Görsel Uyaranlara İlişkin Manyetoensefalografi Sinyallerinin Genelleştirilmiş Regresyon Sinir Ağı ile Sınıflandırılması
Amaç: Bu çalışmanın amacı, beyin aktivitesini çözmek için Manyetoensefalografi (MEG) sinyallerini yapay sinir ağı ilesınıflandırmaktır.Yöntemler: MEG sinyallerini sınıflandırmak için Genelleştirilmiş Regresyon Sinir Ağı (GRSA) kullanılmıştır.Riemannian yaklaşımı ile sinyallerin öznitelikleri çıkarılmış ve 10 katlı çapraz doğrulama tekniği ile GRSA’nındoğruluğu hesaplanmıştır.Bulgular: Çalışmada 9 kız, 7 erkek bireye ait 306 kanaldan kaydedilen MEG verileri kullanılmıştır. Her bireye yaklaşık588 uyaran gösterilmiştir ve böylece tüm veri seti 9414 uyarandan oluşmaktadır. Ortalama spesifite, ortalamaduyarlılık ve ortalama sınıflandırma doğruluğu sırasıyla %75,43, %82,57 ve %79 olarak elde edilmiştir. Bu çalışma veaynı MEG veri setini kullanan diğer çalışmalar tarafından elde edilen sınıflandırma doğrulukları karşılaştırmalı olaraksunulmuştur.Sonuç: GRSA’nın MEG sinyallerinin sınıflandırılmasında kullanılan mevcut yöntemlere başarılı bir alternatifoluşturduğu düşünülmektedir.
Classification of Magnetoencephalography Signals Regarding Visual Stimuli by Generalized Regression Neural Network
Objective: The aim of this study is to classify the magnetoencephalography (MEG) signals with artificial neural network to solve brain activity. Methods: The Generalized Regression Neural Network (GRNN) was used to classify MEG signals. The features of the signals were extracted by the Riemannian approach and the accuracy of the GRNN was calculated by the 10-fold cross validation technique. Results: In the study, MEG data recorded from 306 channels belonging to 7 male subjects and 9 female subjects were used. Approximately 588 stimuli were shown to each individual, so the entire data set is composed of 9414 stimuli. Mean specificity, mean sensitivity and mean classification accuracy were obtained 75.43%, 82.57% and 79%, respectively. The classification accuracies obtained by this study and other studies for same MEG dataset were presented comparatively. Conclusion: GRNN is thought to be a successful alternative to existing methods for classifying MEG signals.
___
- Bascil MS, Tesneli AY, Temurtas F. A study on analog
and digital EEG signal filtering for brain computer
ınterfaces (BCI). Electronic Letters on
Science&Engineering 2014; 10: 1-10.
- Olivetti E, Kia SM, Avesani P. MEG decoding across
subjects. International Workshop on Pattern
Recognition in Neuroimaging 2014; doi:
10.1109/PRNI.2014.6858538.
- Caliskan A, Yuksel ME, Badem H, Basturk A. A deep
neural network classifier for decoding human brain
activity based on Magnetoencephalography.
Elektronika ir Elektrotechnika 2017; 23: 63-7.
- Cetin O, Temurtas F. Öğrenmeli vektör kuantalama ile
beyin bilgisayar arayüzü üzerine bir çalışma. Sakarya
University Journal of Computer and Information
Sciences 2018; 1: 1 7.
- Gulbag A, Temurtas F. A study on quantitative
classification of binary gas mixture using neural
networks and adaptive neuro-fuzzy inference systems.
Sensors and Actuators B 2006; 115: 252-62.
- Abadi MK, Subramanian R, Kia SM, Avesani P, Patras I,
Sebe N. DECAF: MEG-based multimodal database for
decoding affective physiological responses. IEEE Trans.
Affective Computing 2015; 6: 209-22.
- Chan AM, Halgren E, Marinkovic K, Cash SS. Decoding
word and category-specific spatiotemporal
representations from MEG and EEG. NeuroImage 2011;
54: 3028-39.
- Daliri MR. A hybrid method for the decoding of spatial
attention using the MEG brain signals. Biomedical
Signal Processing and Control 2014; 10: 308-12.
- Kia SM, Pedregosa F, Blumenthal A, Passerini A. Grouplevel
spatio-temporal pattern recovery in MEG
decoding using multi-task joint feature learning.
Journal of Neuroscience Methods 2017; 285: 97-108.
- Huttunen H, Kauppi JP, Tohka J. Regularized logistic
regression for mind reading with parallel validation.
ICANN 2011 International Conference on Artificial
Neural Networks, Pattern Analysis, Statistical
Modelling and Computational Learning2 MEG MindReading
2011; 20-24.
- Kia SM, Vega Pons S, Weisz N, Passerini A.
Interpretability of multivariate brain maps in linear
brain decoding: Definition, and heuristic quantification
in multivariate analysis of MEG time-locked effects.
Frontiers in Neuroscience 2017; 10: 619; doi:
10.3389/fnins.2016.00619
- Redcay E, Carlson TA. Rapid neural discrimination of
communicative gestures. Social Cognitive and Affective
Neuroscience 2015; 10: 545-51.
- Ursulean R, Lazar AM. Detrended cross-correlation
analysis of biometric signals used in a new
authentication method. Elektronika ir Elektrotechnika
2009; 89: 55-8.
- Raudonis V, Narvydas G, Simutis R. A classification of
flash evoked potentials based on artificial neural
network. Elektronika ir Elektrotechnika 2008; 81: 31-6.
- Barachant A, Bonnet S, Congedo M, Jutten C.
Multiclass brain–computer interface classification by
Riemannian geometry. IEEE Trans. Biomedical
Engineering 2012; 59: 920-28.
- Yger F, Berar M, Lotte F. Riemannian approaches in
brain computer interfaces: a review. IEEE Trans. Neural
Systems and Rehabilitation Engineering 2017; 25:
1753-62.
- Henson RN, Wakeman DG, Litvak V, Friston KJ. A
parametric empirical Bayesian framework for the
EEG/MEG inverse problem: generative models for
multi-subject and multi-modal integration. Frontiers in
Human Neuroscience 2011; 5: 76.
- Cetin O, Temurtas F, Gulgonul S. An application of
multilayer neural network on hepatitis disease
diagnosis using approximations of sigmoid activation
function. Dicle Medical Journal 2015; 42: 150-7.
- Ulaşabileceği adres:
https://www.kaggle.com/c/decoding-the-human-brain
- Moakher M. A differential geometric approach to the
geometric mean of symmetric positive-definite
matrices. SIAM Journal on Matrix Analysis and
Applications 2005; 26: 735-47.
- Ulaşabileceği adres:
https://github.com/alexandrebarachant/covariancetoolbox
- Pennec X, Fillard P, Ayache N. A Riemannian
framework for tensor computing. International Journal
of Computer Vision 2006; 66: 41-66.
- Specht DF. A general regression neural network. IEEE
Transactions on Neural Networks 1991; 2: 568-76.
- Specht DF, Shapiro PD. Generalization accuracy of
probabilistic neural networks compared with
backpropagation networks. IJCNN-91-Seattle
International Joint Conference on Neural Networks
1991;887-92; doi: 10.1109/IJCNN.1991.155296.
- Bascil MS, Cetin O, Er O, Temurtas F. Olasılıksal sinir
ağının (PNN) parkinson hastalığının teşhisinde
kullanılması. Electronic Letters on Science &
Engineering 2012; 8: 1-10.
- Gorur K, Bozkurt MR, Bascil MS, Temurtas F.
Glossokinetic potential based tongue-machine interface
for 1-D extraction. Australas Phys Eng Sci Med 2018;
41: 379-91.
- Bascil MS, Oztekin H. A study on hepatitis disease
diagnosis using probabilistic neural network. Journal of
Medical Systems 2012; 36: 1603-6.