**5. Conclusion**

*Epilepsy - Advances in Diagnosis and Therapy*

3 Spatial

4 Frequency

5 Frequency

*Summary of EEG analyses.*

domain (EEG segments)

domain DWT (db1,db2)

domain db2 at level 4

**Analysis Database LoD Number of** 

*LoD, Level of decomposition; Tr, Training.*

*Time complexity of EEG analyses.*

6 Db2 at level 4 8 statistical features, 4

have been examined for seizure detection.

calculations, the analyses prove that the performance of SVM with significant features is good when compared with ANN using large number of features as the input. The major contributions of these analyses in view of the existing work are as follows:

1.Two different machine learning algorithms (ANN and SVM) that are based on two different learning methods (error correcting and associative learning)

2.Unique set of features are extracted from the EEG signals for classification.

3.For optimization, genetic algorithm is used for feature selection and proved

4.Accuracies are calculated for raw EEG signal and for all decomposed signals

(GLCM and statistical) SVM (Linear

**Number of features significant**

4 12 12 SVM — 0.09

 Bonn University 3 10 6 ANN 243 0.123 Bonn University 3 10 6 SVM 64 0.023 Bonn University 4 6 6 ANN 184 0.2 Bonn University 4 6 6 SVM 61 0.02 Real-time data 4 16 12 SVM 63 0.0013 Real-time data 4 8 4 SVM 62 0.0011 Validation 4 8 4 SVM — 0.017

and RBF kernel)

ANN and SVM

Linear kernel (99.95%

Db2 wavelet and hybrid features to SVM classifier are the best outcomes (92.16%

accuracy (99.9 accuracy)

successful result with 90% performance accuracy in system validation

accuracy)

accuracy)

SVM Renyi entropy gives better

SVM Relevant features give

**Classifier Tr time** 

**(seconds)**

**Test Time (seconds)**

that the classifier can perform well with relevant features.

**EEG analysis #: Domain Feature extraction Classifier Conclusion** 1 Spatial domain GLCM features ANN 85% Accuracy 2 Spatial domain GLCM features SVM 90% Accuracy

> GLCM, statistical, and hybrid features

Entropy estimation (Shannon, Renyi, and

GLCM features, 4 Renyi entropy estimation (Genetic algorithm for feature selection)

Tsallis)

**features**

**204**

**Table 6.**

**Table 5.**

An epileptic seizure is a symptom due to abnormal and excessive irregular neuronal activity in the brain. EEG test is mainly used for diagnosing epilepsy. EEG includes different types of waveforms with different frequency, amplitude, and spatial distribution. Traditional ways of computations would be less efficient for problem-solving. But, soft computing methods can work in an efficient way for discovering solutions from the given data. Components of soft computing are essential for developing automated expert systems. Early diagnosis of disease can save the life of a person. The approved CAD system is able to provide accurate results. Problem-solving is a challenging task for intelligent entities. It has been proved that "a machine can learn new things." It can adapt to new situations and has an ability to learn from the storage information. Supervised learning technique is used in majority of analyses. Fuzzy logic gives multi-value answers, whereas in machine learning, the system learns from data especially with the control or supervisor. In computational intelligence, evolutionary algorithms are inspired by biological systems and give optimal solution for the problem. "Clean data are greater than more data." Machine learns from data. Quality of data is important rather than quantity of data. This chapter gave an introduction about the components of soft computing and classification in machine learning. From the review of analyses, this chapter concludes that relevant features and less number of features can make the classifier perform well. Accuracies are compared in all decomposed signals and proved that level 4 of decomposition is enough for EEG signal classification. At level 4, the lower frequencies (delta and theta) can be analyzed perfectly because seizures occur mostly at lower frequencies. Also, from the analyses, it has been proved that the time required and memory space for data parameters are less.
