Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.nimbusbci.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Industry Use Cases

This page maps common BCI product scenarios to Nimbus model and workflow choices. It is intentionally pattern-focused; use Basic Examples and Advanced Applications for implementation recipes.

​
Healthcare And Rehabilitation

Scenario: motor imagery rehabilitation after stroke or injury. Recommended pattern:
  • Use CSP features from motor imagery sessions.
  • Start with NimbusLDA for fast baseline feedback.
  • Use confidence thresholds to decide whether feedback should be shown.
  • Track session-level accuracy, confidence, fatigue indicators, and rejection rate.
  • Keep a clinician-visible audit trail for model version, calibration data, and session metrics.
Key docs:

​
Assistive Communication

Scenario: P300 or event-related communication interfaces. Recommended pattern:
  • Use ERP amplitude features in a consistent post-stimulus window.
  • Prefer NimbusQDA when target and non-target distributions overlap.
  • Use posterior confidence to decide between accept, confirm, or repeat.
  • Log rejected selections separately from incorrect selections.
Key docs:

​
Wheelchair Or Device Control

Scenario: safety-critical control from motor imagery or hybrid BCI commands. Recommended pattern:
  • Separate classifier output from command execution.
  • Require higher confidence for movement than for UI navigation.
  • Add a confirmation state for ambiguous commands.
  • Stop the control loop after repeated low-quality chunks.
  • Monitor latency budget end to end: acquisition, preprocessing, inference, and action.
Key docs:

​
Neurofeedback And Training

Scenario: live feedback for attention, motor imagery, or cognitive training. Recommended pattern:
  • Use streaming inference with smoothing in the application layer.
  • Keep the classifier probabilistic and expose confidence.
  • Adapt difficulty based on recent performance, not a single prediction.
  • Use NimbusSTS for Python workflows with long-session drift.
Key docs:

​
Research Studies

Scenario: offline experiments, multi-subject evaluation, or model comparison. Recommended pattern:
  • Use batch processing for reproducible offline analysis.
  • Evaluate with subject-wise or session-wise splits when studying generalization.
  • Save preprocessing and normalization parameters with every experiment.
  • Compare NimbusLDA against NimbusQDA before moving to more specialized models.
Key docs:

​
Product Analytics

For deployed BCI systems, monitor:
MetricWhy It Matters
AccuracyMeasures task success when labels are available.
Mean confidenceTracks model certainty and signal quality.
Rejection rateShows whether thresholds or preprocessing are too strict.
LatencyProtects real-time user experience.
Calibration ageIndicates when retraining may be needed.
Drift indicatorsFlags electrode, fatigue, or session changes.

​
Choosing A Starting Point

Use CaseFirst ModelWorkflow
Motor imagery rehabNimbusLDAStreaming with confidence gates.
P300 communicationNimbusQDABatch calibration, then confidence-gated selection.
Long neurofeedbackNimbusSTSStateful Python streaming.
Julia multinomial tasksNimbusProbitJulia training and streaming workflows.
Cross-subject researchNimbusLDA baselineBatch evaluation with strict normalization.

​
Next Read

Basic Examples

Compact recipes for common BCI tasks.

Advanced Applications

Cross-subject, hybrid, adaptive, and robust deployment patterns.

Model Specification

Choose a model family for your use case.

Development Workflow

Production project structure and guardrails.