| 摘要: | 本計畫的核心技術在於使用人工智慧進行腦電波辨識，並提供反饋(feedback)於使用者，以達成特殊功能與目的。我們基於中央大學電機系過去在腦電波與生醫電子裝置的硬體與生醫指標研究成果，運用人工智慧技術於社會群眾最為相關的臨床醫療、居家看護、以及科學教育三個應用層面。在臨床醫療部分選擇台灣十大死亡原因第三位的腦中風、以及人口最多的失智症作為研究題目，建立腦中風與失智症臨床資料庫，應用人工智慧評估病人的大腦狀態，結合穿顱直流電刺激作為回饋，建立有效的治療策略；在居家照護方面，發展智慧健康管理策略(包含生命特徵偵測、與腦波生物鐘)、腦波人機介面輔具，在智慧健康管理策略部分，將利用受試者的高血壓、血糖、睡眠資料，開發智慧型診斷醫師系統作為給予使用者建議回饋，而腦波人機介面將會針對重度癱瘓病人，給予穿顱直流電刺激作為反饋，提升神經重塑的能力；在教育教學環境部分，開發認知負荷腦波即時偵測系統、與課堂情緒評量偵測系統，以學員狀態提醒與教師反饋作為迴授的途徑。研究的成果，將建立中風、失智症、想像運動腦波資料庫、以及課堂學習腦波資料庫。本計畫擴展腦波人機介面的架構，將人工智慧的腦波迴授架構應用於中風、失智症、健康管理、教育產業、以及腦波控制輔具裝置等領域，提供高準確率的腦波偵測與應用，建立人工智慧腦波產業的應用價值。 ;This project aims to develop artificial intelligence (AI) – based Electroencephalography (EEG) recognition technology. The AI-based brain computer interface (BCI) developments are constructed based on the substantial research capabilities of Department of Electrical Engineering, National Central University, from EEG, medical electronics, to biomarkers. With the help of AI, the proposed BCI system will accurately detect user’s intention or mental status and then apply biofeedback to enhance user’s specific brain function. In this project, three application fields are proposed, including medical care, homecare and scientific education. For medical care aspect, we will design closed-loop BCI for rehabilitation treatments in stroke and cognitive impairment patients. By detecting patients’ brain statuses using AI, transcranial electric stimulation (tES) will be utilized as biofeedback to restore patient’s brain functions. In homecare aspect, we will design AI-based BCI for sever paralyzed patients to communicate with external environments using EEG signals recorded from motor cortex. In addition, vital signs will be also detected to manage patient’s physiological status. In the third part, a real-time BCI with biofeedback system in education environments will be developed. Students’ cognitive loads and emotional statues will be detected using deep learning networks. The mental statues will be feedback to teacher as reference information so that teachers are able to handle students’ classroom performance. With the provision of student’s mental information, teachers can modify their teaching strategy in classroom environment to achieve better teaching performances. The research outcomes of this project will also establish database for stroke patients, cognitive impairment patients, imagery motor EEG databases, and database for student’s cognitive and emotional statuses in classroom environment. For industrial benefits, the established technologies and databases of our AI-based BCIs can be applied to industry of medical care for stroke and cognitive impairment patients, industry of homecare for BCI rehabilitation system and healthcare, and industry of education for achieving better learning performance in students. With accurate detection of user’s intention mental statuses, novel applications using AI-based BCI will be established. |
