DOI QR코드

DOI QR Code

Classification of the presence or absence of underlying disease in EEG Data using neural network

뉴럴네트워크를 이용하여 EEG Data의 기저질환 유무 분류

  • 윤희진 (장안대학교 IT학부 인터넷정보통신과)
  • Received : 2020.11.02
  • Accepted : 2020.12.20
  • Published : 2020.12.28

Abstract

In January 2020, COVID19 plunged the whole planet into a pandemic. This has caused great economic losses and is causing social confusion. COVID19 has a superior infection rate among people with underlying disease such as heart disease, high blood pressure, diabetes, stroke, depression, and cancer. In addition, it was studied that patients with underlying disease had a higher fatality rate than those without underlying disease. In this study, the presence or absence of underlying disease was classified using EEG data. The data used to classify the presence or absence of underlying disease was EEG data provided by Data Science lab, consisting of 33 features and 69 samples. Z-score was used for data pretreatment. Classification was performed using the neural network NEWFM and ZNN engine. As a result of the classification of the presence or absence of the underlying disease, the experimental results were 77.945 for NEWFM and 76.4% for ZNN. Through this study, it is expected that EEG data can be measured, the presence or absence of an underlying disease is classified, and those with a high infection rate can be prevented from COVID19. Based on this, there is a need for research that can subdivide underlying disease in the future and research on the effects of each underlying disease on infectious disease.

2020년 1월, COVID19는 온 지구를 팬데믹에 빠트렸다. 이로 인해 경제적으로 큰 손실을 가져왔으며, 사회적으로 혼란을 일으키고 있다. 이러한 코로나19는 심장병, 고혈압, 당뇨, 뇌졸중, 우울증, 암 등과 같은 기저질환자들에게 감염률이 월등히 높다. 또한, 기저질환자가 기저질환이 없는 사람들보다 치명률이 훨씬 높다고 연구되었다. 본 연구에서는 뇌파데이터를 이용하여 기저질환의 유·무를 분류하였다. 기저질환자 유·무에 대한 분류를 위해 사용된 데이터는 데이터사이언스랩에서 제공하는 뇌파데이터로 33개의 특징과 69개의 샘플로 이루어졌다. 데이터의 전처리는 Z-score를 사용하였다. 분류는 뉴럴네트워크 인 NEWFM와 ZNN엔진을 사용하였다. 실험 결과 기저질환자의 유·무에 대한 분류결과 NEWFM은 77.94%, ZNN은 76.47%의 실험 결과를 얻었다. 이 연구를 통해 뇌파데이터를 측정하고 기저질환의 유무를 분류하고 높은 감염률을 보이는 기저질환자들이 COVID19로부터 예방 할 수 있으리라 기대한다. 이를 기반으로 향후 기저질환에 대한 세분류를 할 수 있는 연구가 필요하고, 각 기저질환이 전염병에 미치는 영향에 대해서도 연구가 필요하다.

Keywords

References

  1. Soo Lin, Hyunjung Lin, Jean-Pierre Despres, "Collateral Damage of the COVID-19 pandemic on Nutritional Quality and Physical Activity:Perspective from South Korea" volume 28, Issue 10. October2020. P,1788-1790. DOI: https://doi.org/10.1002/oby.22935
  2. Miyoung Choi,PhD, Ho Gyun Shin, MPH, Yan Jin, PhD, Joo Youn Kim, PhD, Kwang Hyub Han, MD, "COVID-19 Pandemic-Epidemiology" Joumal of the Korean Medical Association Aug2020;63(8):504-511. DOI: 10.5124/jkma.2020.63.8.504
  3. Sohaib. A.T, Qureshi. S, Hagelback. J, Hilborn. O, and Jercie, P, "Evaluating classifiers for emotion recognition using EEG," Foundations of Augmented Cognition Lecture Notes in Computer Science, vol. 8027, pp. 492-501, 2013. DOI: 10.1007/978-3-642-39454-6_53
  4. Duda. T. O, Hart. P. E, and Stork. D G, "Pattern classification, "Wiley-INterscience, 2000 DOI: https://doi.org/10.1007/BF01237942
  5. Landfield PW, McGaugh JL, Tusa RJ. "Theta Rhythm: A Temporal Correlate of Memory Storage Processes in the Rat" Science 1972:175:87-89 DOI: https:doi.org/10.1126/xcience.175.4017.87
  6. Ole Jensen, Jochen Kaiser Jean-Philippe Lachaux "Human gamma-frequency oscillations associated with attention and memory" Trends in Neurosciences, Volume 30, Issue7, P.317-324 DOI: https://doi.org/10.1016/j.tins.2007.05.001
  7. Hyunju Lee, Dongil Shin dongkyoo Shin "A research on the emotion classification and precision improvement of EEG(Electroencephalogram)data using machine learning algorithm" J,internet computer.Serv. ISSN 1598-0170 pp.27-36. 2019 DOI: 10.7472/jksii.2019.20.5.27
  8. Schnitzler a, Gross J. "Normal and pathological oscillatory communication in the brain." Nat Rev Neurosci 2005; 6:285-296. DOI: 10.1038/nrn1650
  9. Jensen O, Kaiser J, Lachaux JP. Human gamma-frequency oscillations associated with attention and memory. Trends Neurosci 2007;30:317-324. DOI:10.1016/j.tins.2007.05.001
  10. Alexander Craik, Yongtian He Jose L Contreras-Vidal, "Deep learning for electroencephalogram(EEG) classification tasks:a review April 2019. DOI:doi.org/10.1088/1741-2552/abOab5
  11. DataScience Lab
  12. S.H. Lee and J.S. Lim, Parkinson's disease classification using gait characteristics and wavelet-based feature extraction, Expert Syst. Appl. 39 (2012), 7338-7344 DOI:doi.org/10.1016/j.eswa.2012.01.084
  13. Heejin Yoon, Bohyun Wang and JoonS. Lim, Prediction of Time Series Microarray Data using Neurofuzzy Network, Indian Journal of Science and Technology, Vol8(26), IPL0485, October 2015. DOI:10.17485/ijst/2015/v8i26/80728
  14. Sang-Hong Lee, Hyoung J. Jang, Joon S.Lim "Forecasting Short-Term KOSPI Time Series Based on NEWFM" New Directions in intelligent Interactive Multimedia pp.175-184.2008. DOI:doi.org/10/1007/978-3-540-68127-4_18
  15. http://www.wikipedia.org