# BCI-004: BCI2000: A General-Purpose Brain-Computer Interface (BCI) System

> 本页由本地论文卡片生成。请不要直接编辑本页；修改源卡片后运行 `vp run docs:generate`。

> 中文支持说明：本页是中文站点镜像，保留论文标题、证据状态和来源字段的原始表述，避免未经核实的翻译改变证据边界。

> 内部 wiki 边界：本页提供论文卡片证据和用户提供的 PDF 下载入口，仅用于私有阅读；全文抽取文本不发布。

> 来源: `library/paper_cards/BCI-004.md`

## 论文访问

* 内部 PDF: <a href={"/papers/BCI-004.pdf"} download style={{ display: "inline-flex", alignItems: "center", justifyContent: "center", minHeight: "2.25rem", padding: "0.45rem 0.8rem", borderRadius: "6px", backgroundColor: "#047857", color: "#ffffff", fontWeight: 700, lineHeight: 1, textDecoration: "none", boxShadow: "0 1px 2px rgba(15, 23, 42, 0.22)" }}>下载论文 PDF</a>
* DOI / 官方页面: [10.1109/TBME.2004.827072](https://doi.org/10.1109/TBME.2004.827072)
* 部署边界: 这些 PDF 链接只适合私有/受保护的内部 wiki。

## BCI-004: BCI2000: A General-Purpose Brain-Computer Interface (BCI) System

## Metadata

* ID: BCI-004
* Title: BCI2000: A General-Purpose Brain-Computer Interface (BCI) System
* Year: 2004
* DOI / URL: 10.1109/TBME.2004.827072
* Local PDF: 见上方论文访问区块
* Text artifact: local-only path withheld from docs site
* Review status: `extracted`

## Study Type

* Track: BCI / EEG Foundations
* Task: system paper describing a modular BCI research and development platform
* Participants or dataset: multiple BCI implementations are demonstrated; not a single clinical trial
* Hardware: configurable BCI acquisition and application systems, including EEG and other brain-signal modalities
* Channels or sensors: examples include 16-channel and 64-channel acquisition configurations, plus high-rate acquisition examples

## Methods

* Paradigm: general-purpose BCI platform with modular source, signal-processing, application, and operator modules
* Signal processing or model: platform supports feature extraction, translation algorithms, online adaptation, event markers, and different applications
* Training/calibration: operator and parameter modules support configuration and online monitoring; platform supports multiple BCI protocols
* Online/offline: designed for online operation and standardized data storage for offline analysis

## Results

* Metrics: output latency, latency jitter, system clock jitter, processor load, online function across multiple BCI designs
* Main findings: BCI2000 fulfilled real-time requirements in representative implementations; reported average output latency and jitter were low; modular designs supported sensorimotor rhythm, ECoG, SCP, and P300 applications
* Reported limitations: paper describes platform architecture rather than scene-aware BRI, SSVEP-MI fusion, object detection, or robot grasp execution

## Relevance To This Project

* Supports: modular BCI runtime design, source/signal-processing/application/operator separation, online timing/logging discipline, and standardized data storage
* Conflicts with: does not define the SAH-BRI-Grasp runtime or robot-side interfaces
* Design implication: SAH-BRI-Grasp should preserve modular boundaries and log timing/event markers across EEG, stimulus, vision, and robot modules

## Extracted Evidence

| Claim | Status | Evidence Note | Page/Section |
| --- | --- | --- | --- |
| A BCI runtime can be organized into source, signal processing, user application, and operator modules. | verified | The BCI2000 model explicitly defines these four communicating modules. | Section II; Fig. 2 |
| BCI system progress requires systematic comparison of signals, processing methods, applications, and protocols. | verified | The introduction motivates BCI2000 by the need to compare alternative signals, feature extraction methods, translation algorithms, and output applications. | Introduction |
| Real-time timing and jitter should be measured in online BCI systems. | verified | The paper evaluates output latency, output jitter, system clock jitter, and processor load for online operation. | Section III.A |
| BCI2000 demonstrates that modular online BCI operation can satisfy real-time requirements. | verified | Representative implementations had low latency and jitter, and the authors conclude the platform fulfills real-time BCI requirements. | Sections III-IV |
| SAH-BRI-Grasp should version runtime modules and event logs across EEG, stimulus, vision, and robot components. | inferred | This is a design transfer from BCI2000's modular runtime and standardized data storage, not a result from SAH-BRI-Grasp. | Discussion |

## Open Questions

* Should the first implementation use BCI2000-compatible event/log concepts, ROS messages, or a custom local record schema?
* What clock-synchronization method will connect EEG windows, stimulus markers, camera frames, and robot events?
* Which real-time latency and jitter thresholds are acceptable for Exp1-Exp3?