Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/484607
Title: | Development of an approach to study the effect of alternative and complementary therapy on human brain using EEG |
Researcher: | Sharma, Himika |
Guide(s): | Juneja, Mamta |
Keywords: | Digital signal processing EEG Human brain Therapy |
University: | Panjab University |
Completed Date: | 2022 |
Abstract: | The combined effect of yoga and SK meditation on brain signals was investigated in this study. Some of the parameters such as Shannon Entropy (SEn), Renyi Entropy (REn), Maximum (max), Minimum (min), energy, power, hurst component, fractal dimensions, detrended fluctuation analysis and autoregressive parameters have been calculated from each sub-band. One-way Analysis of Variance (ANOVA) has been used to estimate the significant difference of aforementioned parameters before and after daily exercise of combined yoga and SK for 3 months duration. Holm bonferroni adjustment was applied to all extracted p values. The extracted parameters stated the positive effect on human brain after regular practice of combined SK and yoga for 3 months. Results revealed that there is an enhancement in relaxation, concentration, vigilance, positive affect, positive cognition and decrement in chaotic activities of human brain. Furthermore, the signal was decomposed into six sub-bands {(0 4) Hz, (4 8) Hz, (8 16) Hz, (16 32) Hz, (32 64) Hz, (64 128) Hz} by incorporating Discrete Wavelet Transform (DWT). Daubechies (db4) wavelet was used for exploration, along with statistical features such as Variance (Var), Standard Deviation (SD), Kurtosis (kur), Zero Crossing (ZC), max and min were calculated from each sub-band. The Kruskal-Wallis (KW) statistical test was used to validate the derived parameters. The above-mentioned statistical parameters were used to categorise subjects as meditators and non-meditators using Machine Learning (ML) methods such as decision trees, discriminant analysis, logistic regression, Support Vector Machine (SVM), Weighted KNearest Neighbour (KNN), ensemble classifiers, and Artificial Neural Network (ANN). The experimental results showed that the SVM approach had the highest classification accuracy when compared to other classification methods, and that it may be used to build a precise classification system for detecting meditators and non-meditators. |
Pagination: | xii, 121p. |
URI: | http://hdl.handle.net/10603/484607 |
Appears in Departments: | University Institute of Engineering and Technology |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 106.18 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 1.46 MB | Adobe PDF | View/Open | |
03_chapter1.pdf | 315.15 kB | Adobe PDF | View/Open | |
04_chapter2.pdf | 332.97 kB | Adobe PDF | View/Open | |
05_chapter3.pdf | 910.58 kB | Adobe PDF | View/Open | |
06_chapter4.pdf | 1.17 MB | Adobe PDF | View/Open | |
07_chapter5.pdf | 734.75 kB | Adobe PDF | View/Open | |
08_annexures.pdf | 1.08 MB | Adobe PDF | View/Open | |
80_recommendation.pdf | 128.97 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: