Annual Conference of KIPS (한국정보처리학회:학술대회논문집)

  • 2025.05a
  • /
  • Pages.463-465
  • /
  • 2025
  • /
  • 2005-0011(pISSN)
  • /
  • 2671-7298(eISSN)
Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

EEG Statistical Analysis of Coherence Patterns in Depression and Machine Classification

우울증의 일관성 패턴에 대한 뇌파 통계 분석 및 기계 학습 분류

  • Ngumimi Karen Iyortsuun (Department of Artificial Intelligence Convergence, Chonnam National University, South Korea University) ;
  • Soo-Hyung Kim (Department of Artificial Intelligence Convergence, Chonnam National University, South Korea University) ;
  • Soonja Yeom (School of Information and Communication Technology, University of Tasmania) ;
  • Hyung-Jeong Yang (Department of Artificial Intelligence Convergence, Chonnam National University, South Korea University) ;
  • Aera Kim (Department of Artificial Intelligence Convergence, Chonnam National University, South Korea University)
  • ;
  • 김수형 (전남대학교 대학원 인공지능융합학과) ;
  • 염순자 (호주 태즈메이니아 대학교, 정보통신기술대학 ) ;
  • 양형정 (전남대학교 대학원 인공지능융합학과) ;
  • 김애라 (전남대학교 대학원 인공지능융합학과)
  • Published : 2025.05.29
Download PDF
⟨ Previous Next ⟩

Abstract

Depression's neurophysiological basis remains poorly understood, particularly regarding functional connectivity patterns. This study examines EEG coherence differences between 46 major depressive disorder (MDD) patients and 75 healthy controls (HC) from the PRED+CT dataset. Statistical analysis revealed significant reductions in global coherence across all frequency bands in depression, with the strongest effects observed in theta and delta bands. Support Vector Machine classification achieved promising performance, with the best results in the beta band (accuracy=83.28) despite this band showing the smallest statistical effects. This discrepancy suggests that different aspects of coherence data may contribute to biomarker development versus neurophysiological understanding of depression.

Keywords

Acknowledgement

This work was supported by the Institute of Information & Communications Technology Planning & Evaluation (IITP) under the Artificial Intelligence Convergence Innovation Human Resources Development (IITP-2023-RS-2023-00256629) grant funded by the Korea government (MSIT), the ITRC (Information Technology Research Center) support program (IITP-2024-RS-2024-00437718) supervised by IITP, the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (RS-2023-00219107).

References

  1. Luo Y, Tang M, Fan X. Meta analysis of resting frontal alpha asymmetry as a biomarker of depression. npj Mental Health Research. 2025 Jan 17;4(1):2.
  2. Tan W, Liu Z, Xi C, Deng M, Long Y, Palaniyappan L, Yang J. Decreased integration of the frontoparietal network during a working memory task in major depressive disorder. Australian & New Zealand Journal of Psychiatry. 2021 Jun;55(6):577-87.
  3. Yang Y, Cui Q, Pang Y, Chen Y, Tang Q, Guo X, Han S, Fateh AA, Lu F, He Z, Huang J. Frequency-specific alteration of functional connectivity density in bipolar disorder depression. Progress in NeuroPsychopharmacology and Biological Psychiatry. 2021 Jan 10;104:110026.
  4. Cavanagh JF, Bismark AW, Frank MJ, Allen JJ. Multiple dissociations between comorbid depression and anxiety on reward and punishment processing: Evidence from computationally informed EEG. Computational Psychiatry (Cambridge, Mass.). 2019 Jan;3:1.
  5. Jackson-Koku G. Beck depression inventory. Occupational medicine. 2016 Mar 1;66(2):174-5.
  6. Christiano LJ, Fitzgerald TJ. The band pass filter. International economic review. 2003 May;44(2):435-65.
  7. Urigüen JA, Garcia-Zapirain B. EEG artifact removal—state-of-the-art and guidelines. Journal of neural engineering. 2015 Apr 2;12(3):031001.
  8. de Aguiar Neto FS, Rosa JL. Depression biomarkers using non-invasive EEG: A review. Neuroscience & Biobehavioral Reviews. 2019 Oct 1;105:83-93.
  9. Cavanagh JF, Frank MJ. Frontal theta as a mechanism for cognitive control. Trends in cognitive sciences. 2014 Aug 1;18(8):414-21. 
  10. Li L, Wang X, Li J, Zhao Y. An EEG-based marker of functional connectivity: detection of major depressive disorder. Cognitive Neurodynamics. 2024 Aug;18(4):1671-87. 
  11. Huang SS, Yu YH, Chen HH, Hung CC, Wang YT, Chang CH, Peng SJ, Kuo PH. Functional connectivity analysis on electroencephalography signals reveals potential biomarkers for treatment response in major depression. BMC psychiatry. 2023 Aug 1;23(1):554. 
  12. Fingelkurts AA, Fingelkurts AA, Rytsälä H, Suominen K, Isometsä E, Kähkönen S. Impaired functional connectivity at EEG alpha and theta frequency bands in major depression. Human brain mapping. 2007 Mar;28(3):247-61. 
  13. Wei L, Qiu T, Wang Z, Mei W. A functional connectivity metric method for EEG time series via nonlinear symbolization. Biomedical Signal Processing and Control. 2025 May 1;103:107498. 
  14. Suthaharan S. Support vector machine. InMachine learning models and algorithms for big data classification: thinking with examples for effective learning 2016 Jan (pp. 207-235). Boston, MA: Springer US. 
  15. Anguita D, Ghelardoni L, Ghio A, Oneto L, Ridella S. The'K'in K-fold Cross Validation. InESANN 2012 Apr 25 (Vol. 102, pp. 441-446). 
  16. Yacouby R, Axman D. Probabilistic extension of precision, recall, and f1 score for more thorough evaluation of classification models. InProceedings of the first workshop on evaluation and comparison of NLP systems 2020 Nov (pp. 79-91).