Deep Learning-Based Artificial Intelligence Can Differentiate Treatment-Resistant and Responsive Depression Cases with High Accuracy

被引:2
作者
Metin, Sinem Zeynep [1 ]
Uyulan, Caglar [2 ]
Farhad, Shams [3 ]
Erguzel, Tuerker Tekin [4 ]
Turk, Omer [5 ]
Metin, Baris [6 ]
Cerezci, Onder [7 ]
Tarhan, Nevzat [1 ]
机构
[1] Uskudar Univ, Dept Psychiat, Istanbul, Turkiye
[2] Katip Celebi Univ, Dept Mech Engn, Izmir, Turkiye
[3] Uskudar Univ, Dept Neurosci, Istanbul, Turkiye
[4] Uskudar Univ, Fac Engn & Nat Sci, Dept Software Engn, Istanbul, Turkiye
[5] Mardin Artuklu Univ, Dept Comp Technol, Mardin, Turkiye
[6] Uskudar Univ, Med Fac, Neurol Dept, Istanbul, Turkiye
[7] Uskudar Univ, Fac Hlth Sci, Dept Physioterapy & Rehabil, Istanbul, Turkiye
关键词
depression; treatment-resistant depression; deep learning; convolutional neural network; EEG; electroencephalography; CONNECTIVITY; MANAGEMENT; ASYMMETRY; MARKER;
D O I
10.1177/15500594241273181
中图分类号
R74 [神经病学与精神病学];
学科分类号
摘要
Background: Although there are many treatment options available for depression, a large portion of patients with depression are diagnosed with treatment-resistant depression (TRD), which is characterized by an inadequate response to antidepressant treatment. Identifying the TRD population is crucial in terms of saving time and resources in depression treatment. Recently several studies employed various methods on EEG datasets for automatic depression detection or treatment outcome prediction. However, no previous study has used the deep learning (DL) approach and EEG signals for detecting treatment resistance. Method: 77 patients with TRD, 43 patients with non-TRD, and 40 healthy controls were compared using GoogleNet convolutional neural network and DL on EEG data. Additionally, Class Activation Maps (CAMs) acquired from the TRD and non-TRD groups were used to obtain distinctive regions for classification. Results: GoogleNet classified the healthy controls and non-TRD group with 88.43%, the healthy controls and TRD subjects with 89.73%, and the TRD and non-TRD group with 90.05% accuracy. The external validation accuracy for the TRD-non-TRD classification was 73.33%. Finally, the CAM analysis revealed that the TRD group contained dominant features in class detection of deep learning architecture in almost all electrodes. Limitations: Our study is limited by the moderate sample size of clinical groups and the retrospective nature of the study. Conclusion: These findings suggest that EEG-based deep learning can be used to classify treatment resistance in depression and may in the future prove to be a useful tool in psychiatry practice to identify patients who need more vigorous intervention.
引用
收藏
页码:119 / 130
页数:12
相关论文
共 61 条
  • [1] Automated EEG-based screening of depression using deep convolutional neural network
    Acharya, U. Rajendra
    Oh, Shu Lih
    Hagiwara, Yuki
    Tan, Jen Hong
    Adeli, Hojjat
    Subha, D. P.
    [J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2018, 161 : 103 - 113
  • [2] Motor imagery EEG signal classification using image processing technique over GoogLeNet deep learning algorithm for controlling the robot manipulator
    Ak, Ayca
    Topuz, Vedat
    Midi, Ipek
    [J]. BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2022, 72
  • [3] Frontal EEG asymmetry as a promising marker of depression vulnerability: summary and methodological considerations
    Allen, John J. B.
    Reznik, Samantha J.
    [J]. CURRENT OPINION IN PSYCHOLOGY, 2015, 4 : 93 - 97
  • [4] Alsharman N., 2020, J Comput Sci, V16, P620, DOI [10.3844/jcssp.2020.620.625, DOI 10.3844/JCSSP.2020.620.625, 10.3844/jcssp.2020.620]
  • [5] Conceptual and empirical comparison of dimensionality reduction algorithms (PCA, KPCA, LDA, MDS, SVD, LLE, ISOMAP, LE, ICA, t-SNE)
    Anowar, Farzana
    Sadaoui, Samira
    Selim, Bassant
    [J]. COMPUTER SCIENCE REVIEW, 2021, 40
  • [6] High-Gamma: A biological marker for suicide attempt in patients with depression
    Arikan, Mehmet Kemal
    Gunver, Mehmet Guven
    Tarhan, Nevzat
    Metin, Baris
    [J]. JOURNAL OF AFFECTIVE DISORDERS, 2019, 254 : 1 - 6
  • [7] Automated Depression Detection Using Deep Representation and Sequence Learning with EEG Signals
    Ay, Betul
    Yildirim, Ozal
    Talo, Muhammed
    Baloglu, Ulas Baran
    Aydin, Galip
    Puthankattil, Subha D.
    Acharya, U. Rajendra
    [J]. JOURNAL OF MEDICAL SYSTEMS, 2019, 43 (07)
  • [8] Responders to rTMS for depression show increased fronto-midline theta and theta connectivity compared to non-responders
    Bailey, N. W.
    Hoy, K. E.
    Rogasch, N. C.
    Thomson, R. H.
    McQueen, S.
    Elliot, D.
    Sullivan, C. M.
    Fulcher, B. D.
    Daskalakis, Z. J.
    Fitzgerald, P. B.
    [J]. BRAIN STIMULATION, 2018, 11 (01) : 190 - 203
  • [9] Risk factors for treatment resistance in unipolar depression: A systematic review
    Bennabi, D.
    Aouizerate, B.
    El-Hage, W.
    Doumy, O.
    Moliere, F.
    Courtet, P.
    Nieto, I.
    Bellivier, F.
    Bubrovsky, M.
    Vaiva, G.
    Holztmann, J.
    Bougerol, T.
    Richieri, R.
    Lancon, C.
    Camus, V.
    Saba, G.
    Haesbaert, F.
    d'Amato, T.
    Charpeaud, T.
    Llorca, P. M.
    Leboyer, M.
    Haffen, E.
    [J]. JOURNAL OF AFFECTIVE DISORDERS, 2015, 171 : 137 - 141
  • [10] Cai HS, 2016, IEEE INT C BIOINFORM, P1239, DOI 10.1109/BIBM.2016.7822696