Teaching

ECE6809G · 3 credits · 48 contact hours · 16 sessions

Instructor: Yang Xie · yang.xie@sjtu.edu.cn

Fall 2026 · Tuesdays, Periods 2–4 (08:55–11:40), Weeks 2–17

Language: English · Grading: Letter grade

Week 4 is rescheduled to Saturday, 10 October 2026.

COURSE OVERVIEW

This graduate course introduces the scientific principles, engineering methods, and translational challenges of brain-computer interfaces (BCIs). It connects neuroscience foundations with neural-signal acquisition, decoding, system evaluation, and responsible clinical translation.

The course focuses on four areas: motor BCIs, language BCIs, visual neuroprostheses, and closed-loop BCIs.

RECOMMENDED BACKGROUND

Students will benefit from prior coursework or practical experience in machine learning, linear algebra, deep learning, neuroscience, and signals and systems. These subjects are recommended preparation rather than strict prerequisites; essential concepts will be reviewed as needed.

LEARNING OUTCOMES

By the end of the course, students will be able to:

• Explain the neural basis, signal sources, and system architecture of major BCI paradigms.

• Compare invasive and noninvasive BCIs and select appropriate performance, usability, safety, and clinical evaluation metrics.

• Interpret representative research in motor BCIs, language BCIs, visual neuroprostheses, and closed-loop BCIs.

• Build, evaluate, and communicate a reproducible neural-decoding workflow using the Neural Latents Benchmark (NLB).

• Critically assess a BCI device or application, including its evidence, limitations, risks, and future directions.

WEEKLY SCHEDULE

Week 2 — BCI overview, taxonomy, system loop, and evaluation

Week 3 — Neural signals, recording, data quality, and preprocessing

Week 4 — Neuronal basis of motor control

Week 5 — Motor BCI: invasive and noninvasive · Motor BCI literature discussion

Week 6 — Movement-decoding algorithms and the NLB benchmark · NLB project launch

Week 7 — Neuronal basis of language

Week 8 — Language BCI: handwriting and speech · Language BCI literature discussion

Week 9 — Neuronal basis of vision

Week 10 — Visual prosthesis · Visual-input literature discussion

Week 11 — NLB project progress presentations and feedback · NLB mid-term review

Week 12 — Closed-loop BCI

Week 13 — Closed-loop case studies: depression and epilepsy · Closed-loop BCI literature discussion

Week 14 — Clinical translation, safety, regulation, and neuroethics

Week 15 — NLB project presentation and peer review · NLB peer review

Week 16 — Individual BCI device or application presentations I · Final presentation

Week 17 — Individual BCI device or application presentations II and course synthesis · Final presentation

LEARNING ACTIVITIES

• Lectures and guided discussion, with neuroscience foundations preceding each application area.

• Collaborative literature presentations in motor BCIs, language BCIs, visual neuroprostheses, or closed-loop BCIs.

• A staged NLB algorithm project focused on reproducible neural decoding, evaluation, interpretation, and peer review.

• An independent final presentation introducing, critically evaluating, and projecting the future of one BCI device or application.

ASSESSMENT

• Class participation and literature discussion — 20%

  Collaborative paper presentation 12%; preparation and discussion 8%.

• Biweekly reading cards or case memos — 20%

• NLB algorithm project — 35%

  Environment and baseline 5%; plan 5%; progress 5%; final code and results 15%; interpretation and reproducibility 5%.

• Individual BCI device or application presentation — 20%

  Presentation and slides only; no technical report.

• NLB peer review — 5%

FINAL PRESENTATION AND AI USE

The final presentation is completed independently. AI tools may be used, but students must submit a brief AI-use statement in Canvas identifying the tools and versions used; the stages supported; slides containing AI-assisted content; how papers, numbers, and conclusions were verified; AI errors found and corrected; and the student’s own core analysis and judgment. AI use does not replace source verification, critical analysis, or responsibility for every claim.

REFERENCE BOOKS

1. Wolpaw, J. R., & Wolpaw, E. W. (Eds.). Brain-Computer Interfaces: Principles and Practice. Oxford University Press, 2012.

2. Chaudhary, U. Expanding Senses using Neurotechnology, Volume 1: Foundation of Brain-Computer Interface Technology. Springer, 2025.

3. Chaudhary, U. Expanding Senses using Neurotechnology, Volume 2: Brain Computer Interfaces and Their Applications. Springer, 2025.