Design Philosophy Bridges the gap between subjective and objective mental health assessments.
1. Traditional evaluations rely on subjective scales, prone to bias or low patient insight. This system uses medical-grade hardware and AI to transform EEG signals into quantifiable indicators, enabling precise early screening.
2. Technical Principles Integrates EEG processing with AI learning models: Signal Acquisition: Captures 90s resting-state, non-invasive EEG data, filtering out muscle artifacts in real time. AI Model: Trained on clinical data from over 400 cases across three major medical centers for feature cross-comparison.
3. Functional Features Efficient: The workflow (setup, collection, analysis) completes within 15 minutes with instant reports. Quantifiable: Outputs an intuitive 1–10 SEA Score; higher scores indicate stronger physiological risk features. Safe: Non-invasive testing with zero pharmaceutical or radiation risks ensures high compliance. Compliant: Cleared by regulatory authorities (TFDA and US FDA).
4. Applications Clinical Support: Provides objective data for psychiatrists/neurologists, optimizing decisions for patients with low self-awareness. Corporate EAP: Screens employee mental health to risk-stratify stress and fatigue in high-pressure groups. Early Screening: Deploys at primary care sites to identify at-risk populations early, bridging the final mile of digital healthcare.
The EEG device received US FDA 510(k) clearance in December 2020 (K number: K201747) and subsequently obtained the Taiwan Food and Drug Administration (TFDA) medical device clearance in March 2021 (Cert. No. 007081). This milestone established the company as Taiwan's first manufacturer of medical-grade EEG hardware. Furthermore, the Stress EEG Assessment (SEA) system—developed using depression-related EEG datasets in collaboration with three major domestic medical centers—secured the TFDA Software as a Medical Device (SaMD) clearance in November 2023 (Cert. No. 008104).
The product underwent a rigorous Verification and Validation (V&V) plan during both the development and mass production stages. Hardware Verification: Passed specialized medical standard testing for EEG (Electroencephalogram). 100% inspection is performed on cosmetics, packaging, and accessories. Software Validation (4 Stages): Installation Qualification (IQ): Verifies that the system installation manual and items comply with specifications. Operational Qualification (OQ): Validates the software architecture planning and operational performance. Functional Qualification (FQ): Validates file types/formats, data import/export, data storage, and access permission settings. Performance Qualification (PQ): Validates the overall stability and controllability of the software under the specified operating environment. Third-Party Inspection & Certification: The contract manufacturer is certified under ISO 13485:2016 Medical Devices – Quality Management Systems (Certificate No.: MD 770476)
The product implements a data-driven quality management system covering incoming materials, semi-finished goods, and final outgoing inspections: Incoming Inspection & Critical Components: 100% full inspection and testing are applied. The sampling plan adopts the MIL-STD-105E-AQL II standard with an Acceptance Quality Limit (AQL) of 0.4% to ensure high quality stability. Semi-finished Software QC: A Checksum validation is enforced after copying/burning software to USB drives or discs. The target metric is a 100% perfect match; any discrepancy in characters results in a "Fail." Mandatory USB media protection and antivirus/malware scanning are executed with comprehensive logs. Final Quality Control (FQC): Includes hardware power-on tests, UI login checks, real-time waveform display confirmation, execution of the full software testing workflow, and verification of report uploading and web-portal viewing.
A comprehensive risk management system and regulatory compliance framework are established across incoming materials, manufacturing, finished goods, and post-market stages: Medical Device Risk Management: Strictly complies with ISO 14971:2019, IEC/TR 80002-1 (Software Hazard Identification), and ISO/TR 24971:2020. It encompasses discussions on intended use/misuse, identification of hazardous situations, evaluation of injury severity and probability, implementation of risk control measures, residual risk evaluation, risk-benefit analysis, and completeness reviews. Cybersecurity Risk Control: Complies with the "Cybersecurity Guidance for Medical Device Manufacturers," covering risk analysis/assessment, residual risk identification, and risk validation. Software Requirement Specifications (SRS) and architecture diagrams are developed for process hazard analysis. Compliant Manufacturing & Quality Patrols: Manufacturing and contract manufacturing fully comply with Taiwan's Medical Devices Act (including factory registration exemption, medical device manufacturer license, QMS Manufacturing License No. QMS2055, and contract manufacturing approval No. CM0004441). The contract manufacturer strictly executes quality patrols and trial production reviews (DHF document audits) based on tech-transfer standard operating procedures.
A robust system for post-market surveillance, software lifecycle management, and crisis response is established: Post-Market Surveillance (PMS): Standardized workflow covering information collection, adverse event reporting, product recall/disposal, and CAPA implementation. Cybersecurity Monitoring & Response: 15-Day Notification: Users will be notified within 15 days of vulnerability detection with temporary risk controls. Regulatory Reporting: Uncontrolled risks causing serious adverse events or data breaches will be reported immediately to health authorities. Amendment Registration: Modifications involving specifications or performance will be registered per regulatory laws. Software Maintenance & Lifecycle: Annual maintenance and software anomaly change control are strictly enforced. Upon End-of-Life (EOL), patient data is completely wiped during decommissioning. Crisis Management System: Structured into three phases: Prevention/drills, Emergency Response (escalation containment and rapid resolution), and Recovery (performance evaluation). Regulatory Compliance: Zero non-compliance or violation records within the past three years.
Chun-Hung Chang, Alexander T. Sack,Che-Sheng Chu. EEG Biomarkers of Accelerated Prolonged Intermittent Theta-Burst Stimulation in Depression: Machine Learning with a Minimal-Electrode Montage. Brain Stimulation. In press. https://doi.org/10.1016/j.brs.2026.103135
