리스크 관리를 제안한다. 제안하는 방법은 세단계로 진행한다. 첫 번째는 전처리 및 특징추출 단계이다. 이는 웨어러블 디바이스를 이용하여 라이프로그를 수집하여 불완전데이터, 에러, 잡음, 모순된 데이터를 제거하며 결측 값을 처리한다. 그 다음 특징추출을 위해 주성분 분석을 통해 중요 변수를 선택하고, 상관계수와 공분산을 통해 데이터 간의 관계와 유사한 데이터들의 분류를 진행한다. 또한 라이프로그에서 추출한 특징을 분석하기 위해 시간의 흐름을 고려하여 K-means 알고리즘을 통해 동적 군집을 진행한다. 새로운 데이터는 오차 제곱합의 증가분을 기반으로 유사성 거리 측정 방법을 통해 군집을 진행하고, 시간의 흐름을 고려하여 군집에 대한 정보를 추출한다. 따라서 특징 군집을 통해 헬스 의사결정 시스템을 이용하여 신체적 특성, 생활습관, 질병여부, 헬스케어 이벤트 발생위험, 예상 정도 등의 요소를 통해 리스크를 관리할 수 있다. 성능평가는 Precision, Recall, F-measure을 사용하여 제안하는 방법과 퍼지방법, 커널기반 방법을 비교한다. 평가결과 제안하는 방법이 우수하게 평가된다. 따라서 제안하는 방법을 통해 유병자와의 유사도를 이용하여 정확한 사용자의 잠재적 건강 위험을 예측 및 적절한 관리가 가능하다.
Abstract
In this paper, we propose health risk management using feature extraction and cluster analysis considering time flow. The proposed method proceeds in three steps. The first is the pre-processing and feature extraction step. It collects user’s lifelog using a wearable device, removes incomplete data, errors, noise, and contradictory data, and processes missing values. Then, for feature extraction, important variables are selected through principal component analysis, and data similar to the relationship between the data are classified through correlation coefficient and covariance. In order to analyze the features extracted from the lifelog, dynamic clustering is performed through the K-means algorithm in consideration of the passage of time. The new data is clustered through the similarity distance measurement method based on the increment of the sum of squared errors. Next is to extract information about the cluster by considering the passage of time. Therefore, using the health decision-making system through feature clusters, risks able to managed through factors such as physical characteristics, lifestyle habits, disease status, health care event occurrence risk, and predictability. The performance evaluation compares the proposed method using Precision, Recall, and F-measure with the fuzzy and kernel-based clustering. As a result of the evaluation, the proposed method is excellently evaluated. Therefore, through the proposed method, it is possible to accurately predict and appropriately manage the user’s potential health risk by using the similarity with the patient.
참고문헌 (17)
- Joo-Chang Kim, "Depression Index Service Using Knowledge Based Crowdsourcing in Smart Health", Wireless Personal Communications, 93(1), pp.255~268 (2017), Springer Science and Business Media LLC[DOI]
- Kyungyong Chung, "Edge computing health model using P2P-based deep neural networks", Peer-to-Peer Networking and Applications, 13(2), pp.694~703 (2020), Springer Science and Business Media LLC[DOI]
- Hoill Jung, "Associative context mining for ontology-driven hidden knowledge discovery", Cluster Computing, 19(4), pp.2261~2271 (2016), Springer Science and Business Media LLC[DOI]
- Ji-Won Baek, "Context Deep Neural Network Model for Predicting Depression Risk Using Multiple Regression", IEEE Access, 8, pp.18171~18181 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Chang-Min Kim, "Driver Facial Expression Analysis Using LFA-CRNN-Based Feature Extraction for Health-Risk Decisions", Applied Sciences, 10(8), pp.2956~ (2020), MDPI AG[DOI]
- Ji-Won Baek, "Hybrid clustering based health decision-making for improving dietary habits", Technology and Health Care, 27(5), pp.459~472 (2019), IOS Press[DOI]
- Kyungyong Chung, "Knowledge-based dynamic cluster model for healthcare management using a convolutional neural network", Information Technology and Management, 21(1), pp.41~50 (2020), Springer Science and Business Media LLC[DOI]
- Hoill Jung, "Ontology-driven slope modeling for disaster management service", Cluster Computing, 18(2), pp.677~692 (2015), Springer Science and Business Media LLC[DOI]
- Korea Health Statistics 2018: Korea National Health and Nutrition Examination Survey (KNHANES Ⅶ-3), , Korea Centers for Disease Control and Prevention, ~ (2017)
- "Health Insurance Review & Assessment Service"[URL]
- Ji-Soo Kang, "Activity Recommendation Model Using Rank Correlation for Chronic Stress Management", Applied Sciences, 9(20), pp.4284~ (2019), MDPI AG[DOI]
- Hoill Jung, "Life style improvement mobile service for high risk chronic disease based on PHR platform", Cluster Computing, 19(2), pp.967~977 (2016), Springer Science and Business Media LLC[DOI]
- Ji-Soo Kang, "PrefixSpan Based Pattern Mining Using Time Sliding Weight From Streaming Data", IEEE Access, 8, pp.124833~124844 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Davide Chicco, "The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation", BMC Genomics, 21(1) (2020), Springer Science and Business Media LLC[DOI]
- Z. Zivkovic., "Improved adaptive Gaussian mixture model for background subtraction", Proc of the 17th International Conference on Pattern Recognition, 2, pp.28~31 (2004)
- Meng Tang, "Kernel Cuts: Kernel and Spectral Clustering Meet Regularization", International Journal of Computer Vision, 127(5), pp.477~511 (2019), Springer Science and Business Media LLC[DOI]
- Dong-Hoon Shin, "Decision Boundary-Based Anomaly Detection Model Using Improved AnoGAN From ECG Data", IEEE Access, 8, pp.108664~108674 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]