Attention-Based Convolutional Recurrent Deep Neural Networks for the Prediction of Response to Repetitive Transcranial Magnetic Stimulation for Major Depressive Disorder

被引:15
作者
Shahabi, Mohsen Sadat [1 ]
Shalbaf, Ahmad [1 ]
Nobakhsh, Behrooz [1 ]
Rostami, Reza [2 ]
Kazemi, Reza [3 ]
机构
[1] Shahid Beheshti Univ Med Sci, Sch Med, Dept Biomed Engn & Med Phys, Tehran, Iran
[2] Univ Tehran, Dept Psychol, Tehran, Iran
[3] Inst Cognit Sci Studies, Dept Cognit Psychol, Tehran, Iran
关键词
Transfer learning; convolutional neural networks; long-short term memory; attention mechanism; treatment outcome prediction; repetitive transcranial magnetic stimulation; THETA CONNECTIVITY; BRAIN; EEG; COMPLEXITY; DIAGNOSIS; GRAPH; RTMS;
D O I
10.1142/S0129065723500077
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Repetitive Transcranial Magnetic Stimulation (rTMS) is proposed as an effective treatment for major depressive disorder (MDD). However, because of the suboptimal treatment outcome of rTMS, the prediction of response to this technique is a crucial task. We developed a deep learning (DL) model to classify responders (R) and non-responders (NR). With this aim, we assessed the pre-treatment EEG signal of 34 MDD patients and extracted effective connectivity (EC) among all electrodes in four frequency bands of EEG signal. Two-dimensional EC maps are put together to create a rich connectivity image and a sequence of these images is fed to the DL model. Then, the DL framework was constructed based on transfer learning (TL) models which are pre-trained convolutional neural networks (CNN) named VGG16, Xception, and EfficientNetB0. Then, long short-term memory (LSTM) cells are equipped with an attention mechanism added on top of TL models to fully exploit the spatiotemporal information of EEG signal. Using leave-one subject out cross validation (LOSO CV), Xception-BLSTM-Attention acquired the highest performance with 98.86% of accuracy and 97.73% of specificity. Fusion of these models as an ensemble model based on optimized majority voting gained 99.32% accuracy and 98.34% of specificity. Therefore, the ensemble of TL-LSTM-Attention models can predict accurately the treatment outcome.
引用
收藏
页数:13
相关论文
共 68 条
  • [1] Abdelhameed AM, 2018, IEEE INT NEW CIRC, P139, DOI 10.1109/NEWCAS.2018.8585542
  • [2] 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
  • [3] A Novel Depression Diagnosis Index Using Nonlinear Features in EEG Signals
    Acharya, U. Rajendra
    Sudarshan, Vidya K.
    Adeli, Hojjat
    Santhosh, Jayasree
    Koh, Joel E. W.
    Puthankatti, Subha D.
    Adeli, Amir
    [J]. EUROPEAN NEUROLOGY, 2015, 74 (1-2) : 79 - 83
  • [4] Computer-Aided Diagnosis of Depression Using EEG Signals
    Acharya, U. Rajendra
    Sudarshan, Vidya K.
    Adeli, Hojjat
    Santhosh, Jayasree
    Koh, Joel E. W.
    Adeli, Amir
    [J]. EUROPEAN NEUROLOGY, 2015, 73 (5-6) : 329 - 336
  • [5] Complexity of weighted graph: A new technique to investigate structural complexity of brain activities with applications to aging and autism
    Ahmadlou, Mehran
    Adeli, Hojjat
    [J]. NEUROSCIENCE LETTERS, 2017, 650 : 103 - 108
  • [6] Complexity of functional connectivity networks in mild cognitive impairment subjects during a working memory task
    Ahmadlou, Mehran
    Adeli, Anahita
    Bajo, Ricardo
    Adeli, Hojjat
    [J]. CLINICAL NEUROPHYSIOLOGY, 2014, 125 (04) : 694 - 702
  • [7] Spatiotemporal Analysis of Relative Convergence of EEGs Reveals Differences Between Brain Dynamics of Depressive Women and Men
    Ahmadlou, Mehran
    Adeli, Hojjat
    Adeli, Amir
    [J]. CLINICAL EEG AND NEUROSCIENCE, 2013, 44 (03) : 175 - 181
  • [8] Fractality analysis of frontal brain in major depressive disorder
    Ahmadlou, Mehran
    Adeli, Hojjat
    Adeli, Amir
    [J]. INTERNATIONAL JOURNAL OF PSYCHOPHYSIOLOGY, 2012, 85 (02) : 206 - 211
  • [9] [Anonymous], 2017, P IEEE C COMP VIS PA
  • [10] Real-time epileptic seizure recognition using Bayesian genetic whale optimizer and adaptive machine learning
    Anter, Ahmed M.
    Abd Elaziz, Mohamed
    Zhang, Zhiguo
    [J]. FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2022, 127 : 426 - 434