標題: 可攜式EEG/EKG/FNIRS腦神經影像系統研發暨其整合型生醫感測處理晶片系統設計---總計畫(I)
System Development and SoC Design of a Truely Portable Neuroimaging System Based on EEG/EKG/FNIRS Multisensors(I)
作者: 方偉騏
Fang Wai-Chi
國立交通大學電子工程學系及電子研究所
關鍵字: EEG;EKG;fNIRS;單晶片系統設計;低複雜度生醫運算引擎;可攜式3D 腦神經影像系統;EEG;EKG;fNIRS;SoC design;low complexity biomedical process engine;Portable 3D NeuronImage System
公開日期: 2008
摘要: 近幾年來,由於腦科學研究的蓬勃發展,開拓了人類在科學研究上新的一頁。而除了原本的EEG 信號分析以外,腦神經研究的專家們已經把研究觸角延伸至研究腦細胞活動對於EEG 信號的相關性 研究。為了達到此研究目的,EEG/fMRI 技術被研發出來以同時偵測腦電波信號以及血氧濃度變化特 徵。然而,由於fMRI 的先天設計上,不論是從成像原理面向、設備複雜度面向、信號雜訊面向以及 設備成本面向上,其成效及結果都非常的差,導致EEG/fMRI 在研究上非常不容易普及。為了克服 EEG/fMRI 的發展困境,本整合型計劃引進目前應用於腦神經影像的最新技術-fNIRS,進一步的創造 新型腦科學研究的技術。而更進一步的考量血液動力學上對於血氧濃度變化的影響,本計畫亦整合 EKG 心電圖信號進入本系統,期望能達到同時以神經內科以及循環內科的角度來分析人類在不同行為 之下的腦波狀態。此三年期整合計劃將藉由微電子、電機資訊,醫學資訊與工程等跨領域專家的合作, 開發可攜式EEG/EKG/fNIRS 腦神經影像系統,以及對應該系統的生醫感測晶片系統設計。本計畫分 為四個子計畫來執行:子計畫一其目的在於協助總計劃進行生物醫學方面的相關性研究,並協助建立 其餘子計畫系統之生物規格及驗證方法。預計將透過雛型元件以及初步實驗,進而訂定其生物安全標 準及規格,並將其結果統整建構專屬於EEG/EKG/fNIRS 的生物暨電子系統認證資料庫。在此同時, 將原有適用於個人電腦的軟體工具改寫成為適用於嵌入式平台之3D 腦神經影像工具,並引領本計畫 開發之元件進行動物實驗、人體實驗以及最後的科學驗證。子計畫二負責進行其複合生理訊號感測系 統單晶片搭配乾式電極的設計與製作來偵測腦波圖、心電圖與近紅外線之生理訊號。主要使用微機電 製程及前端類比電路設計技術,搭配SIP 整合乾式電極技術,進而完成偵測近紅外線(fNIRS)之互補式 金屬氧化電晶體影像感測器、可程式化增益放大器、整合EEG/EKG 之前級儀表放大濾波器、生醫應 用之整合型類比數位轉換器與等主要元件。子計畫三在設計一顆低功率生醫訊號處理及影像重建系統 單晶片(SoC),應用在可攜式的腦神經影像儀之核心硬體。此晶片系統架構主體為一個新穎的二維獨立 成份分析(2D-ICA)影像處理器並結合資訊壓縮模組以及短距通訊模組,通過收集來自前端類比電路所 獲得的fNIRS、EEG 及EKG 的訊號,於本晶片系統進行即時的生醫訊號處理及影像重建運算,將處 理完的結果送至後端的science station。子計畫四負責進行整合生醫訊號處理晶片的設計與製作和後端 3D 運算顯示平台的開發。主要基於可程式邏輯閘陣列 (FPGA)和後端3D 演算法技術,完成EEG 信號 處理晶片架構設計、HRV 晶片架構設計、可適性影像補插晶片等主要晶片元件,並且結合所開發出來 的3D 顯示演算法來完成生醫訊號處理及嵌入式平台開發。總計劃將有效的整合四個子計畫個別所發 展的成果,透過WLAN 無線通訊技術以及SIP 技術,達到最小系統體積及陣列感測的最終目標。此 研究預期將是微電子、電機資訊,醫學資訊與工程之跨領域的成功整合。對晶片系統設計領域以及生 物醫學領域具有觀念與應用突破的貢獻,其地位足夠提升我國生醫儀器產業水準,進而成為引領國際 的主流之一。未來將可廣泛應用於腦功能病變治療以及相關認知神經科學的相關應用,讓科技更貼近 人類的生活必須,增進生活之幸福與安全感。
Since the brain research has been flourishing development in recent years, Scientists begin to study the activiation of brain cell betewwn EEG signal. To attach this purpose, EEG/fMRI technology has been developed to detect the EEG signal and blood oxygenation-level dependent at the same time. However, fMRI has lots of weak spot on image formation of image, equipment complexity, noise issue and equipment cost. To overcome above problems, this project integrates fNIRS and EEG measurement technology for novel creation on brain research. Further more, considered the the influence of hemodynamic factor, the project also combines EKG signal measurement into system. This milestone would drive EEG research on relationship between different biomedical signals into more phases with the circulatory system. This is a three years project with cowork for different field expert including biomedic, IC design, MEMS, SoC design, computer science, and system integration, etc. It would be divided into four sub-projects to carry out. First sub-project carries on the dependence result between equipment and biomedicine signal. It helps other sub-project to set up the biological specifications through prototype simulation and first phase experimentation. The result can help to setup the biological safety standard and specification, and generalize the qualification database of EEG/EKG/fNIRS. At the mean while, the original tool which base on PC will rewrite to a 3D neuron-image tool which is suitable for embedded platform. Further more; it will help the project to develop carries on animal's experiment, human experiment and the last science to verify. Second sub-project will invent a novel circuit with MEMS electrode which includes EEG, EKG, and fNIRS sensing circuits. With the process of MEMS,mixmode circuit design, and System In Package(SIP)integrated with dry electrode, the sub-project will implement image sensor for infrared ray (fNIRS) detect, programmable gain amplifier, and instrument ampliger with programmable bandpass filter integrated for EEG/EKG signal. Sub-project three proposes a biomedical signal processing and image reconstruction SoC (System on a Chip) design with low power consumption architecture. The novel 2D Independent Component Analysis (ICA) image processor can merge raw image data acquired from neighbor image sensors as well as its own image and string out clear fNIR node image that has been removed biological noise. We introduce a super node concept in our design, with which all nodes clear image that have been processed by the 2D-ICA image processor will pipeline sending to the science station through a specific super node under 802.11g protocol. A low power consumption with high bandwidth utility goal will be reached under the architecture. Sub project four designs a platform to catch on the response for front-end node. It may base on 3D visualized algrothm of neronimage, and implement several ips include EEG processing, HRV processing, and resizing circuits with FPGA. Finally, the main project integrated the sub-projects above through WLAN wireless communication technology and SIP technology. This research will be the successful integration on microelectronics, computer science, and biomedical industry, and deeply improve relative industry in our country.
官方說明文件#: NSC97-2220-E009-051
URI: http://hdl.handle.net/11536/102544
https://www.grb.gov.tw/search/planDetail?id=1685164&docId=290420
Appears in Collections:研究計畫