Neural Computing and Applications

, Volume 27, Issue 6, pp 1607–1616 | Cite as

Artificial intelligence approach to classify unipolar and bipolar depressive disorders

  • Turker Tekin Erguzel
  • Gokben Hizli Sayar
  • Nevzat Tarhan
Original Article


Machine learning approaches for medical decision-making processes are valuable when both high classification accuracy and less feature requirements are satisfied. Artificial neural networks (ANNs) successfully meet the first goal with its adaptive engine, while nature-inspired algorithms are focusing on the feature selection (FS) process in order to eliminate less informative and less discriminant features. Besides engineering applications of ANN and FS algorithms, medical informatics is another emerging field using similar methods for medical data processing. Classification of psychiatric disorders is one of the major focus of medical informatics using artificial intelligence approaches. Being one of the most debilitating psychiatric diseases, bipolar disorder (BD) is frequently misdiagnosed as unipolar disorder (UD), leading to suboptimal treatment and poor outcomes. Thus, discriminating UD and BD at earlier stages of illness could therefore help to facilitate efficient and specific treatment. The use of quantitative electroencephalography (EEG) cordance as a biomarker has greatly enhanced the clinical utility of EEG in psychiatric and neurological subjects. In this context, the paper puts forward a study using two-step hybridized methodology: particle swarm optimization (PSO) algorithm for FS process and ANN for training process. The noteworthy performance of ANN–PSO approach stated that it is possible to discriminate 31 bipolar and 58 unipolar subjects using selected features from alpha and theta frequency bands with 89.89 % overall classification accuracy.


Artificial intelligence Artificial neural network Particle swarm optimization Unipolar and bipolar disorders 



The authors would like to express their thanks to NPIstanbul Hospital for providing the required EEG data.


  1. 1.
    Merikangas KR, Akiskal HS, Ankst J et al (2007) Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry 64(5):543–552CrossRefGoogle Scholar
  2. 2.
    Angst J, Azorin JM, Browden CL et al (2012) Diagnostic criteria for bipolarity based on an international sample of 5,635 patients with DSM-IV major depressive episodes. Eur Arch Psychiatry 262(1):3–11Google Scholar
  3. 3.
    Bowden CL (2010) Diagnosis, treatment, and recovery maintenance in bipolar depression. J Clin Psychiatry 71(1):e01CrossRefGoogle Scholar
  4. 4.
    Goldberg JF, Harrow M, Whiteside JE (2001) Risk for bipolar illness in patients initially hospitalized for unipolar depression. Am J Psychiatry 158(8):1265–1270CrossRefGoogle Scholar
  5. 5.
    Bowden CL (2001) Strategies to reduce misdiagnosis of bipolar depression. Psychiatr Serv 52(1):51–55CrossRefGoogle Scholar
  6. 6.
    Lee PS, Chen YS, Hsieh JC et al (2010) Distinct neuronal oscillatory responses between patients with bipolar and unipolar disorders: a magnetoencephalographic study. J Affect Disord 123(1–3):270–275CrossRefGoogle Scholar
  7. 7.
    Hirschfeld RM, Calabrese JR, Weissman MM et al (2003) Screening for bipolar disorder in the community. J Clin Psychiatry 64(1):53–59CrossRefGoogle Scholar
  8. 8.
    Phillips ML, Frank E (2006) Redefining bipolar disorder: toward DSM-V. Am J Psychiatry 163:1135–1136CrossRefGoogle Scholar
  9. 9.
    Almeida JR, Versace A, Mechelli A et al (2009) Abnormal amygdala-prefrontal effective connectivity to happy faces differentiates bipolar from major depression. Biol Psychiatry 66:451–459CrossRefGoogle Scholar
  10. 10.
    Lawrence NS, Williams AM, Surguladze S et al (2004) Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry 55:578–587CrossRefGoogle Scholar
  11. 11.
    Phillips ML, Drevets WC, Rauch SL et al (2003) Neurobiology of emotion perception II: implications for major psychiatric disorders. Biol Psychiatry 54:515–528CrossRefGoogle Scholar
  12. 12.
    Ritchie MD, White BC, Parker JS et al (2003) Optimization of neural network architecture using genetic programming improves detection and modeling of gene-gene interactions in studies of human diseases. BMC Bioinformatics 4:28CrossRefGoogle Scholar
  13. 13.
    Freitas AA (2002) A survey of evolutionary algorithms for data mining and knowledge discovery. In: Ghosh A, Tsutsui S (eds) Advances in evolutionary computation. Springer, Berlin, pp 819–845Google Scholar
  14. 14.
    Pena-Reyes CA, Sipper M (2000) Evolutionary computation in medicine: an overview. Artif Intell Med 9(1):1–23CrossRefGoogle Scholar
  15. 15.
    Chang YH, Zheng B, Wang XH et al (1999) Computer-aided diagnosis of breast cancer using artificial neural networks: comparison of backpropagation and genetic algorithms. Proceedings of the International Joint Conference on Neural Networks, Washington, DC, USA, vol 5. IEEE Press, Washington (DC), pp 3674–3679Google Scholar
  16. 16.
    Koza JR (1992) Genetic programming: on the programming of computers by means of natural selection. MIT Press, Cambridge (MA)zbMATHGoogle Scholar
  17. 17.
    Tan KC, Tay A, Lee TH et al (2002) Mining multiple comprehensible classification rules using genetic programming. IEEE Congress on Evolutionary Computation. Honolulu, HI, pp 1302–1307Google Scholar
  18. 18.
    Witten IH, Frank E (1999) Data mining: practical machine learning tools and techniques with Java implementations. Kaufmann (Morgan), San Francisco (CA)Google Scholar
  19. 19.
    Wong ML, Lam W, Leung KS et al (2000) Discovering knowledge from medical databases using evolutionary algorithms. IEEE Eng Med Biol Mag 19(4):45–55CrossRefGoogle Scholar
  20. 20.
    Leslie S (1987) Neurometric quantitative EEG features of depressive disorders. In: Takahashi R, Flor Henry P, Gruzier J, Niwa S (eds) Cerebral dynamics, laterality and psychopathology. Elsevier, Amsterdam, pp 1–17Google Scholar
  21. 21.
    Lucek P, Hanke J, Reich J et al (1998) Multi-locus nonparametric linkage analysis of complex trait loci with neural networks. Hum Hered 48:275–284CrossRefGoogle Scholar
  22. 22.
    Ottenbacher KJ, Smith PM, Illig SB et al (2001) Comparison of logistic regression and neural networks to predict rehospitalization in patients with stroke. J Clin Epidemiol 54(11):1159–1165CrossRefGoogle Scholar
  23. 23.
    Delen D, Walker G, Kadam A (2005) Predicting breast cancer survivability: a comparison of three data mining methods. Artif Intell Med 34(2):113–127CrossRefGoogle Scholar
  24. 24.
    Jaimes F, Farbiarz J, Alvarez D et al (2005) Comparison between logistic regression and neural networks to predict death in patients with suspected sepsis in the emergency room. Crit Care 9(2):150–156CrossRefGoogle Scholar
  25. 25.
    Adam P, Pawel M, Jaroslaw J (2012) Comparison of evolutionary computation techniques for noise injected neural network training to estimate longitudinal dispersion coefficients in rivers. Expert Syst Appl 39:1354–1361CrossRefGoogle Scholar
  26. 26.
    Guyon I, Gunn S, Zadeh LA (2006) Feature extraction, foundations and applications. Springer, BerlinCrossRefzbMATHGoogle Scholar
  27. 27.
    Hassan R, Othman RM, Saad P, Kasim S (2011) A compact hybrid feature vector for an accurate secondary structure prediction. Inf Sci 181:5267–5277CrossRefGoogle Scholar
  28. 28.
    Maldonado S, Weber R, Basak J (2011) Kernel-penalized SVM for feature selection. Inf Sci 181:115–128CrossRefGoogle Scholar
  29. 29.
    Satchidananda D, Royb R, Choc SB et al (2012) An improved swarm optimized functional link artificial neural network (ISO-FLANN) for classification. J Syst Eng 85:1333–1345Google Scholar
  30. 30.
    Young RC, Biggs JT, Ziegler E et al (1978) A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry 133:429–435CrossRefGoogle Scholar
  31. 31.
    Leuchter AF, Uijtdehaage SH, Cook IA et al (1999) Relationship between brain electrical activity and cortical perfusion in normal subjects. Psychiatry Res 90(2):125–140CrossRefGoogle Scholar
  32. 32.
    Hughes JR, John ER (1999) Conventional and quantitative electroencephalography in psychiatry. J Neuropsychiatry Clin Neurosci 11:190–208CrossRefGoogle Scholar
  33. 33.
    Niedermeyer E, Silva L (2004) Electroencephalography. Basic principles, clinical applications, and related fields, 5th edn. Lippincott Williams & Wilkins, PhiladelphiaGoogle Scholar
  34. 34.
    Leuchter AF, Cook IA, Lufkin RB et al (1994) Cordance: a new method for assessment of cerebral perfusion and metabolism using quantitative electroencephalography. Neuroimage 1:208–219CrossRefGoogle Scholar
  35. 35.
    Nuwer MR, Lehmann D, da Silva FL et al (1999) IFCN guidelines for topographic and frequency analysis of EEGs and EPs. Electroencephalogr Clin Neurophysiol 52:15–20Google Scholar
  36. 36.
    Hjorth B (1975) An on-line transformation of EEG scalp potentials into orthogonal source derivations. Electroencephalogr Clin Neurophysiol 39:526–530CrossRefGoogle Scholar
  37. 37.
    Cook IA, O’Hara R, Uijtdehaage S et al (1998) Assessing the accuracy of topographic EEG mapping for determining local brain function. Electroencephalogr Clin Neurophysiol 107:404–414CrossRefGoogle Scholar
  38. 38.
    Bares M, Novak T, Brunovsky M et al (2012) The change of QEEG prefrontal cordance as a response predictor to antidepressive intervention in bipolar depression. A pilot study. J Psychiatr Res 46:219–225CrossRefGoogle Scholar
  39. 39.
    Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182zbMATHGoogle Scholar
  40. 40.
    Gheyas IA, Smith LS (2010) Feature subset selection in large dimensionality domains. Pattern Recogn 43:5–13CrossRefzbMATHGoogle Scholar
  41. 41.
    Blum A, Langley P (1997) Selection of relevant features and examples in machine learning. Artif Intell 97(1–2):245–271MathSciNetCrossRefzbMATHGoogle Scholar
  42. 42.
    Dash M, Liu H (1997) Feature selection for classification. Intell Data Anal 1:131–156CrossRefGoogle Scholar
  43. 43.
    Chuang LY, Tsai SW, Yang CH (2011) Improved binary particle swarm optimization using catfish effect for feature selection. Expert Syst Appl 38:12699–12707CrossRefGoogle Scholar
  44. 44.
    Kennedy J, Eberhart RC (1995) Particle swarm optimization. In: Proceedings of the 1995 IEEE International Conference on Neural Networks, IEEE, Piscataway, pp 1942–1948Google Scholar
  45. 45.
    Shi Y, Eberhart R (1998) A modified particle swarm optimizer. In: International Conference on Evolutionary Computation, IEEE, pp 69–73Google Scholar
  46. 46.
    Lin SW, Chen SC (2009) PSOLDA: a particle swarm optimization approach for enhancing classification accuracy rate of linear discriminant analysis. Appl Soft Comput 9:1008–1015CrossRefGoogle Scholar
  47. 47.
    Peer ES, Van Den Bergh F, Engelbrecht AP (2003) Using neighbourhood with the guaranteed convergence PSO. In: Proceedings of the 2003 IEEE Swarm Intelligence Symposium, pp 235–242Google Scholar
  48. 48.
    Jakob SV, Jacques R (2002) Particle swarms extensions for improved local, multimodal, and dynamic search in numerical optimization. M.Sc. Thesis, Department of Computer Science, Aarhus University, Aarhus C, DenmarkGoogle Scholar
  49. 49.
    Ratnaweera A, Halgamuge SK, Watson CH (2004) Self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients. IEEE Trans Evol Comput 8(3):240–255CrossRefGoogle Scholar
  50. 50.
    Pasupuleti S, Battiti R (2006) The gregarious particle swarm optimizer (G-PSO). In: Eighth Annual Conference on Genetic and Evolutionary Computation, pp 67–74Google Scholar
  51. 51.
    Kennedy J, Mendes R (2002) Population structure and particle swarm performance. Proc Congr Evol Comput 2:1671–1676Google Scholar
  52. 52.
    Zhang W, Xie X (2003) DEPSO: hybrid particle swarm with differential evolution operator. In: IEEE International Conference on Systems, Man and Cybernetics, Washington DC, pp 3816–3821Google Scholar
  53. 53.
    Atyabi A, Samadzadegan S (2011) Particle swarm optimization: a survey, applications of swarm intelligence. Nova Scientific Publishers, Hauppauge, pp 167–178Google Scholar
  54. 54.
    Atyabi A, Luerssen MH, Powers DM (2013) PSO-based dimension reduction of EEG recordings: implications for subject transfer in BCI. Neurocomputing 119:319–331CrossRefGoogle Scholar
  55. 55.
    Marren A, Harston C, Pap R (1990) Handbook of neural computating applications. Academic Press Inc., San DiegoGoogle Scholar
  56. 56.
    Principe JC, Euliano NR, Lefebvre WC (2000) Neural and adaptive systems: fundamentals through simulations. Wiley, New YorkGoogle Scholar
  57. 57.
    Kunhimangalama R, Ovallath S, Joseph PK (2013) Computer aided diagnostic problem solving: identification of peripheral nerve disorders. IRBM 34:244–251CrossRefGoogle Scholar
  58. 58.
    Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by error propagation. In: Rumelhart DE, McClelland JL (eds) Parallel distributed processing: explorations in the microstructure of cognition, vol 1. The MIT Press, Cambridge, pp 318–362Google Scholar
  59. 59.
    Asensio-Cuesta S, Diego-Mas JA, Alcaide-Marzal J (2010) Applying generalised feed forward neural networks to classifying industrial jobs in terms of risk of low back disorders. Int J Ind Ergon 40:629–635CrossRefGoogle Scholar
  60. 60.
    Kaladjian A, Jeanningros R, Azorin JM et al (2009) Reduced brain activation in euthymic bipolar patients during response inhibition: an event-related fMRI study. Psychiatry Res 173(1):45–51CrossRefGoogle Scholar
  61. 61.
    Langenecker SA, Kennedy SE, Guidotti LM et al (2007) Frontal and limbic activation during inhibitory control predicts treatment response in major depressive disorder. Biol Psychiatry 62(11):1272–1280CrossRefGoogle Scholar
  62. 62.
    Anand A, Li Y, Wang Y (2009) Resting state corticolimbic connectivity abnormalities in unmedicated bipolar disorder and unipolar depression. Psychiatry Res 171(3):189–198CrossRefGoogle Scholar
  63. 63.
    Lawrence SN, Williams AM, Surguladze S et al (2004) Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry 55:578–587CrossRefGoogle Scholar
  64. 64.
    Taylor JV, Clark L, Furey ML et al (2008) Neural basis of abnormal response to negative feedback in unmedicated mood disorders. Neuroimage 42:1118–1126CrossRefGoogle Scholar
  65. 65.
    Sheline YI, Price JL, Yan Z (2010) Resting-state functional MRI in depression unmasks increased connectivity between networks via the dorsal nexus. Proc Natl Acad Sci 107(24):11020–11025CrossRefGoogle Scholar
  66. 66.
    Marchand WR, Lee JN, Johnson S (2013) Differences in functional connectivity in major depression versus bipolar II depression. J Affect Disord 150(2):527–532CrossRefGoogle Scholar
  67. 67.
    Bares M, Novak T, Brunovsky M (2012) The change of QEEG prefrontal cordance as a response predictor to antidepressive intervention in bipolar depression. A pilot study. J Psychiatr Res 46(2):219–225CrossRefGoogle Scholar
  68. 68.
    Brooks JO, Wang PW, Bonner JC et al (2009) Decreased prefrontal, anterior cingulate, insula, and ventral striatal metabolism in medication-free depressed outpatients with bipolar disorder. J Psychiatry Res 43:181–188CrossRefGoogle Scholar
  69. 69.
    Hosokawa T, Momose T, Kasai K (2009) Brain glucose metabolism difference between bipolar and unipolar mood disorders in depressed and euthymic states. Prog Neuropsychopharmacol 33:243–250CrossRefGoogle Scholar
  70. 70.
    Drevets WC, Price JL, Furey ML (2008) Brain structural and functional abnormalities in mood disorders: implications for neurocircuitry models of depression. Brain Struct Funct 213:93–118CrossRefGoogle Scholar
  71. 71.
    Savitz J, Drevets WC (2009) Bipolar and major depressive disorder: neuroimaging the developmental-degenerative divide. Neurosci Biobehav R 33:699–771CrossRefGoogle Scholar
  72. 72.
    Leuchter AF, Cook IA, Hamilton SP et al (2010) Biomarkers to predict antidepressant response. Curr Psychiatry Rep 12:553–562CrossRefGoogle Scholar
  73. 73.
    Bares M, Brunovsky M, Kopecek M et al (2008) Early reduction in prefrontal theta QEEG cordance value predicts response to venlafaxine treatment in patients with resistant depressive disorder. Eur Psychiatry 23:350–355CrossRefGoogle Scholar
  74. 74.
    Bares M, Brunovsky M, Novak T et al (2010) The change of prefrontal QEEG theta cordance as a predictor of response to bupropion treatment in patients who had failed to respond to previous antidepressant treatments. Eur Neuropsychopharmacol 20:459–466CrossRefGoogle Scholar
  75. 75.
    Kopecek M, Tislerova B, Sos P et al (2008) QEEG changes during switch from depression to hypomania: a case report. Neuroendcrinol Lett 29:295–302Google Scholar
  76. 76.
    Noonan SK, Haist F, Muller RA (2009) Aberrant functional connectivity in autism: evidence from low frequency BOLD signal fluctuations. Brain Res 1262:48–63CrossRefGoogle Scholar
  77. 77.
    Clementz BA, Sponheim SR, Iacono WG, Beiser M (1994) Resting EEG in first episode schizophrenia patients, bipolar psychosis patients, and their first-degree relatives. Psychophysiology 31:486–494CrossRefGoogle Scholar
  78. 78.
    Degabriele R, Lagopoulos J (2009) A review of EEG and ERP studies in bipolar disorder. Acta Neuropsychiatr 21:58–66CrossRefGoogle Scholar
  79. 79.
    Tas C, Cebi M, Tan O, Hizli Sayar G, Tarhan N, Brown EC (2015) EEG power, cordance and coherence differences between unipolar and bipolar depression. J Affect Disord 172:185–190CrossRefGoogle Scholar
  80. 80.
    Brooks J, Po WW, Ketter T (2010) Functional brain imaging studies in bipolar disorder: focus on cerebral metabolism and blood flow. In: Yatham LN, Maj M (eds) Bipolar disorder. Wiley, Chichester, pp 200–209CrossRefGoogle Scholar
  81. 81.
    Haldane M, Frangou S (2006) Functional neuroimaging studies in mood disorders. Acta Neuropsychiatr 18:88–99CrossRefGoogle Scholar
  82. 82.
    Culha AF, Osman O, Dogangun Y et al (2008) Changes in regional cerebral blood flow demonstrated by 99mTc-HMPAO SPECT in euthymic bipolar patients. Eur Arch Psychiatry Clin Neurosci 258:144–151CrossRefGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2015

Authors and Affiliations

  1. 1.Department of Computer Engineering, Faculty of Engineering and Natural SciencesUskudar UniversityUskudar/IstanbulTurkey
  2. 2.Department of PsychiatryNPIstanbul HospitalIstanbulTurkey
  3. 3.Department of Psychology, Faculty of Humanities and Social SciencesUskudar UniversityIstanbulTurkey

Personalised recommendations