Academic Journal of Computing & Information Science, 2022, 5(8); doi: 10.25236/AJCIS.2022.050815.
Weihao Zhang1, Changyu Fan2, Jicheng Xie2, Xinyu Zhang2
1School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, 210023, China
2School of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, 210023, China
This paper is an attempt to distinguish depressed patients from healthy samples mainly by analyzing magnetoencephalogram data of depressed patients and healthy people under different emotional picture stimuli. The following conclusions are mainly drawn: First, the relationship of information storage and condition entropy is obtained firstly through the formula derivation, and the higher the information storage value, the lower the condition entropy. The healthier the biological system, the higher the complexity. Second, the conditional entropy values of the magnetoencephalography channel were mostly higher in the healthy samples than in the depressed patients under the emotional picture stimulation. Accordingly, the conditional entropy values of the frontal region under negative stimulation were lower in depressed patients than in healthy subjects. Third, under both the AR model and the ARFI model, the magnetoencephalogram information storage values were higher in depressed patients than in healthy samples. The difference was more pronounced especially near the frontal regions for both. The variability of the results obtained under the ARFI model was even more pronounced when depressed patients and healthy samples were more discriminated near the left frontal region than the right frontal region, as revealed by the study of line plots of information storage values across left and right frontal regions channels.
Depression; ARFI; MEG; Conditional Entropy; AR
Weihao Zhang, Changyu Fan, Jicheng Xie, Xinyu Zhang. The AR and ARFI-Based Models Utility of Magnetoencephalography Assisted Diagnosis of Depression. Academic Journal of Computing & Information Science (2022), Vol. 5, Issue 8: 101-107. https://doi.org/10.25236/AJCIS.2022.050815.
 F. Zhang, Y. Wang, et al. “Statistical analysis of 5109 cases of central nervous system tumors by who new classification (in chinese)”, Journal of clinical and Experimental Pathology, 2004, vol. 20, pp. 688-691.
 R. Lang. “Research on segmentation, classification and recognition technology of midbrain tumors in computer-aided diagnosis (in chinese)”, Beijing University of technology, 2018.
 N. S. Zulpe, V. P. Pawar. “GLCM textural features for Brain Tumor Classification.” International Journal of Computer Science Issues, 2012, vol. 9, pp. 354-359.
 J. Sachdeva, V. Kumar, I. Guptas, et al. “Segmentation, feature extraction, and multiclass brain tumor classification.” Journal of Digital Imaging, 2013, vol. 26, pp. 1141-1150.
 L. Zhao, K. Jia. “Multiscale CNNs for Brain Tumor Segmentation and Diagnosis.” Comput Math Methods Med, 2016, pp. 1-7.
 S. Das, et al. “Brain Tumor Classification Using Convolutional Neural Network”, 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT), 2019.
 H. Ke, D. Chen, X. Li, et al. “Towards Brain Big Data Classification: Epileptic EEG Identification with a Lightweight VGGNet on Global MIC.” IEEE Access, 2018.
 S. Qian, C. Ning, Hu. Y. “MobileNetV3 for Image Classification.” 2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE), IEEE, 2021.
 Br35h: brain tumor detection 2020, Model Whale. https://www.heywhale.com/mw/ dataset/61 d3e5682d30dc001701f728/ content.
 J. Li “A deep learning based classification study of retinal fundus diseases (in chinese).” Nanchang University, 2020
 D. Kingma, J. Ba. “Adam: A Method for Stochastic Optimization.” Computer Science, 2014.
 D. Magee. “A Sequential Scheduling Approach to Combining Multiple Object Classifiers Using Cross-Entropy.” Springer Berlin Heidelberg. Springer Berlin Heidelberg, 2003.