Nimbus Blog & Research Updates
Insights on BCI, neurotechnology, and probabilistic AI from our research team.
Semi-Supervised Calibration for EEG BCI: Extending Your Decoder with Unlabeled Streaming Data
Most BCI calibration protocols stop the moment you collect enough labeled trials. This post explains how to keep learning after calibration ends — using high-confidence streaming epochs to extend a Bayesian decoder with no extra labeled data.
We Validated Studio Against Hand-Written Reference Pipelines — Exact Parity on Core Deterministic Paths
Core deterministic pipelines matched exactly; the tangent-space LOSO and xDAWN comparisons differed by 0.06 and 0.01 percentage points, while deep models remained within seed-to-seed training noise.
Zero-Calibration BCI with Nimbus Studio's Indicator Nodes: When You Don't Need a Trained Decoder
Most BCI pipelines assume a trained decoder — but Nimbus Studio's indicator nodes let you stream brain-state signals into live apps with zero per-user calibration. This post explains when to reach for bandpower_indicator and blink_detector instead of a classifier, and how INDICATOR and EVENT streams wire into interactive applications.
Weekly Neurotech & BCI Digest — September 28, 2026
Cross-subject generalization, invasive BCI foundation models, neuromorphic edge inference, a new inner-speech dataset, and major ecosystem funding — the week's most technically significant neurotech developments.
Minimal Montage, Maximum Signal: EEG Channel Selection for Deployable BCI Pipelines
Most BCI pipelines are designed with lab-grade EEG headsets in mind — 64 or 128 channels, clean recordings, patient subjects. But the moment you move toward a wearable device or a clinical product, channel count becomes a hard constraint. This post explains how to select a minimal, informative electrode subset without sacrificing decoder performance.
Weekly Neurotech & BCI Digest — September 21, 2026
This week: China releases the world's first AI-enabled BCI medical device standard; in-ear EEG hits a platform inflection with IDUN's Guardian 4; MOABB 1.7.2 lands as the benchmark suite for reproducible BCI research; and the Graz BCI 2026 conference wraps with hardware and closed-loop highlights.
The Personalizer Node: Bayesian Personalization for Frozen EEG Trunks in Nimbus Studio
A step-by-step guide to wiring Nimbus Studio's Personalizer node — connecting frozen encoders like REVE or EEGNet to a Bayesian head, reading BrainState outputs, and enabling online partial_fit adaptation during live sessions.
What's New in Nimbus Studio: September 2026
Apps and Create mode headline Nimbus Studio's August–September: a one-click Apps Hub with per-user profiles and a P300 Mind Reader journey, AI Create with an App Node sandbox, the Unified Project Canvas, plus a full science audit and the Nimbus Personalizer node on nimbus-bci 0.6.0.
Weekly Neurotech & BCI Digest — September 14, 2026
This week: ABILITY Neurotech's infrared optical BCI records its first human neural data; Tether Evo publishes a cross-subject speech-decoding model; China's NEO device becomes the world's first invasive BCI cleared for routine clinical sale; and geopolitical competition around BCI heats up.
Nimbus SDK 0.6.0: Personalization that keeps up with the session
nimbus-bci 0.6.0 brings session-shaped personalization: a Bayesian head over a frozen trunk, cheap online adaptation, portable profiles, and paradigm recipes for motor imagery, P300, and Riemannian geometry.
ZUNA in the Pipeline: Foundation Model Denoising and Channel Reconstruction for EEG Preprocessing
Most EEG preprocessing chains rely on bandpass filters and ICA — but both assume the signal structure they are trying to recover. ZUNA takes a different approach: a generative foundation model that denoises raw EEG and imputes missing channels from learned population-scale representations, before a single spatial filter is applied.
Weekly Neurotech & BCI Digest — August 31, 2026
This week: Precision Neuroscience's Layer 7 enables real-time thought-based computer control without brain penetration, ABILITY Neurotech clears Dutch ethics approval for chronic ECoG trial, UC Davis reports 3,800 h of independent BCI home use in Nature Medicine, a wave of EEG foundation-model benchmarks converges, and neurotech H1 2026 funding hits $2.64B.
Passive BCIs: Decoding Mental Workload from EEG with Spectral Features and Bayesian Classifiers
Most BCI paradigms require deliberate effort — imagining movement, counting targets, or fixating on a flickering stimulus. Passive BCIs are different: they decode cognitive states continuously, without any voluntary action from the user.
Information Transfer Rate: The Engineering Metric That Should Drive Your BCI Design Decisions
Most BCI pipelines are benchmarked on accuracy alone — but accuracy without speed is an incomplete picture. This post unpacks ITR, explains the accuracy-speed-class-count tradeoff, and shows how to compute and optimize it with the Nimbus Python SDK and Nimbus Studio.
Conductor: Adaptive BCI Without Constant Recalibration
A clearer guide to how Conductor diagnoses EEG drift, separates useful uncertainty from signal noise, reduces calibration fatigue, and keeps online BCI adaptation safe with ShadowGuard.
Weekly Neurotech & BCI Digest — August 17, 2026
This week: Tether Evo's durability breakthrough for implantable BCIs, a touch-sensation milestone in quadriplegia, a fully endovascular neural interface preprint, the UET175 stroke EEG dataset, and the global race to register commercial BCI devices.
Weekly Neurotech & BCI Digest — August 10, 2026
This week: ABILITY Neurotech's first-in-human optical BCI at TU Munich, Fluent's sub-scalp speech-decoding implant raise, a million-trial EEG-image dataset, China's BCI Five-Year Plan, and Motif's FDA-approved depression BCI trial.
What's New in Nimbus: July 2026 SDK Update
The Personalizer paper ships, v0.5.0 lands a trunk-agnostic Bayesian head, and the road to 0.6 adds BaLoRA, composed uncertainty, and a real CORAL transform.
Beyond Neural Decoding: Fusing EEG with Behavioral Signals in Nimbus Studio
Most BCI pipelines use EEG alone — but keypresses, mouse events, and reaction times are rich, millisecond-precise signals. This post explains how to wire Nimbus Studio's Behavioral Recording node into a hybrid pipeline and why Bayesian decoders benefit from the fusion.
Weekly Neurotech & BCI Digest — July 27, 2026
This week: a landmark Nature Medicine study shows an intracortical BCI enabling full-time employment for a paralyzed user, Universal Brain's FDA-cleared EEG cap targets psychiatry, Gestala and Hemispheric raise $114M combined, and BCI funding hits $939M YTD.
Active Learning for BCI Calibration: How Bayesian Uncertainty Drives Smarter Trial Selection
Most BCI calibration protocols present trials in a fixed, predetermined order — but Bayesian classifiers already know which trials would teach them the most. This post explains how to use NimbusSDK's posterior uncertainty to adaptively select calibration trials, cutting the time to a deployment-ready decoder.
SSVEP Decoding with CCA: Building an SSVEP Classifier Pipeline in Nimbus Studio
Motor imagery needs subject-specific spatial filters and labelled calibration. SSVEP is different: CCA matches EEG to known flicker sinusoids without learning those filters — then Nimbus Studio classifies the correlation features with LDA (or Classic ML) for a complete pipeline.
Weekly Neurotech & BCI Digest — July 13, 2026
This week: Nature Medicine documents the first truly independent long-term BCI user, BrainCo's $2.8B wearable bet reshapes the China BCI landscape, Neuropype 2026 lands deep-net pipeline support, and generative AI decoding gets a landmark review.
Benchmarking Your BCI Pipeline on Public Datasets: A MOABB-Style Workflow in Nimbus Studio
Most BCI pipelines are validated on private recordings — but honest generalization requires public benchmarks. This guide walks through MOABB-style multi-subject evaluation in Nimbus Studio using the Public Data node, cross-validation, and Bayesian classifiers from the Nimbus Python SDK.
Nimbus Studio & Nimbus SDK: June 2026 Product Updates
June updates for Nimbus Studio and Nimbus SDK: adaptive deploy, zero-calibration apps, Hub AI Create, researcher calibration tools, and Windows desktop support.
Weekly Neurotech & BCI Digest — July 6, 2026
This week: Paradromics' first-in-human wireless BCI at U of Michigan, BrainGate hits 22 WPM, Motif Neurotech enters its first depression trial, low-density EEG decoding breakthroughs, and the clinical trial surge continues.
From Training to Streaming: Deploying a Live EEG Decoder with the Nimbus Python SDK
Most BCI pipelines train a model offline — then hit a wall when it's time to deploy. This post walks through the full journey from batch training to live EEG streaming using the Nimbus Python SDK, covering the chunk loop, confidence gating, and online adaptation.
Weekly Neurotech & BCI Digest — June 29, 2026
This week: long-term independent speech BCI in Nature Medicine, China's NEO reimbursement milestone, end-to-end BIT decoder hits new SOTA, and ARIA's scalable neurotechnologies programme.
The EU AI Act Is Now in Force: What High-Risk BCI Classification Means for Your Decoder
With the EU AI Act now live, AI-powered BCIs are formally classified as high-risk systems — triggering transparency, explainability, and human oversight requirements. This post explains what that means in practice and how Nimbus SDK's probabilistic architecture maps to the regulation's core demands.
Weekly Neurotech & BCI Digest — June 22, 2026
This week: BrainGate's 22 WPM typing neuroprosthesis in Nature Neuroscience, Paradromics' full Connect-One clinical implant, Motif Neurotech's FDA IDE for depression, and the EU AI Act's high-risk classification of BCIs.
EEG Functional Connectivity as a BCI Feature: Phase Coupling, Coherence, and Bayesian Decoding
Functional connectivity metrics like PLV and coherence capture synchronized EEG network dynamics that per-channel spatial filters miss. This post explains when connectivity features help, how to extract them with Nimbus Studio's Functional Connectivity node, and how to pair them with Bayesian classifiers.
Weekly Neurotech & BCI Digest — June 15, 2026
This week: Columbia's 65,536-electrode single-chip BISC implant, Paradromics' first-in-human Connexus recording, $335M in BCI funding, and China's historic first commercial invasive BCI approval.
Epoch Design and Trial Structure in BCI: How Your Protocol Shapes Your Decoder
This post explains how trial length, baseline windows, overlap stride, and event alignment interact — and how Nimbus Studio's Trial Protocol, Epoching, Baseline Correction, Data Augmentation, and Evaluation Plan nodes let you iterate without rewriting your pipeline.
Weekly Neurotech & BCI Digest — June 8, 2026
This week: generative AI breaks the BCI calibration ceiling, in-ear EEG hits the consumer market, and $1.75B concentrates around category leaders.
Confidence-Gated BCI: How Entropy and Rejection Policies Keep Decoders Reliable in the Real World
A decoder that outputs a label on every trial sounds robust — until a noisy session proves otherwise. This post shows how to use entropy, calibration, and rejection policies from the Nimbus Python SDK to build BCI systems that know when to act and when to abstain.
Weekly Neurotech & BCI Digest — June 2, 2026
A curated weekly roundup of the most important developments in neurotechnology and brain-computer interfaces for ML/BCI engineers and applied researchers.
May at Nimbus Studio: EEG Devices and Interactive Games
A redesigned Home screen with Projects & Lab sharing, desktop app updates, live BrainBit/PiEEG streaming, Interactive Games, new deep learning models and evaluation improvements.
Meet Your BCI Co-Pilot: The AI Assistant in Nimbus Studio
The Nimbus Studio AI assistant explains your pipeline, reviews runs against benchmarks, and applies fixes from the execution terminal.
BCI Calibration with Nimbus Studio: From Hardware to Trained Decoder
Most BCI calibration workflows are ad hoc — raw recordings, manual labeling, hand-coded loaders. This post walks through Nimbus Studio's end-to-end calibration pipeline and shows how Bayesian priors in the Nimbus Python SDK cut the labeled-data requirement significantly.
Weekly Neurotech & BCI Digest — May 26, 2026
Your weekly roundup of the most important developments in brain-computer interfaces and neurotechnology — from China's AI implant push to record-breaking funding in Q1 2026.
Decoding Under Drift: How NimbusSTS Tracks Brain State Across Sessions
A practical look at NimbusSTS — Nimbus SDK's experimental latent-state classifier — and how Extended Kalman Filter-style inference handles EEG non-stationarity that trips up static decoders.
Weekly Neurotech & BCI Digest — May 11, 2026
This week: a low-power differencing feature beats SOTA intracortical decoding at the silicon level, a wireless cortical visual prosthesis reaches its third human implant, China's BCI sector raised ¥3.8B in Q1, and a new real-world wearable EEG benchmark dataset drops at SenSys 2026.
April in Nimbus Studio: Introducing Nimbus Academy, Better Visualization Modals
Monthly product update for Nimbus Studio: Nimbus Academy learning flow, scikit-learn model node, and improved data views + smoother imports.
Expected Free Energy: How Active Inference BCI Systems Balance Exploration and Exploitation
Expected Free Energy (EFE) is the core quantity Active Inference agents minimize when selecting actions. This post unpacks EFE into its epistemic and pragmatic components, shows why it naturally produces exploration-exploitation balance, and connects the math to real-time BCI control loops.
The Free Energy Principle, Ground Up: An Intuition for BCI Engineers
The Free Energy Principle is the theoretical bedrock of Active Inference, but most introductions bury the intuition in notation. This post builds FEP from first principles — surprise, Markov blankets, and the perception-action loop — so BCI engineers have a solid mental model before reaching for the math.
Closing the Loop on the Brain: Designing Neurofeedback Systems with Active Inference
Neurofeedback BCIs don't decode intent — they feed brain state back to the brain itself. This post explains why probabilistic AI and Active Inference are uniquely suited to neurofeedback: calibrated feedback signals, online adaptation to the user's learning curve, and a generative model that spans signal quality, brain state, and neural dynamics.
Weekly Neurotech & BCI Digest — April 27, 2026
This week: Cambridge's neuromorphic memristor chip, Cornell's grain-of-salt implant, Science Corp's biohybrid first-in-human push, Flow Neuroscience's at-home depression device hitting shelves, a new ADHD multimodal dataset, and the shape of neurotech's $2B funding year.
Why Nimbus SDK? Beyond scikit-learn and pyRiemann for BCI
Nimbus SDK is built for uncertainty-aware, streaming BCI workflows rather than generic offline classification. This post explains where Nimbus fits next to scikit-learn and pyRiemann, when to use each stack, and why diagnostics like entropy, calibration, and ITR matter in real BCI systems.
From ELBO to EEG: A Practical Guide to Variational Inference for BCI Engineers
A hands-on explainer for ML/BCI engineers: what variational inference is, why the ELBO is the right objective for probabilistic decoding, and how it underlies every Active Inference model you'll build.
Weekly Neurotech & BCI Digest — April 20, 2026
This week's digest covers a multi-region intracortical BCI for 3D VR navigation, a Cambridge memristor chip that slashes AI energy use, China's new national BCI standards, and a strong mid-April funding round led by Beacon Biosignals' $97M Series B extension.
Continual Learning in BCI: Handling Neural Drift with Online Bayesian Updates
EEG signals shift over time — across sessions, electrode placements, and mental states. This post explains how online Bayesian updating and Active Inference give BCI pipelines a principled way to track and adapt to neural drift without full retraining.
Active Sensing in BCI: How Active Inference Closes the Loop on Uncertainty
Active sensing is Active Inference's most underrated feature: rather than passively waiting for data, the BCI system chooses what to observe next. This post explains how expected free energy drives stimulus selection, why it produces better decoders than passive paradigms, and how to implement it in practice.
Reactive Message Passing for BCI: How RxInfer.jl Brings Active Inference to Real Time
Most Active Inference tutorials stop at the math. This post goes further — explaining how reactive message passing turns the Free Energy Principle into a runtime algorithm, and why RxInfer.jl makes it practical for real-time BCI pipelines.
Weekly Neurotech & BCI Digest — April 13, 2026
This week's curated digest for ML/BCI engineers: EEG foundation models, calibration-free decoding, Columbia's single-chip implant, g.tec Spring School, and the expanding BCI clinical ecosystem.
EEG Foundation Models in Practice: What REVE Brings to BCI Preprocessing
Classical EEG preprocessing relies on hand-tuned filters and spatial methods that break across sessions and subjects. This post explains how foundation models like REVE work, how they fit into the Nimbus Studio preprocessing stack alongside CSP and causal filtering, and why a learned preprocessing layer is a natural complement to probabilistic downstream models.
Active Inference vs. Deep Learning for BCI: Why Uncertainty Quantification Changes Everything
Deep learning dominates ML benchmarks, but BCI pipelines face a different challenge: uncertainty is not noise to suppress — it is signal to leverage. This post compares deep learning and Active Inference for neural decoding, and shows how Nimbus Studio and NimbusSDK make the probabilistic approach practical.
Choosing the Right Bayesian Classifier for Your BCI Pipeline
NimbusSDK ships four Bayesian classifiers, but which one belongs in your pipeline? This guide breaks down NimbusLDA, NimbusQDA, NimbusSoftmax, and NimbusSTS — what they assume, where they excel, and when to switch — so you can make the right call before training a single model.
Weekly Neurotech & BCI Digest — Week of April 6, 2026
This week: Neuralink's automated surgery push, Columbia's single-chip implant, China's BCI clinical trials, a new visual imagery EEG dataset, MOABB benchmarks, WEF regulation report, and a $400M+ March funding wave.
March at Nimbus Studio: EEG calibration and stronger research workflows
March updates: live EEG calibration, multi-subject benchmarking, semantic channel mapping, and reliability improvements.
NimbusSTS in Practice: Handling EEG Drift (Without Recalibration)
A practical, non-repetitive guide to NimbusSTS: when to use adaptive state-space decoding, how the predict/update loop maps to EEG drift, how to tune adaptation speed, and how to deploy in Nimbus Studio.
Weekly Neurotech & BCI Digest — March 30, 2026
This week: China lands the world's first commercial BCI approval, BrainGate hits a new typing speed record in Nature Neuroscience, foundation models for EEG decoding gain traction, and Beijing's five-year plan redraws the competitive map for neural interfaces.
Implementing Closed-Loop Active Inference BCI Control in Nimbus Studio
Hands-on guide to building a real-time closed-loop Active Inference control pipeline in Nimbus Studio: loop topology, update signals, and deployment tips.
What Is Active Inference? A Practical Primer for BCI Engineers
Active Inference is the theoretical engine behind Nimbus BCI — but for engineers coming from ML or classical neuroscience, the concepts can feel abstract. This primer breaks down generative models, free energy minimization, and real-time belief updating in practical terms, with direct connections to RxInfer, NimbusSDK, and Nimbus Studio.
Cross-Session BCI Transfer with Bayesian Priors: Reuse, Adapt, and Personalize
Focuses on cross-session transfer: how to carry priors across days/users, update with minimal calibration data, and avoid full retraining in practice.
Weekly Neurotech & BCI Digest — March 23, 2026
This week's top stories in neurotech: China's landmark first commercial BCI approval, BrainGate's high-speed iBCI typing results in Nature Neuroscience, Neuralink's Blindsight visual cortex push, a $400M+ funding surge, and a look at hardware-agnostic SDK tooling.
Beyond Binary: Multi-Class BCI Decoding with Bayesian Softmax and NimbusSoftmax
Most BCI tutorials assume two classes — left hand vs. right hand, P300 vs. non-P300. But real-world applications demand more. This post explains how Bayesian Multinomial Logistic Regression works, why it outperforms classical softmax for neural data, and how to build a multi-class decoder with NimbusSoftmax in Nimbus Studio.
Decoding P300 ERPs with Bayesian QDA: A Practical Guide for BCI Engineers
P300-based BCIs are among the most clinically validated paradigms — but classical LDA classifiers struggle when class distributions overlap and signal-to-noise ratios are low. This post explains how Bayesian Quadratic Discriminant Analysis (NimbusQDA) improves P300 detection through uncertainty quantification and flexible covariance modeling, and shows how to scaffold a complete speller pipeline in Nimbus Studio.
Bayesian CSP and Motor Imagery: Building a Robust EEG Decoder with NimbusLDA
Motor imagery BCI relies on spatial filtering and classification — but classical pipelines leave uncertainty on the table. This post walks ML and BCI engineers through combining Common Spatial Patterns (CSP) with Bayesian LDA using NimbusLDA to build decoders that are accurate, calibrated, and production-ready.
Neural Drift and Why It Breaks Your BCI Classifier (And How Adaptive Bayesian Models Fix It)
EEG signals shift within and across sessions, silently breaking classifiers trained on static data. This post explains why neural drift happens, how Bayesian adaptive models like NimbusSTS track and correct for it in real time, and what this means for building BCI systems that work outside the lab.
Weekly Neurotech & BCI Digest — March 16, 2026
This week's curated neurotech digest covers China's landmark first-ever commercial BCI approval, Neuralink's high-volume production push, FDA regulatory friction, and the state of open-source BCI tooling heading into mid-2026.
Explainable by Design: Why Probabilistic BCI Models Are Built for FDA Approval
Most BCI classifiers are accurate in the lab but opaque to regulators. This post explains how probabilistic AI and Active Inference produce transparent, auditable models that align with FDA's Bayesian guidance and accelerate the path to clinical deployment.
From Classifier to Agent: How Generative Models Are Redefining BCI Decoding
Most BCI systems treat neural decoding as a classification problem — but generative models and Active Inference reframe it as inference over a world model. This post walks ML and BCI engineers through the conceptual shift from discriminative classifiers to generative agents, and explains why it produces more robust, adaptive decoders.
Active Inference for Closed-Loop BCI: The Self-Correcting Architecture
A conceptual architecture guide to closed-loop BCI with Active Inference: the perception–action loop, expected free energy, and how self-correction emerges when actions are chosen to reduce uncertainty and prediction error.
Weekly Neurotech & BCI Digest — Mar 9, 2026
This week's digest covers China's national BCI policy push and NeuroXess's surface-mesh human trials, the emerging non-invasive ultrasound wave led by Merge Labs and Nudge's $100M raise, a systematic review on LLM-BCI integration, a new visual imagery flickering-pattern paradigm, and the FDA's draft Bayesian guidance and what it means for BCI trial design.
From EEG to Action: Building Your First Real-Time BCI Pipeline with Nimbus Studio
Building a real-time BCI pipeline from scratch takes weeks of boilerplate. This guide walks ML and BCI engineers through assembling a complete, production-ready pipeline using Nimbus Studio's visual interface and Bayesian SDK — from raw EEG to live predictions with confidence scores.
The Free Energy Principle, Demystified: A Practical Guide for BCI Engineers
The Free Energy Principle is the theoretical backbone of Active Inference — and increasingly, of state-of-the-art BCI systems. This post strips away the philosophy and explains what it actually means for engineers building real-time neural decoders.
Factor Graphs and Message Passing: The Engine Behind Real-Time Bayesian BCI
Most Bayesian inference algorithms are too slow for real-time BCI — unless you structure your model as a factor graph and let message passing do the heavy lifting. This post explains how belief propagation works, why it's ideal for neural decoding pipelines, and how RxInfer and Nimbus Studio make it practical.
Uncertainty Quantification in BCI: Why Confidence Scores Matter as Much as Accuracy
Most BCI classifiers report a single prediction — but for real-world deployment, knowing how confident that prediction is can be just as critical as getting it right. This post explains uncertainty quantification from first principles and shows how Bayesian models expose calibrated confidence scores for safer, more adaptive BCI systems.
Weekly Neurotech & BCI Digest — Mar 2, 2026
This week's digest covers MIT's transcranial focused ultrasound roadmap for deep-brain neuromodulation, Rice University's dye-free molecular atlas of an Alzheimer's brain, NeuroXess reaching 54 human implants with China's first battery-integrated BCI chip, CorTec's second human implantation, Meta's Neural Band accessibility research with University of Utah, the geopolitical acceleration of China's BCI ecosystem, and the 2026 Queen Elizabeth Prize awarded to neural interface pioneers.
Nimbus Studio: February 2026 Product Updates
February was a major release month for Nimbus Studio: the new desktop app, cloud GPU offloading, and the ZUNA foundation model preprocessing node.
Within-Session Non-Stationarity in EEG BCIs: An Adaptive Bayesian Approach (NimbusSTS)
Deep technical dive into within-session non-stationarity (drift) in EEG BCIs and how online state-space Bayesian adaptation (NimbusSTS) maintains performance.
Weekly Neurotech & BCI Digest — Mar 1, 2026
This week's digest covers inner speech decoding from motor cortex at 125k-word vocabulary, new electrode-scaling insights challenging cortical dimensionality theory, Columbia's single-chip neural implant, Neuralink's GB-PRIME trial reaching 7 participants and its push toward high-volume production, Paradromics entering clinical trials, and Morgan Stanley's $80B market outlook.
Introduction to Active Inference for BCI
A technical but accessible introduction to Active Inference for brain-computer interfaces (BCI), explaining the perception-action loop, variational free energy, and how probabilistic modeling enables robust real-time decoding and adaptive control.
Causal Filters for Streaming BCI: What Changes When You Go from Batch to Real-Time
Most BCI pipelines are built and validated offline — then silently break when deployed to streaming. The culprit is almost always a zero-phase filter. This guide explains the causal/acausal distinction, how it propagates through Nimbus Studio's preprocessing chain, and how to build a deploy-safe pipeline from the start.
Cleaning EEG Before It Reaches Your Decoder: ICA, Artifact Rejection, and EOG Removal in Nimbus Studio
A practical guide to the three core artifact-removal nodes in Nimbus Studio — ICA, EOG Removal, and Artifact Rejection — and how to sequence them into a preprocessing pipeline that delivers clean epochs to any Bayesian classifier.
Expected Free Energy as a Decision Variable: Why Your BCI Needs a World Model
Most BCI classifiers output a label. Active Inference outputs a decision under uncertainty — and the difference is expected free energy. This post explains what EFE is, why it matters for real-time BCI control, and how probabilistic world models change the architecture of a decoding pipeline.
Expected Free Energy in BCI: How Active Inference Balances Exploration and Exploitation
Most BCI decoders optimize for accuracy. Active Inference optimizes for something deeper: minimizing expected surprise over future states. This post explains Expected Free Energy, its two components — epistemic value and pragmatic value — and why the distinction matters for robust, adaptive BCI control.