Electroarthrography (EAG)
Electroarthrography (EAG)
A novel biosignal approach to detect early cartilage pathology through load-induced electrical potentials in joints.
Overview
Articular cartilage generates measurable bioelectric potentials under mechanical loading — a phenomenon known as streaming potentials. Our lab investigates these cartilage-generated electrical signals using the OpenBCI Cyton platform, aiming to develop a non-invasive diagnostic tool for early cartilage degeneration.
The key innovation is the Electroarthrography (EAG) concept: recording and analyzing joint electrical activity during dynamic movement to characterize cartilage health, analogous to how ECG assesses cardiac function.
Core Research Project (Funded)
핵심연구 (기초연구사업)
무릎 연골의 생체전기 신호 분석을 통한 하중 감시 기술 개발 Development of Load Monitoring Technology Using Bioelectric Signal Analysis of Knee Cartilage
- Funding: 과학기술정보통신부 기초연구사업 — 핵심연구 (유형A)
- Period: 2026.03 ~ 2030.02 (4 years)
- PI: Jaehyun Lee (이재현)
- Investigator: Seungheon Han (한승헌)
- External Advisor: Year 3-4
4-Year Roadmap
| Year | Goal | Key Deliverables | KPI |
|---|---|---|---|
| Year 1 (2026-2027) | Signal reproducibility & protocol standardization | SCI paper 1, conference 2 | n=20, test-retest ICC ≥ 0.7 |
| Year 2 (2027-2028) | EAG-to-load quantitative model | SCI papers 1-2, patent 1 | Load estimation R² ≥ 0.6, n=50 |
| Year 3 (2028-2029) | Clinical application (OA patients) | SCI papers 1-2, conference 2 | OA vs. normal AUC ≥ 0.75 |
| Year 4 (2029-2030) | Integrated real-time monitoring system | SCI paper 1, patent registration | Real-time prototype completed |
Research Questions
- Does EAG produce reproducible responses to joint loading changes?
- Can we quantitatively estimate load magnitude and direction from EAG signals?
- Do cartilage lesions (e.g., osteoarthritis) alter EAG signal patterns?
- Can EAG-based load monitoring be practically applied to rehabilitation protocols?
Target KPIs (4-Year Cumulative)
| Metric | Target |
|---|---|
| SCI Publications | 5-7 papers |
| Conference Presentations | 8-10 |
| Patents | 2-3 filed, ≥1 registered |
| Subjects | 50 healthy + 20 OA patients |
| Load Estimation Accuracy | R² ≥ 0.6 |
| OA Classification | AUC ≥ 0.75 |
Current Technical Status (TRL 2)
| Component | Status |
|---|---|
| Signal Acquisition | OpenBCI 8-ch EAG + Kinvent GRF simultaneous recording |
| Signal Processing | LP filter (5Hz), drift correction, mirror padding |
| Synchronization | 2-stage sync (94% event-based + 6% xcorr fallback), 100% success rate |
| Event Detection | GRF weight-shift auto-detection, EAG inflection auto-matching |
| Data | 44 healthy adults recruited, single-leg stance task |
Projects
| Project | Stage |
|---|---|
| Cartilage-generated bioelectric potentials during dynamic joint movement | 📜 Published |
| Intra-articular pressure characteristics of knee joints | 📜 Published, 👑 Award |
| EAG-GRF synchronization (event-based + cross-correlation) | 📊 Active |
| EAG test-retest reliability study | 💡 Year 1 |
| EAG-based load estimation model (ML) | 💡 Year 2 |
| OA vs. normal cartilage EAG comparison | 💡 Year 3 |
| Real-time EAG monitoring prototype | 💡 Year 4 |
Methods & Tools
- Hardware: OpenBCI Cyton board (8-channel, 250Hz), Kinvent K-Plate (GRF)
- Signal Processing: Lowpass filtering (5Hz), drift correction (detrend/moving average), baseline stabilization
- Synchronization: 2-stage — event-based trigger alignment + cross-correlation fallback
- ML Pipeline (Year 2+): Feature extraction (time/frequency domain) → Random Forest, Gradient Boosting, LSTM
- Software: Python (BrainFlow SDK, NumPy, SciPy, Matplotlib)
Key Publications
| Year | Title | Journal |
|---|---|---|
| 2025 | Cartilage-generated bioelectric potentials induced by dynamic joint movement | BMC Musculoskelet Disord |
| 2022 | Intra-articular pressure characteristics of the knee joint | J Orthop Res |
Awards
- 👑 Excellent Poster Award, "Cartilage-Generated Electric Potentials Induced by Dynamic Joint Movement," 52nd KARM Fall International Conference, 2024
- 👑 Best Oral Presentation, "Pressure Characteristics of The Knee Joints," KANMS Spring Conference, 2019
Open Source
- EAGstatic GitHub Repository — OpenBCI BrainFlow EAG data analyzer with EMG noise removal and signal visualization