在使用ADJUST 和 ICMARC 的特徵算法從最後實驗結果可以發現到特徵對於訊號 之前的差異性在變異數分析的結果是具有參考性質的，實驗當中的赫斯特指數不管在分 類器特徵挑選上還是變異數分析中得到的結果是屬於偏差的，儘管在一些情況中有得到 分類效果的提升，但得到效益很低，那麼我們可以知道利用變異數分析可以明確的發現 這種分析法對於訊號特徵的分析結果是很好的，在不同的分類器上儘管訊號之間具有差 異的比較性，但分類器對於特徵的適合性也會造成不一樣的結果，在學習分類器上隨機 森林、支援向量機、樸素貝葉斯對於訊號這種特徵是屬於較好的方式，雖然最後最好的 結果中沒有最鄰近演算法，但從實驗表格中可以發現最鄰近的分類沒有最好但普遍看來 都是不錯的。在資料挑選的時候，雖然在ADUST 和 ICMARC 個別挑選出雜訊有時出現 錯誤，但兩種訊號分類器一起使用挑選出的結果幾乎都是好的，極少數部分才會出錯，
兩種分類的方式有所互補可以讓分類效果更好，但也會隨著時間效益變低，多個方法代 表特徵越多，那我們在使用檢定的方式來做特徵的挑選，可以得到更高效益的特徵，在 使用分類器檢測學習效果的好壞，利用這樣的方式使能做出更有效益及更準確的訊號分 類器。
43
參考文獻
[1] A. Mogonon, J. Jovicich, L. Bruzzone, and M. Buiatti “ADJUST: An automatic EEG artifact detector based on the joint use of spatial and temporal features” ,Psychophysiology, no.48, P.229–240,2011.
[2] L. F. Frølich” Statistical evaluation of features in classification problems with applications to detection of hypoglycemic conditions based on EEG data”, August 12, 2011.
[3] L. F. Frølich, T. S. ANDERSEN, and M. MøRUP,” Classification of independent components of EEG into multiple artifact classes”, Psychophysiology, no.52, pp.32–45, 2015.
[4] I. Winkler, S. Haufe, and M. Tangermann,”Automatic Classification of Artifactual ICA-Components for Artifact Removal in EEG Signals”, Behavioral and Brain Functions ,2011.
[5] F.C. Viola, J. Thorne, B. Edmonds, T. Schneider, T. Eichele, and S. Debener,” Semi-automatic identification of independent components representing EEG artifact”,Elsevier, Volume 120, Issue 5, Pages 868–877,2009.
[6] Immrama Institute. The International 10-20 System of Electrode Placement. Mar. 2010.
[7] CARRIE A. JOYCE, IRINA F. GORODNITSKY, and MARTA KUTAS , Automatic removal of eye movement and blink artifacts from EEG data using blind component separation, Psychophysiology, 41, 313–325,2004.
[8] Yuan Zou, Student Member, IEEE, Viswam Nathan, Student Member, IEEE, and Roozbeh Jafari, Senior Member, IEEE , Automatic Identification of Artifact-Related Independent Components for Artifact Removal in EEG Recordings, IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, VOL. 20, NO. 1, JANUARY 2016.
[9] Suguru Kanoga, Keio University, Yasue Mitsukura, Keio University, ICA-Based Positive Semidefinite Matrix Templates for Eye-Blink Artifact Removal from EEG Signal with
44
Single-Electrode.
[10] C.Y. Sai, N. Mokhtar, H. Arof, P. Cumming, and M. Iwahashi, Automated Classification and Removal of EEG Artifacts with SVM and Wavelet-ICA, 2016.
[11] Michael Tangermann et al. “Classification of artifactual ICA components”. In: Int J Bioelectromagnetism 11.2 , pp. 110–114,2009.
[12] L. Shoker, S. Sanei, and J. Chambers. “Artifact removal from electroencephalograms using a hybrid BSS-SVM algorithm”. In: Signal Processing Letters, IEEE 12.10 pp. 721 – 724,2005.
[13] Maria Anastasiadou, Avgis Hadjipapas, Manolis Christodoulakis, Eleftherios S.
Papathanasiou, Savvas S. Papacostas and Georgios D. Mitsis, ”Detection and Removal of Muscle artifacts from Scalp EEG Recordings in Patients with Epilepsy,” IEEE 14th International Conference on Bioinformatics and Bioengineering,2014.
[14] Saleha Khatun, Ruhi Mahajan, and Bashir I. Morshed,” Comparative Analysis of Wavelet Based Approaches for Reliable Removal of Ocular Artifacts from Single Channel EEG”.
[15] S. Jirayucharoensak P. Israsena, Automatic Removal of EEG Artifacts Using ICA and Lifting Wavelet Transform, 2013 International Computer Science and Engineering Conference (ICSEC): ICSEC 2013 English Track Full Papers.
[16] Aapo Hyvarinen. Independent component analysis: a tuturial. May 2010.
[17] Arnaud Delorme and Scott Makeig. “EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis”. In: Journal of Neuroscience Methods 134, pp. 9–21.2004.
[18] Anusha Zachariah, Jinu Jai,and Geevarghese Titus, “Automatic EEG Artifact Removal by Independent Component Analysis Using Critical EEG Rhythms”, 2013 International Conference on Control Communication and Computing.
[19] Gang Wang, Member, IEEE, Chaolin Teng, Kuo Li, Zhonglin Zhang, and Xiangguo Yan,
45
“The Removal of EOG Artifacts From EEG Signals Using Independent Component Analysis and Multivariate Empirical Mode Decomposition”, IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, VOL. 20, NO. 5, SEPTEMBER 2016.
[20] C.Y. Sai, N. Mokhtar, H. Arof, P. Cumming and M. Iwahashi, Automated Classification and Removal of EEG Artifacts with SVM and Wavelet-ICA, Citation information: DOI 10.1109/JBHI.2017.2723420, IEEE Journal of Biomedical and Health Informatics
[21] Breiman, “RANDOM FORESTS”, Statistics Department University of California Berkeley, CA 94720 January 2001.
[22] Breiman, “Statistical Modeling: The Two Cultures”, Statistical Science, Vol. 16, No. 3 (Aug., 2001), pp. 199-215.
[23] Sebastian Mika, Gunnar Rätsch, Jason Weston, Bernhard Schölkopf, and Klaus-Robert Müller,” Fisher Discriminant Analysis With Kernels ”,1999.
[24] Schölkopf, Bernhard, Burges, Christopher J.C. ,and Smola, “Advances in Kernel Methods:
Support Vector Learning”, MIT Press, Cambridge, MA, 1999.
[25] Harry Zhang,” The Optimality of Naive Bayes”, Faculty of Computer Science University of New Brunswick Fredericton, New Brunswick, Canada E3B 5A3.
[26] Begg, C. B. and Gray, R.” Calculation of polychotomous logistic regression parameters using individualized regressions”, Biometrika, vol.71, pp.11-18,1984.
[27] N. S. Altman,” An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression”, Pages 175-185 ,1990.