Technology

Core Technology

Brain-Computer Interface

A Brain-Computer Interface (BCI) establishes a direct communication pathway between the brain and external devices — without requiring any physical movement. SIU BehaveNeu™ uses a non-invasive EEG-based passive BCI that reads electrical potentials generated by neurons during cognitive and emotional activity.

Our system operates in the passive BCI paradigm, continuously monitoring mental states without requiring deliberate user input — enabling seamless integration into everyday activities.

Non-Invasive

No surgical implants — safe, comfortable, dry electrode design

Passive BCI

Continuous monitoring without requiring deliberate user action

Wireless

Bluetooth 5.0 LE with sub-10ms latency for real-time data

BCI Signal Pipeline

Neural Signal Source

Neurons fire → electrical potentials (μV range)

↓

EEG Electrodes (×8)

Ag/AgCl dry electrodes capture scalp potentials at 256 Hz

↓

Signal Conditioning

24-bit ADC, bandpass filter (0.5–50 Hz), notch at 50 Hz

↓

BLE 5.0 Transmission

Encrypted wireless data stream to mobile device

↓

AI Model Inference

LSTM-CNN processes 1-second epochs → cognitive state label

↓

Insight Delivery

Neurofeedback + Dashboard + Cloud Storage

Hardware

EEG Technology

Medical-grade electroencephalography in a lightweight, comfortable wearable form factor.

8 EEG Channels

International 10-20 electrode placement system: Fp1, Fp2, F3, F4, C3, C4, P3, P4 — covering frontal, central, and parietal regions.

256 Hz Sampling

24-bit ADC resolution with up to 256 samples/second per channel — sufficient for capturing all clinically relevant brain rhythms.

Dry Ag/AgCl Electrodes

Self-adhesive silver/silver-chloride dry electrodes with impedance check — no conductive gel required, ready in <60 seconds.

Signal Processing

Hardware bandpass filter (0.5–50 Hz), 50/60 Hz notch filter, Common Average Reference (CAR) montage, ICA artifact removal.

8-Hour Battery

Rechargeable 500 mAh LiPo battery with wireless charging; sleep mode extends standby to 3 days.

Comfort Design

82g lightweight headband with adjustable sizing (52–60 cm head circumference); hypoallergenic silicone contact points.

Detected Brainwave Bands

Delta

0.5–4 Hz

Deep sleep, unconscious processing

Theta

4–8 Hz

Creativity, meditation, drowsiness

Alpha

8–13 Hz

Relaxation, calm focus, closed eyes

Beta

13–30 Hz

Active thinking, focus, alertness

Gamma

30–50 Hz

High cognition, perception binding

Artificial Intelligence

AI & Machine Learning

SIU BehaveNeu™ uses a proprietary hybrid LSTM-CNN architecture — combining the temporal sequence modeling power of Long Short-Term Memory networks with the spatial feature extraction of Convolutional Neural Networks.

Models are trained on our internal dataset of 50,000+ labeled EEG epochs from 500+ subjects, augmented with transfer learning from public datasets (DEAP, DREAMER, SEED).

Attention Level Detection

94.2%

Stress Classification

91.7%

Cognitive Load Estimation

93.1%

Emotion Recognition

87.8%

Fatigue Detection

89.5%

Model Architecture — LSTM-CNN Hybrid

Input Layer 8 × 256 EEG time-series matrix (1-second epoch)

CNN Block ×3 Conv1D(64) → BatchNorm → ReLU → MaxPool — spatial feature extraction

LSTM Block ×2 Bidirectional LSTM(128) → Dropout(0.3) — temporal sequence modeling

Attention Layer Self-attention mechanism weights relevant time windows

Dense Head FC(256) → FC(128) → Softmax(n_classes) — multi-task classification

2.1M

Parameters

<500ms

Inference

TFLite

On-Device

Training

Neurofeedback Protocol

Closing the loop — teaching the brain to regulate itself through real-time neural feedback.

1

Baseline Recording

A 5-minute eyes-open/closed baseline captures your personal EEG signature for personalized calibration.

2

State Detection

Real-time AI classifies your mental state every second — attention, relaxation, stress, or distraction.

3

Feedback Signal

Visual, auditory, or haptic feedback is provided when your brain is in the target state (e.g., high alpha = calm focus).

4

Operant Conditioning

Over multiple sessions, the brain learns to self-regulate — reducing aberrant patterns associated with ADHD, anxiety, and stress.

Neurofeedback Loop

Brain Activity

EEG Capture

AI Analysis

Feedback

↻ Continuous real-time loop at 1 Hz feedback rate

Platform

Mobile App & Cloud Analytics

SIU BehaveNeu™ Mobile App

Available on iOS 14+ and Android 10+, the app is the primary interface for neurofeedback sessions, real-time EEG visualization, and daily cognitive tracking.

Live EEG waveform display with band power meters
AI-guided neurofeedback training sessions
Daily cognitive score & mental wellness index
Personalized exercise library (meditation, focus, sleep)
Progress analytics with weekly/monthly trends
Clinician portal for remote patient monitoring

Cloud Analytics Platform

HIPAA-compatible secure cloud infrastructure with enterprise-grade analytics for clinicians, researchers, and institutional administrators.

Multi-subject longitudinal EEG data warehouse
Population-level cognitive analytics & benchmarking
Customizable alert thresholds for clinical monitoring
FHIR-compliant EHR integration (HL7 support)
One-click research dataset export (EDF+, CSV, JSON)
Role-based access: Patient / Clinician / Researcher / Admin

System

System Architecture

Hardware Layer

NeuroBand Headset Dry EEG Electrodes 24-bit ADC BLE 5.0 Radio IMU Sensor

Edge Processing Layer

Real-time Noise Filter Artifact Rejection Band Power Extraction Feature Engineering TFLite Inference

Mobile Layer

iOS & Android App BLE SDK Local SQLite DB Neurofeedback Engine Push Notifications

API Layer

RESTful API Gateway WebSocket Streams OAuth 2.0 / JWT Rate Limiting SDK (Python, R)

Cloud Layer

AWS / Azure Hosted EEG Data Warehouse ML Training Pipeline FHIR EHR Bridge Analytics Dashboard

The Science & Technology