facebook
twitter
vk
instagram
linkedin
google+
tumblr
akademia
youtube
skype
mendeley
Wiki
Page translation
 

BRAIN DATA ANALYSIS AND MANAGEMENT

BRAIN DATA ANALYSIS AND MANAGEMENT
Ilieva Roumiana, assoc. prof. in automated systems for data processing and management, ph.d. of technical science, associate professor

Stanislava Asparuhova Petrova, meng student in electronic management, elfe

Technical University of Sofia, Bulgaria

Petia Georgieva, head of signal processing lab (spl), ieeta, ph.d. of technical science

University of Aveiro, Bulgaria

Conference participant

Machine Learning (ML) techniques have been extensively applied in bioinformatics. In this paper, we choose RapidMiner software to analyze brain data (EEG signals) in order to discriminate human emotions while subjects were observing images. Five ML classification algorithms were studied: k-Nearest Neighbor (kNN), Naive Bayes, Support Vector Machine, Artificial Neural Networks and Decision Tree. kNN and ensemble classifiers achieved above 80% accuracy with test data, and very encouraging results taking into account that brain signals are highly nonstationary and noisy and therefore quite challenging data for analysis and management.

 

I. Introduction

Machine Learning (ML) is a subfield of artificial intelligence that is concerned with the design, analysis, implementation, and applications of programs that learn from examples [2]. Learning from data is commercially and scientifically important. ML consists of methods that extract automatically interesting knowledge (patterns, models, relationships) in large databases.

The goal of this work is to find a reliable ML algorithm (or a combination of ML techniques) able to discriminate positive and negative human emotions based on Electroencephalogram (EEG) signals. The EEG data was collected while subjects were exposed to images which typically provoke positive and negative emotions.

This paper is organized as follows. Section 2 shortly introduces the ML techniques studied. Section 3 describes the dataset, the acquisition process and the metrics used to analyze the classifier performance. In Section 4 are summarized the classification results and finally our conclusions are presented in Section 5.

II. Machine Learning techniques

The ML classification techniques applied in the present study are briefly introduced.

1. K-Nearest Neighbor (k-NN)

K-NNis a widely applied classifier [1]. The class of a new example (object) is defined based on the majority vote of the K nearest training examples with known class labels (Fig.1). K is usually a small positive integer unpair number to avoid that two classes have the same votes. kNN computes the most common class among the k Nearest Neighbors of the new example andassigns this class to the example. For K=1 the object is simply assigned to the class of its nearest neighbor.

2. Support Vector Machine (SVM)

SVM is a classification technique that defines the hyperplane that maximizes the margin between two classes [1]. There may be several options (lines) that separate two classes (+/- on Fig. 2). SVM determines the closest objects between the two classes (termed support vectors) and the optimal separation line is the one that maximizes the distance between the classes. For linearly non separable classes (Fig. 3) the concept of kernel induced feature space is formulated. The Kernel SVM transforms the original data into a higher dimensional feature space where data is already linearly separable and then applies the same procedure as described above.

3. Artificial Neural Network(ANN)

ANN is a mathematical architecture inspired by the structure and functionality of biological neural networks [1]. An ANN consists of layers, typically one input, one hidden and one output layer (Fig. 4). Each layer has a number of parallel processing elements (PE), termed neurons (or nodes), which are mathematical functions that mimic at a macro scale the dynamical behavior of biological neurons. Due to their properties of adaptation, noise filtering and parallel processing, the ANNs are a powerful framework for classification and regression. Their main disadvantage is the usually long processing time for networks with high dimension (high number of PE).

 

4. Decision Tree (DT)

DT is a classification technique based on the principle of complex problem division into a sequence of subproblems and thus generating a decision tree (Fig.5).

III. EEG Signal Classification

1. EEG signals acquisition.

The goal of this study is to distinguish emotional bio-signals evoked by viewing selected affective pictures from the International Affective Picture System (IAPS) [5].EEG-based brain-computer interfaces consist of very typical components, each of which performs its own critical function. Figure 1 shows the process cycle. Firstly, a stimulus set and test protocol is needed (1). During testing (2), the test subject will be exposed to the stimuli according to the test protocol. The resulting voltage changes in the brain are then recorded (3) as an electroencephalogram, from which noise and artifacts are removed (4). The resulting data will be analyzed (5) and relevant features (like power spectra) will be computed. Based on a test set from these features a classi?er will be trained (6), and the rest of the data will be classi?ed using this classi?er. This step provides an interpretation of the original raw brain signals. The feedback step will not be used during this research, but is shown for completeness.

Figure 6: Brain Study Spiral Model

A total of 26 female volunteers participated in the study. 21 channels of EEG were recorded - Frontal and Parietal (FP), Frontal (F); Temporal (T), Central (C), Parietal (P) and Occipital (O) channels. A total of 24 of high arousal (>6) images with positive valence (7.29+/-0.65) and negative valence (1.47+/-0.24) were selected. Each image was presented 3 times in a pseudo- random order and each trial lasted 3500ms: during the first 750ms, a fixation crosses was presented, then one of the images during 500ms and at last a black screen during the 2250ms. Three schemes were implemented by choosing three different filters and detecting n maximums and n minimums at the output of the filters.

i) Butterworth filter of fourth order with passband [0.5 - 15]Hz. 12 features are stored according to the latency (time of occurrence) and amplitude of the first  3 maximums and minimums (Fig. 7a).

ii) Butterworth filter of fourth order with Delta band [0.5 - 4]Hz. 8 features are stored according to the latency and amplitude of the first two  maximums and minimums (Fig. 7b).

iii) Butterworth filter of 4th order with Theta band [4 - 8]Hz . 12 features are stored according to the latency and amplitude of the first 3 maximums and minimums (Fig. 7c).

 

2. Classifier performance metrics

The classifier performance is analyzed by the Confusion Matrix [3]. The basic structure of a Confusion Matrix for a two class problem is presented on Table 1.

Table 1: Confusion Matrix

Where TP (true positive) and TN (true negative) are the numbers of correctly classified positive and negative examples respectively, while FP (false positive) and FN (false negative) are the numbers of wrongly classified positive and negative examples respectively. The following performance measures are determined from the Confusion Matrix.

Accuracyis the fraction of all (positive and negative) correctly classified examples [3]:

Accuracy 

Precisionis the fraction of correctly classified positive examples from all classified as positive.  

Precision

Specificityis the fraction of correctly classified negative examples from all classified as negative.

Specificity

Recallis the fraction of positive examples correctly classified positive examples  from all positive examples.                               

Recall 

Whileaccuracy is a performance measure more typical for balanced data (data with similar number of examples of all classes), the other measures (precision, specificity and recall) are more adequate for unbalanced data.

IV. EEG classification results with Rapid Miner (RM)

RM is an open-source system for data mining [4]. It is available as stand-alone application for data analysis and as a data mining engine for integration into other products. RM is an environment for machine learning, data mining, text mining, predictive analytics and business analytics.

1. Data Preprocessing

Before information extraction and classification the data was conveniently preprocessed, filtered and stored as follows.

Data normalization: In order to escape computational problems before loading into RM.

Data Cleaning and storing: Fig. 8 depicts the process of i) data loading into RM with the operator ReadCSV, ii) filtering (removing) the lines with missing values (that correspond to lines with zeros) with the operator FilterExamples and iii) storing the cleaned data into BrainStudy Repository by the operator Store.

Figure 8: The process of data loading, filtering and storing.

The same process was applied for the three datasets corresponding to Filter 1, Filter2 and Filter3.

2. Classification based on all attributes and all channels

The goal is to find the best binary (two class) classifier based on all attributes and all channels. This scenario is implemented for three data sets Filter1 (ClassAB_Data), Filter2 (ClassABdelta_Data), Filter3 (ClassABteta_Data). The five classifiers were compared. On Figure 9 is summarized the process of simultaneous training and testing of the five classifiers with the same data set loaded by the operator Retrieve Data. The operator Multiply provides the same data to each classifier represented by the X-Validation operator.

Figure 9:  X-Validation of five classifiers applied to ClassAB_Data.

After playing the process, we obtained all values for accuracy, recall and precision from the confusion matrix. These performance measures are summarized in Table 2 for 5 classifiers, applied to 3 data sets. The highest classification rate is obtained for KNN classifier trained with CLassABdelta dataset.

Table 2: Performance measures of 5 classifiers, applied to 3 data sets.

C. Classifier optimization by feature selection of data set with all channels

The three datasets considered have between 8 and 12 features. Now we want to explore the possibility of improving the classification applying feature selection (reduction) procedures. Forward Selection and Backward Elimination are among the most typical operators in RM for extracting the most influential features.

KNN classifier optimization

In order to illustrate the plausibility of feature selection procedure, KNN classifier is optimized by both operators. The performance results in terms of confusion matrix and the associated values for accuracy, precision and recall are summarized in Fig. 10. Here, Forward Selection leads to higher accuracy compared to the Backward Elimination.

NN (neural network) classifier optimization

The effect of feature selection procedure depends strongly on the classification algorithm. On Figure 11 are presented the results of NN classifier optimization. In this case the latency 2 is the remover feature and the Backward Elimination leads to higher accuracy compared to the Forward Selection.

D.Classification based on all attributes and selected channels

Taking into account that Parietal and Occipital channels are responsible for visual processing we have isolated in Group1 only Parietal channels, in Group2 only Occipital channels and in Group3 the combination of Parietal and Occipital channels from the complete dataset ABdelta data (Table 3, Table 4, Table 5).The highest classification rate was obtained for KNN classifier trained with CLassABdelta dataset.

E. Ensemble classification

Ensemble classification uses a combination of n-learned classifiers M1, M2,…Mn, in order to build an improved composite model [2]. The ensemble method “Bagging” was applied in this study where the final result is obtained by the principle of majority vote.

Figure 12: Implementation of ensemble classification with Bagging operator.

On Fig. 12 is depicted the sequence of nested processes to implement ensemble bagging classification with the five classifiers. In the previous study, DT and NB classifiers exhibited worse performance (see Table 2), therefore they were removed. The results are summarized on Fig. 13 a) and b). The best ensemble classification (SVM, NN, KNN) does not provide significant advantages compared to the KNN from Table 3.

VI. Conclusions

In this paper a number of Machine Learning methods are studied and applied to a challenging classification problem of discriminatinghuman emotions based on EEG brain data. Among five classifiers, K-Nearest Neighbor (kNN) provided the best discrimination (84% accuracy) as an individual classification model. Ensemble classification (combination of SVM, NN and KNN classifiers) achieved slightly better results (85%). This study has shown that the preprocessing step on the row data collected from the EEG machine is crucial for extracting discriminative patterns. First, we need to choose the frequency band that adequately reflects the affective brain states. Our conclusions are that apparently [0.5-4] Hz is the most corresponding band. Then, the most suitable features need to be identified. The classification is clearly affected by the choice of temporal (selected amplitudes and latencies) and spatial (selected channels) features. Last but not least the classification algorithm is also important for the emotional patterns recognition. In case no one individual classifier is satisfactory, a mixture of weak classifiers (termed ensemble classification)   may be a reasonably good alternative. All required data analysis steps fit nicely to the modular structure of RM software platform and make the pipeline of procedures quite clear and intuitive. We are confident that the methodology studied in this paper can be easily adapted to other problems in bioinformatics or other private and public sectors, such as banking, insurance and medicine.

Acknowledgement

The research, described in this paper, was carried out within the framework of the contract № ИУНФ 13000. All experiments are conducted during Erasmus study period of the third author in the University of Aveiro (UA) under the supervision of the rest two authors. The Erasmus scholarship provided by Technical University of Sofia and the excellent working conditions in UA are highly acknowledged. We would like to express thanks to the PsyLab from UA, and particularly to Dr. Isabel Santos, for providing the data set.

 

References:

  1. Han, J., M. Kamber, Data Mining Concepts and Technique, University of Illinois at Urbana- Champaign, San Francisco, CA94111, Second Edition, 2006.
  2. Bramer, M., Principles of Data Mining, Springer, London, 2007.
  3. Pang-Ning T., M. Steinback, V. Kumar, Introduction to Data Mining, 2003.
  4. RapidMiner 5 software platform. Website: http://rapid-i.com/content/view/181/190/.
  5. Frantzidis, C., et al., “On the classification of emotional biosignals evoked while viewing affective pictures: An integrated Data-Mining-Based approach for Healthcare applications.
  6. Shoikova E., A. Peshev, M. Krumova, ”ePortfolio – Identity аnd Professional Development”, ICEIRD 2011, 5-7 May 2011, Ohrid, Macedonia, Proceedings CD, pp. 1061- 1068
  7. Nikolova, I., W. Jahn, Die Еinf?hrung und Кonkurenzvorteile des  TQM, Wissenschaftliche konferenz Innovationen und wettbewerbsf?higkeit,  Karlsruher Institut f?r Technologie, TU Sofia,TU Braunschweig, FOM Essen
  8. Tamosiuniene, R., K. Angelov, Project and Programme Management and Evaluation, С., 2011
Comments: 2

Chiglintsev Alexander

Уважаемые авторы! Для упрощения ситуации сбора данных вы оперировали только двумя категориями эмоций - положительными и отрицательными, и то вы при деятельности мозга столкнулись с трудностями в виде помех. Использовали метод ЭЭГ - очень широко применяемый в психофизиологических исследованиях и достаточно стандарными математическими способами. Исходя из вышеуказанного мне как практическому врачу интересно какое применение вы можете найти в медицине. На мой взгляд - только в психофизиологических академических исследований, но не в практической медицине. Так причем здесь выбранная вами тематика конференции.

Lakhtin Yuriy Vladimirovich

Dear author! It is difficult to assess your work. This may make the experts Biophysics Neuroscience. I wish you success in your studies. Regards, Yu. Lakhtin.
Comments: 2

Chiglintsev Alexander

Уважаемые авторы! Для упрощения ситуации сбора данных вы оперировали только двумя категориями эмоций - положительными и отрицательными, и то вы при деятельности мозга столкнулись с трудностями в виде помех. Использовали метод ЭЭГ - очень широко применяемый в психофизиологических исследованиях и достаточно стандарными математическими способами. Исходя из вышеуказанного мне как практическому врачу интересно какое применение вы можете найти в медицине. На мой взгляд - только в психофизиологических академических исследований, но не в практической медицине. Так причем здесь выбранная вами тематика конференции.

Lakhtin Yuriy Vladimirovich

Dear author! It is difficult to assess your work. This may make the experts Biophysics Neuroscience. I wish you success in your studies. Regards, Yu. Lakhtin.
PARTNERS
 
 
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
Would you like to know all the news about GISAP project and be up to date of all news from GISAP? Register for free news right now and you will be receiving them on your e-mail right away as soon as they are published on GISAP portal.