腦神經人機界面及應用

Abstract

近年來在醫學診斷和神經生物學研究中，腦電波訊號(Electroencephalogram, EEG)已成為非常有用的非侵入式生理訊號工具，主要因為它能在千分之一秒(milliseconds)的時間裡提供極高的生理訊號解析度，直接反映出細胞群體中動態的變化。在所有量測大腦造影的醫療工具中，量測腦電波訊號最不受任何限制，因為在量測過程中，受試者不需受到固定身體和保持頭不動等限制。然而，若將現今市面上的腦波監測系統應用到日常生活中卻會深深受到許多限制，例如：需在頭皮上塗抹導電膠才能量測腦電波訊號，系統缺乏高精確度的量測，即時訊號處理和有效去除雜訊等功能，皆是主要腦波監測系統的缺失。因此，為了解決這些缺失，本論文主要目的在開發，設計和測試一可直接應用於日常生活環境裡的腦神經人機界面，可讓使用者在日常生活的自然情況下方便使用生理訊號監測系統，即使是在多變的環境中做不同的工作任務，亦能直接擷取大腦活動變化。更重要的是，本論文為了探討此創新的移動式無線腦神經人機界面的應用，我們亦建構一環繞式虛擬實境動態駕駛環境，此環境可真實的模擬日常生活的駕駛情境，無論是應用於神經科學基礎研究上或是應用於日常生活中的警示提醒，皆能有效地測試此腦神經人機界面的效能。在本論文中，我們提出三個與日常生活相關的應用研究，但絕不僅侷限於這些而已，這三個研究分別為：(1) 在虛擬實境動態駕駛環境裡探討受試者駕車時的認知狀態的變化；(2) 探討利用聲音回饋來維持駕駛員的精神狀態和注意力是否集中的腦神經變化；(3) 在虛擬實境動態駕駛環境裡探討動態刺激對身體感覺和知覺的腦神經變化。對目前許多探討複雜腦功能的研究來說，本論文探討的應用能在生活環境裡受限最小並提出許多重要新穎的發現，這些研究成果亦可有效提升正常人每天在反覆工作任務環境下的工作能力表現，也可應用於腦傷、生病或身體不健全等病人更細部的動態認知狀態研究。否則這些成果至今僅能請受試者來傳統腦波實驗室參與實驗，並要求受試者固定身體，頭不能亂動，眼睛不能亂瞄等限制受試者行為。一旦有了創新的可攜式無線腦神經人機界面，這些限制將能一一破除，我們相信這能為認知神經科學和人機界面互動等研究開啟另一新的頁章。

Electroencephalogram is a powerful non-invasive tool widely used by for both medical diagnosis and neurobiological research because it provides high temporal resolution in milliseconds which directly reflects the dynamics of the generating cell assemblies, and it is the only brain imaging modality that does not require the head/body to be fixed. However, the lack of availability of EEG monitoring system capable of high-definition recording, online signal processing and artifact cancellation, without use of conductive gels applied to the scalp, has long thwarted the applications of EEG monitoring in the workplace. This dissertation describes a design, development and testing of a neural human machine interface that allows assessment of brain activities of participants actively performing ordinary tasks in natural body positions and situations within a real operational environment. More importantly, this dissertation also discuss the implications of this innovative mobile wireless brain imaging technology in neuroscience and neuro-technology, through three sample studies: (1) cognitive-state monitoring of participants performing realistic driving tasks in the virtual-reality-based dynamic driving simulator; (2) the efficacy and neural correlates of auditory feedback delivered to participants to maintain participants attention and alertness; (3) the neural correlates of kinesthetic sensation and perception in the dynamic driving simulator. Results of these studies provide many new insights into the understanding of complex brain functions of participants performing ordinary/routine tasks in a minimum constrained environment. These results also allow a better appreciation of the limitations of normal human performance in repetitive task environments, and may allow more detailed study of changes in cognitive dynamics in brain-damaged, diseased, or genetically abnormal individuals. Furthermore, these findings might be difficult, if ever possible, to obtain in a standard EEG laboratory where participants are asked to limit their eye blinks, teeth clinching, head/ body movements. We, thus, believe this work opens a new chapter in neuro-cognitive human-machine interface/interaction.