| 摘要: | 大腦與心臟是有著相互連接的，在大腦中控制自主神經系統來對心臟進行調節，控制心臟跳動的速度。心臟會將心臟內部的訊息透過迷走神經或是脊髓傳遞給大腦，讓大腦可以更好的去控制心臟跳動的速度。大腦會根據不同的訊息來控制心臟跳動的速度，其中生理上的改變以及心理上的改變都會改變心律的改變，甚至於只是想像也可以讓心律產生改變。因此不同的階段下，大腦與心臟之間的關係也會產生變化，而本研究的目的在於檢測不同階段下腦電與心電的改變，以及其相關性的變化。
在本研究中召集了20個受試者，給予他們一份購物清單並記住，接著讓他們進入一個虛擬的超市中挑選清單上的物品。在這個任務階段的前後都各收5分鐘的放鬆階段的資料來做比對。總共有三個階段，分別為前測放鬆、任務以及後測放鬆階段。在所有的階段中都接收5個頻道的腦電資料、2個頻道的眼動資料以及3個頻道的心電資料。並對這些資料進行相關性、t檢定以及機器學習的方式進行分析。
在本研究中測試了不同階段中，腦電與心電特徵的變化以及兩者的相關性。腦電的特徵中則是在專注階段中的變化比放鬆階段要來的大。並且專注階段則是由交感神經佔優勢。在前測放鬆階段的放鬆程度比起後測階段要來的高。在相關性的結果中，在放鬆階段beta頻段控制心跳的特徵(nn20)，到了專注的階段全部的頻段都與該心跳有關。Theta頻段則是與心臟受到交感神經與副交感神經的調節相關(SDNN、SD2)，存在於專注階段與後測放鬆階段。Alpha頻段則是與放鬆階段的交感(LF)與副交感神經(HF)的活性有關，在前測階段中控制交感神經的活性，在後測階段中控制副交感神經的活性。機器學習中的結果則表明，在不同階段的腦電與心電特徵都是可以分辨的出來的，甚至是在不同的放鬆階段，其中MLP模型的表現最好，使用混合的模型則可以達到最高的準確度。
;The brain and heart are physiologically related, with the brain modulate the heart rate via the autonomic nervous system. The heart transfer signals through the vagus nerve or spinal cord to the brain, allowing the brain to better modulate the heart. Physiological and psychological changes can both affect heart rate, even just through imagination. Therefore, the relationship between the brain and heart can change in different state, and the aim of this study is to investigate the changes in brain and heart activity and their correlation in different state.
Twenty participants were recruited for this study and given a shopping list to remember, then asked to select the items on the list in a virtual supermarket. Beside this task, data was collected during 5-minute resting state before and after the task. Thus, there were three stages: pre-resting, attentional(task), and post-resting state. Five-channel electroencephalography(EEG), two-channel Electrooculogram(EOG) and three-channel Electrocardiography(ECG) were recorded in all state. Pearson correlation analysis was used to assess the relationship between EEG features and heart rate variability (HRV). Two-sample t-tests and machine learning methods were used to compare the different states.
This study examined the changes in EEG and HRV features and their correlation in different states. There was significant change in EEG features during the attentional compared to resting state. During the attentional state, the sympathetic nervous system dominated. In terms of correlation, beta band in the resting state was related to heart rate feature(nn20), while all bands in the attentional state were related to heart rate feature. Theta band was related to sympathetic and parasympathetic modulation of heart (SDNN, SD2), and was present in the attentional and post-resting state. Alpha band was related to sympathetic (LF) and parasympathetic (HF) activity during resting state, modulation sympathetic activity during the pre-resting state and parasympathetic activity during the post-resting state. Machine learning results showed that brain and heart activity features in different state could be distinguished, even in different resting state. Among them, the MLP model performed the best, and the use of a fusion model achieved the highest accuracy. |
