Pallathadka Harikumar, Gardanova Zhanna R, Al-Tameemi Ahmed Read, Al-Dhalimy Aiman Mohammed Baqir, Kadhum Eftikhaar Hasan, Redhee Ahmed Huseen
Manipur International University, Imphal, Manipur, India.
Pirogov Russian National Research Medical University, Moscow, Russia.
Iran J Psychiatry. 2024 Jul;19(3):327-336. doi: 10.18502/ijps.v19i3.15808.
Dementia is a broad term referring to a decline in problem-solving abilities, language skills, memory, and other cognitive functions to a degree that it significantly disrupts everyday activities. The underlying cause of dementia is the impairment or loss of nerve cells and their connections within the brain. The particular symptoms experienced are contingent upon specific regions of the brain affected by this damage. In this research, we aimed to investigate the nonlinear dynamics of the mixed demented brain compared to healthy subjects using electroencephalogram (EEG) analysis. For this purpose, EEG was recorded from 66 patients with mixed dementia and 65 healthy subjects during rest. After signal preprocessing, sample entropy and Katz fractal dimension analyses were applied to the preprocessed EEG data. Analysis of variance with repeated measures was utilized to compare the nonlinear dynamics of brain activity between dementia and healthy states and partial correlation analysis was employed to explore the relationship between EEG complexity measures and cognitive and neuropsychiatric symptoms of patients. Based on repeated measures ANOVA, there was a significant main effect between groups for both Katz fractal dimension (F = 4.10, P = 0.01) and sample entropy (F = 4.81, P = 0.009) measures. Post hoc comparisons revealed that EEG complexity was significantly reduced in dementia mainly in the occipitoparietal and temporal areas (P < 0.05). MMSE scores were positively correlated with EEG complexity measures, while NPI scores were negatively correlated with EEG complexity measures, mainly in the occipitoparietal and temporal areas (P < 0.05). Moreover, using a KNN classifier, all significant complexity measures yielded the best classification performance with an accuracy of 98.05%, sensitivity of 97.03% and specificity of 99.16% in detecting dementia. This study demonstrated a unique dynamic system within the brain impacted by dementia that results in more predictable patterns of cortical activity mainly in the occipitoparietal and temporal areas. These abnormal patterns were associated with patients' cognitive capacity and neuropsychiatric symptoms.
痴呆是一个广义术语，指解决问题能力、语言技能、记忆力及其他认知功能下降到显著干扰日常活动的程度。痴呆的根本原因是大脑内神经细胞及其连接受损或丧失。具体出现的症状取决于受此损伤影响的大脑特定区域。在本研究中，我们旨在通过脑电图（EEG）分析，调查与健康受试者相比，混合性痴呆大脑的非线性动力学。为此，在静息状态下记录了66例混合性痴呆患者和65名健康受试者的脑电图。信号预处理后，将样本熵和卡茨分形维数分析应用于预处理后的脑电图数据。采用重复测量方差分析比较痴呆状态和健康状态下大脑活动的非线性动力学，采用偏相关分析探讨脑电图复杂性测量与患者认知及神经精神症状之间的关系。基于重复测量方差分析，卡茨分形维数（F = 4.10，P = 0.01）和样本熵（F = 4.81，P = 0.009）测量在组间均有显著的主效应。事后比较显示，痴呆患者脑电图复杂性主要在枕顶叶和颞叶区域显著降低（P < 0.05）。简易精神状态检查表（MMSE）评分与脑电图复杂性测量呈正相关，而神经精神症状问卷（NPI）评分与脑电图复杂性测量呈负相关，主要在枕顶叶和颞叶区域（P < 0.05）。此外，使用K近邻分类器，所有显著的复杂性测量在检测痴呆时均产生了最佳分类性能，准确率为98.05%，灵敏度为97.03%，特异性为99.16%。本研究证明了大脑内受痴呆影响的独特动态系统，该系统导致皮质活动出现更可预测的模式，主要在枕顶叶和颞叶区域。这些异常模式与患者的认知能力和神经精神症状相关。
