Please wait a minute...
Journal of Integrative Neuroscience  2018, Vol. 17 Issue (1): 11-18    DOI: 10.31083/JIN-170033
Research article Previous articles | Next articles
Functional analysis of ADHD in children using nonlinear features of EEG signals
Shiva Khoshnoud1, Mohammad Ali Nazari1, Mousa Shamsi1, *()
1 Biomedical Engineering Faculty, Sahand University of Technology, Tabriz, Iran;
2 Cognitive Neuroscience Laboratory, Department of Psychology, University of Tabriz, Tabriz, Iran
Download:  PDF(394KB)  ( 970 ) Full text   ( 37 )
Export:  BibTeX | EndNote (RIS)      

Attention deficit hyperactivity disorder is a neurodevelopmental condition associated with varying levels of hyperactivity, inattention, and impulsivity. This study investigated brain function in children with attention deficit hyperactivity disorder using measures of nonlinear dynamics in electroencephalogram signals during rest. During eyes-closed resting, 19 channel electroencephalogram signals were recorded from 12 ADHD and 12 normal age-matched children. The multifractal singularity spectrum, the largest Lyapunov exponent, and approximate entropy were employed to quantify the chaotic nonlinear dynamics of these electroencephalogram signals. As confirmed by Wilcoxon rank sum test, the largest Lyapunov exponent over left frontal-central cortex exhibited a significant difference between attention deficit hyperactivity disorder subjects and the age-matched control groups. Further, mean approximate entropy was significantly lower in attention deficit hyperactivity disorder subjects in prefrontal cortex. The singularity spectrum was also considerably altered in attention deficit hyperactivity disorder subjects when compared to control children. Evaluation of these features was performed with two classifiers: a support vector machine and a radial basis function neural network. For better comparison, subject classification based on frequency band power was assessed using the same types of classifiers. Nonlinear features provided better discrimination between attention deficit hyperactivity disorder and control than band power features. Under four-fold cross-validation testing, the support vector machine gave 83.33% accurate classification results.

Key words:  Attention deficit hyperactivity disorder      EEG signals      largest Lyapunov exponent      approximate entropy      multifractal DFA      classification     
Submitted:  23 January 2017      Accepted:  08 May 2017      Published:  15 February 2018     
*Corresponding Author(s):  Mousa Shamsi     E-mail:

Cite this article: 

Shiva Khoshnoud, Mohammad Ali Nazari, Mousa Shamsi. Functional analysis of ADHD in children using nonlinear features of EEG signals. Journal of Integrative Neuroscience, 2018, 17(1): 11-18.

URL:     OR

Fig.1.  Overall procedure for functional brain dynamic analysis of ADHD and control children using nonlinear dynamical features of EEG signals. First step: Signal was preprocessed to remove any noise and artefacts. Followed by calculation of LLE, ApEn, and multifractal spectra for every channel of the recorded EEG signals. In parallel with nonlinear features, frequency band power estimates were also calculated. Subsequently, the principal component coefficients of these features were computed and the 15 largest components of each feature category employed as final features. Finally, subject group classification was implemented with two classifiers: a support vector machine classifier and a radial basis function neural network.

Fig.2.  Results of the Wilcoxon rank sum test for the mean ApEn feature between ADHD and control group at channels: (a) Fp1, (b) Fp2. Mean ApEn in subjects with ADHD was significantly lower than control in these two scalp electrodes.

Fig.3.  Black disks indicate channels with multifractal spectrum height differences (p-value < 0.05) between ADHD and normal subjects. The multifractal spectrum for ADHD subjects was significantly higher than the control group over these scalp electrodes.

Fig.4.  Group classification using frequency band powers.

Table 1  Classification using nonlinear dynamical properties
Classifier Nonlinear features
Largest Approximate Multi fractal
Lyapunov entropy spectrum
SVM 70.83 59 79.17
RBF 79.17 75 66.67
Table 2  Classification using combined features
Band power features Nonlinear features
Classifier Without With PCA Without With PCA
PCA (%) (%) PCA (%) (%)
SVM 79.17 79.17 83.33 83.33
RBF 83.33 70.83 70.83 66.67
[1] Sadock BJ, Kaplan HI, Sadock VA ( 2007) Kaplan & Sadock’s Synopsis of Psychiatry: Behavioral Sciences/Clinical Psychiatry (10th ed.). Philadelphia, PA, Lippincott Williams & Wilkins.
[2] Jukić V, Arbanas G ( 2014) Diagnostic and Statistical Manual of Mental Disorders (DSM-5®). Jastrebarsko, Naklada Slap.
[3] Ahmadlou M, Adeli H ( 2010) Wavelet-synchronization methodology: a new approach for EEG-based diagnosis of ADHD. Clinical EEG & Neuroscience 41( 1), 1-10.
doi: 10.1177/155005941004100103 pmid: 20307009
[4] Ahmadlou M, Adeli H ( 2011) Fuzzy synchronization likelihood with application to attention-deficit/hyperactivity disorder. Clinical EEG & Neuroscience 42( 1), 6-13.
[5] Ahmadlou M, Adeli H ( 2011) Functional community analysis of brain: A new approach for EEG-based investigation of the brain pathology. Neuroimage 58( 2), 401-408.
doi: 10.1016/j.neuroimage.2011.04.070 pmid: 21586331
[6] Adámek V, Valeš F ( 2012) Graph theoretical analysis of organization of functional brain networks in ADHD. Clinical EEG & Neuroscience 43( 1), 5-13.
doi: 10.1177/1550059411428555 pmid: 22423545
[7] Barry RJ, Clarke AR, Johnstone SJ ( 2003) A review of electrophysiology in attention-deficit/hyperactivity disorder: I. Qualitative and quantitative electroencephalography. Clinical Neurophysiology 114( 2), 171-183.
doi: 10.1016/S1388-2457(02)00362-0
[8] Barry RJ, Clarke AR, Mccarthy R, Selikowitz M ( 2002) EEG coherence in attention-deficit/hyperactivity disorder: a comparative study of two DSM-IV types. Clinical Neurophysiology 113( 4), 579-585.
doi: 10.1016/S1388-2457(02)00036-6 pmid: 11956003
[9] Ghassemi F, Hassan Moradi M, Tehrani-Doost M, Abootalebi V ( 2012) Using non-linear features of EEG for ADHD/normal participants’ classification. Procedia-Social and Behavioral Sciences 32, 148-152.
doi: 10.1016/j.sbspro.2012.01.024
[10] Lazzaro I, Gordon E, Whitmont S, Meares R, Clarke S ( 2001) The modulation of late component event related potentials by pre-stimulus EEG theta activity in ADHD. International Journal of Neuroscience 107( 3, 4), 247-264.
doi: 10.3109/00207450109150688 pmid: 11328694
[11] Lazzaro I, Gordon E, Whitmont S, Plahn M, Li W, Clarke S, Dosen A, Meares R ( 1998) Quantified EEG activity in adolescent attention deficit hyperactivity disorder. Clinical Electroencephalogr 29( 1), 37-42.
doi: 10.1177/155005949802900111 pmid: 9472424
[12] Lubar JF ( 1991) Discourse on the development of EEG diagnostics and biofeedback for attention-deficit/hyperactivity disorders. Biofeedback and Self-regulation 16( 3), 201-205.
doi: 10.1007/BF01000016 pmid: 1932259
[13] Mann CA, Lubar JF, Zimmerman AW, Miller CA, Muenchen RA ( 1992) Quantitative analysis of EEG in boys with attention-deficit-hyperactivity disorder: Controlled study with clinical implications. Pediatric Neurology 8( 1), 30-36.
doi: 10.1016/0887-8994(92)90049-5 pmid: 1558573
[14] Sohn H, Kim I, LeeW, Peterson BS, Hong H, Chae JH, Hong S, Jeong J ( 2010) Linear and non-linear EEG analysis of adolescents with attentiondeficit/ hyperactivity disorder during a cognitive task. Clinical Neurophysiology Official Journal of the International Federation of Clinical Neurophysiology 121( 11), 1863-1870.
doi: 10.1016/j.clinph.2010.04.007 pmid: 20659814
[15] Swartwood JN, Swartwood MO, Lubar JF, Timmermann DL ( 2003) EEG differences in ADHD-combined type during baseline and cognitive tasks. Pediatric Neurology 28( 3), 199-204.
doi: 10.1016/S0887-8994(02)00514-3 pmid: 12770673
[16] Tcheslavski GV, Beex AA ( 2006) Phase synchrony and coherence analyses of EEG as tools to discriminate between children with and without attention deficit disorder. Biomedical Signal Processing & Control 1( 2), 151-161.
doi: 10.1016/j.bspc.2006.08.001
[17] Stam CJ ( 2005) Nonlinear dynamical analysis of EEG and MEG: review of an emerging field. Clinical Neurophysiology 116( 10), 2266-2301.
doi: 10.1016/j.clinph.2005.06.011 pmid: 16115797
[18] Dutta S, Ghosh D, Samanta S, Dey S ( 2014) Multifractal parameters as an indication of different physiological and pathological states of the human brain. Physica A Statistical Mechanics & Its Applications 396( 2), 155-163.
doi: 10.1016/j.physa.2013.11.014
[19] Zorick T, Mandelkern MA ( 2013) Multifractal Detrended Fluctuation Analysis of Human EEG: Preliminary Investigation and Comparison with the Wavelet Transform Modulus Maxima Technique. Plos One 8( 7), e68360.
doi: 10.1371/journal.pone.0068360 pmid: 3700954
[20] Weiss B, Clemens Z, Bódizs R, Vágó Z, Halász P ( 2009) Spatiotemporal analysis of monofractal and multifractal properties of the human sleep EEG. Journal of Neuroscience Methods 185( 1), 116-124.
doi: 10.1016/j.jneumeth.2009.07.027 pmid: 19646476
[21] Fetterhoff D, Kraft RA, Sandler RA, Ioan O, Sexton CA, Marmarelis VZ, Hampson RE, Deadwyler SA ( 2015) Distinguishing cognitive state with multifractal complexity of hippocampal interspike interval sequences. Frontiers in Systems Neuroscience 9, 130.
doi: 10.3389/fnsys.2015.00130 pmid: 4585000
[22] Gradshteyn IS, Ryzhik IM ( 2014) Table of Integrals, Series, and Products. San Diego, Academic Press.
[23] Rosenstein MT, Collins JJ, Luca CJD ( 1993) A practical method for calculating largest Lyapunov exponents from small data sets. Physica D-nonlinear Phenomena 65( 1, 2), 117-134.
doi: 10.1016/0167-2789(93)90009-P
[24] Takens F ( 1981) Detecting strange attractors in turbulence. In, D. Rand and LS. Young(Eds) Dynamical Systems and Turbulence(pp. 366-381), Groningen, Rijksuniversiteit Groningen. Mathematisch Instituut.
[25] Kennel MB, Brown R, Abarbanel HD ( 1992) Determining embedding dimension for phase-space reconstruction using a geometrical construction . Physical review A, Atomic, molecular, and optical physics 45( 6), 3403-3411.
[26] Pincus SM ( 1991) Approximate entropy as a measure of system complexity. Proceedings of the National Academy of Sciences of the United States of America 88( 6), 2297-2301.
doi: 10.1073/pnas.88.6.2297
[27] Sanei S, Chambers JA ( 2013) EEG Signal Processing. Somerset, Wiley.
[28] Kantelhardt JW, Zschiegner SA, Koscielny-Bunde E, Havlin S, Bunde A, Stanley HE ( 2002) Multifractal detrended fluctuation analysis of nonstationary time series. Physica A Statistical Mechanics & Its Applications 316( 1-4), 87-114.
doi: 10.1016/S0378-4371(02)01383-3
[29] Semmlow JL, Griffel B ( 2009) Biosignal and Medical Image Processing. Boca Raton, CRC Press.
[30] Vapnik VN, Lerner AY ( 1963) Recognition of Patterns with help of Generalized Portraits. Avtomat i Telemekh 24( 6), 774-780.
[31] Vapnik VN ( 1999) The Nature of Statistic Learning Theory. New York, Springer.
[32] Schwenker F, Kestler HA, Palm G ( 2001) Three learning phases for radial-basis-function networks. Neural Networks 14( 4), 439-458.
doi: 10.1016/S0893-6080(01)00027-2 pmid: 11411631
[33] Burioka N, Cornélissen G, Maegaki Y, Halberg F, Kaplan DT, Miyata M, Fukuoka Y, Endo M, Suyama H, Tomita Y ( 2005) Approximate entropy of the electroencephalogram in healthy awake subjects and absence epilepsy patients. Clinical EEG & Neuroscience 36( 3), 188-193.
doi: 10.1177/155005940503600309 pmid: 16128154
[34] Ikawa M, Nakanishi M, Furukawa T, Nakaaki S, Hori S, Yoshida S ( 2010) Relationship between EEG dimensional complexity and neuropsychological findings in Alzheimer’s disease. Psychiatry & Clinical Neurosciences 54( 5), 537-541.
doi: 10.1046/j.1440-1819.2000.00749.x pmid: 11043802
[35] Clarke AR, Barry RJ, Mccarthy R, Selikowitz M ( 2010) Electroencephalogram differences in two subtypes of attentiondeficit/ hyperactivity disorder. Psychophysiology 38( 2), 212-221.
doi: 10.1111/1469-8986.3820212 pmid: 11347867
[36] Aman CJ, Roberts RJ, Pennington BF ( 1998) A neuropsychological examination of the underlying deficit in attention deficit hyperactivity disorder: frontal lobe versus right parietal lobe theories. Developmental Psychology 34( 5), 956-969.
doi: 10.1037//0012-1649.34.5.956 pmid: 9779742
[1] Rekha Sahu, Satya Ranjan Dash, Lleuvelyn A Cacha, Roman R Poznanski, Shantipriya Parida. Epileptic seizure detection: a comparative study between deep and traditional machine learning techniques[J]. Journal of Integrative Neuroscience, 2020, 19(1): 1-9.
[2] Fatemeh Zareayan Jahromy, Atena Bajoulvand, Mohammad Reza Daliri. Statistical algorithms for emotion classification via functional connectivity[J]. Journal of Integrative Neuroscience, 2019, 18(3): 293-297.
[3] Matteo Fraschini, Margherita Lai, Luca Didaci. Comparison of functional connectivity metrics using an unsupervised approach: a source resting-state EEG study[J]. Journal of Integrative Neuroscience, 2018, 17(4): 393-396.
[4] Xian Liu, Chunyu Zhao, Chengxia Sun, Zhanming Chen, Tasawar Hayat, Ahmed Alsaedi. A new closed-loop strategy for detection and modulation of epileptiform spikes based on cross approximate entropy[J]. Journal of Integrative Neuroscience, 2018, 17(3): 271-280.
[5] Ivan Grbatinić, Nebojsa Milosevic, Dusica Maric. Translaminar neuromorphotopological clustering and classification of dentate nucleus neurons[J]. Journal of Integrative Neuroscience, 2018, 17(2): 105-124.
No Suggested Reading articles found!