The Intelligence of Healthcare.
A vertically integrated ecosystem of instruments, microfluidic discs, and AI-driven LIMS — engineered to eliminate diagnostic bottlenecks at the point of care.
Stage Locked.
Every BiQadx hardware platform is milestone-gated from architecture through controlled pilot. No clinical performance claims are made at any stage until a signed, evidence-backed design review opens the next gate.
Four non-negotiable principles that govern how BiQadx develops and communicates about its products at every stage.
Performance claims are scoped to the current stage only. R&D bench data cannot be presented as clinical evidence. Every milestone is gated — promotion requires a signed design review.
Each stage transition requires a documented evidence package: IQ/OQ records, analytical validation data, HFE study reports, and risk review sign-off before the gate is opened.
All design controls, DHF artefacts, and CAPA records are maintained within an ISO 13485-aligned QMS. Every record links directly back to the originating user need and risk item.
Regulatory pathway is planned from Concept stage. Class I and Class II IVD classification decisions are locked early, shaping the entire validation evidence strategy from Day 1.
Stage Governance.
Every BiQadx product follows a disciplined, milestone-gated stage model. Claims and capabilities are scoped to the current stage only — gates are opened only when a signed evidence package is approved.
Design inputs documented under ISO 13485. Architecture brief drafted. Risk register initiated. No hardware built — only validated need and architecture intent.
Gate criteria: Design input freeze, risk classification assigned, DHF structure approved.
- User needs elicited via clinician interviews; traceability matrix started
- Risk classification (Class I / II IVD) assigned per IVDR 2017/746
- FMEA draft initiated across hardware, chemistry, and software domains
- Technology readiness level (TRL 1–2) assessment completed
Parallel hardware, chemistry, and software development tracks active — all gated by design review. Bench data is internal and does not constitute a clinical claim.
Gate criteria: Bench prototype validated against design inputs, reagent formulation locked.
- Bench prototype validated against all design inputs in DHF
- Reagent formulation locked for lyophilisation stability trials
- Software architecture review complete; API contract frozen
- Analytical performance (LoD, precision, linearity) studies initiated — no clinical comparisons
Full-system bench integration testing underway. IQ/OQ/PQ protocols drafted. Human factors engineering studies and simulated-use evaluations initiated.
Gate criteria: IQ/OQ signed-off, HFE study complete, analytical validation package approved.
- IQ / OQ / PQ protocol documents drafted and under internal review
- Inter-lot precision & accuracy studies initiated across 3 reagent lots
- Human factors engineering study (simulated-use) conducted
- Electrical safety and EMC pre-compliance testing initiated
Controlled external site deployment at partner institutions. Operator training and certification. CAPA system fully active. Post-market surveillance initiated.
Gate criteria: Site qualification complete, PMS plan active, validation report approved.
- Site qualification & operator certification protocols completed
- Post-market surveillance plan activated; complaint handling workflow live
- CAPA workflow integrated with LIMS alert and escalation system
- Regulatory submission strategy initiated under ISO 13485 QMS
Method First.
Three interlocking disciplines form the non-negotiable backbone of every BiQadx product — architecture, evidence-gated staging, and AI-augmented intelligence. In that order. Always.
Architecture First
Every platform begins with a rigorous design input process. Requirements traceability matrices, FMEA, and IVD risk classification are completed before a single component is specified — architecture drives hardware, not the reverse.
- ISO 13485-aligned design controls enforced from Day 1 of DHF creation
- Full Design History File (DHF) maintained at every stage — no retroactive documentation
- FMEA covers hardware failure modes, process failure modes, and software fault trees
- User needs and intended use locked before design output specification begins
Methods, Then Claims
BiQadx develops in transparent, milestone-gated public stages — Concept, R&D, Prototype, Pilot. No clinical performance claims are made until validated, stage-appropriate evidence is complete and reviewed.
- Stage-specific evidence packages required and signed off before stage promotion
- Regulatory strategy aligned with IVDR 2017/746; Class I / II IVD pathway pre-decided
- Analytical validation (LoD, precision, linearity) precedes any comparative clinical study
- All bench data labelled as R&D-stage only — no surrogate clinical performance
AI Discovery Layer
The Unified LIMS & AI Stack interprets instrument telemetry and assay data in real-time, building a continuously improving diagnostic intelligence engine that spans across all platform families.
- Edge-deployed synthetic neural net — no cloud dependency for core inference loop
- HL7 FHIR R4 structured output for seamless EHR and clinical workflow integration
- Continuous retraining pipeline with locked evaluation sets — no silent model drift
- All AI outputs are decision-support only — clinician remains accountable for diagnosis
No component is specified before design inputs, FMEA, and risk classification are complete.
Every performance statement is bounded by the current stage's evidence package — no shortcuts.
All AI/ML outputs are flagged as decision-support only. Diagnostic responsibility remains with the clinician.
Diagnostic Domains.
Five clinical application areas — each with defined biomarker panels, detection technologies, and time-to-result targets. All at R&D or Concept stage. No clinical performance claims.
BiQadx evaluates new application areas as evidence matures.
Sepsis ID
Rapid, culture-independent pathogen identification and resistance gene detection directly from whole blood or bronchoalveolar lavage, enabling actionable antimicrobial selection before conventional blood culture results are available.
Multiplex real-time PCR + electrochemical lateral flow on POCT cartridge
- 1Simultaneous gram-positive, gram-negative, and fungal detection via multiplexed PCR primer pools in a single cartridge pass
- 2Integrated resistance gene panel covering mecA (MRSA), vanA/B (VRE), blaKPC, blaNDM — all detected in parallel without culture
- 3Procalcitonin electrochemical sandwich immunoassay co-interpreted with nucleic acid result to classify bacteraemic vs. non-bacteraemic sepsis (R&D concept only)
Unified Ecosystem.
Six interconnected modules spanning hardware, chemistry, consumables, and digital intelligence — designed to operate as a single cohesive platform.
Live Intelligence.
R&D publications and concept insights — rotating live from our engineering teams. Research & Concept phases only.
Overcoming Fluidic Resistance in High-Density Disc Arrays
An analysis of the fluid dynamics challenges encountered during the R&D phase of the EtherX High-Density Disc, focusing on viscosity variations in whole blood samples.
Zero-Trust Architecture for Point-of-Care Diagnostics
How the BiQadx LIMS Intelligence Node ensures total data segregation and AES-256 encryption at the edge, preparing the ecosystem for future compliance standards.
Benchtop Calibration Mechanics for Dr. POCT
A technical deep dive into the automated onboard optical calibration sequence. Reducing human-error interventions from prototype to pilot stage.
Lyophilized Reagent Stability in Extreme Environments
Evaluating the thermal degradation curve of our proprietary freeze-dried assay beads across humidity gradients (10% to 95% RH).
Capillary Action Modeling using Finite Element Analysis
Implementing COMSOL Multiphysics to accurately predict sample routing times through 50-micron diagnostic channels on the Stratos Cartridge.
Expert Core.
Five frontier scientific concepts powering the BiQadx diagnostic architecture — each grounded in peer-reviewed physics, chemistry, and computational biology.
Acoustic Cell Lysis-Free Separation
Standing acoustic wave fields generated at 2 MHz within microfluidic channels exert radiation pressure proportional to cell compressibility. Cells with differing acoustic contrast factors stratify into pressure nodes vs. antinodes without membrane disruption — enabling label-free, lysis-free separation of leukocytes from erythrocytes at >97% purity in under 90 seconds.
- Piezoelectric PZT-5A transducers bonded directly to PDMS-glass hybrid channel at matched resonant frequency
- Acoustic contrast factor (Φ) differences exploited: Φ_leukocyte ≈ +0.062 vs Φ_erythrocyte ≈ +0.015
- No sheath fluid required — reduces sample dilution and downstream assay matrix interference
- Downstream optical lysis detection validates separation quality per sample before assay proceeds
Lyophilised Reagent Stability Engineering
Lyophilisation of diagnostic reagents (enzymes, antibodies, oligonucleotide probes) into glassy disaccharide matrices reduces molecular mobility below the glass transition temperature. The Tg >55°C matrix enables >24-month ambient-temperature storage — eliminating cold chain dependence for POCT deployment in low-resource settings.
- Trehalose:sucrose 4:1 ratio optimised via DSC-measured Tg across 5 formulation variants
- Residual moisture content <1% confirmed by Karl Fischer titration per finished lot
Centrifugal Disc Geometry Optimisation
Centrifugal microfluidic discs exploit rotational frequency as the sole pumping mechanism. Siphon valve geometry — defined by capillary burst frequency (ω²r∆ρ > 2γcosθ/r²) — enables sequential reagent release at programmable spin speeds without active valves, eliminating external pump hardware and reducing dead volume to <2 µL per chamber.
- Siphon geometry modelled in COMSOL Multiphysics; critical spin frequency ωc verified against contact angle (θ = 68°) and surface tension (γ = 71 mN/m)
- Two-ply polycarbonate disc with pressure-sensitive adhesive layer bonding — autoclavable up to 134°C, 18 psi
Edge-Deployed Synthetic Neural Net
A quantised transformer-style architecture (INT8 precision, 4-bit weight compression) runs entirely on-device ARM Cortex-M7 microcontrollers — no cloud inference required. The model interprets multi-channel fluorescence decay curves, extracting analyte concentration via learned non-linear calibration manifolds that outperform traditional 4PL fits in low-signal regimes.
- Fluorescence decay curve fitting via learned 8-layer attention stack — beats 4PL R² by +0.021 in low-signal assay benchmarks
- All model outputs labelled decision-support only — clinician accountability preserved
Polymer Autofluorescence Suppression
Common diagnostic polymers (PMMA, PS) exhibit intrinsic fluorescence emission peaked at 440–500 nm from carbonyl and aromatic residue impurities — directly overlapping FAM/FITC detection windows. Cyclic olefin polymer (COP) substrates reduce polymer background emission by 8× at 488 nm excitation, enabling sub-picomolar fluorescence detection without time-gating.
- COP Tg ≈ 136°C enables autoclave compatibility at 121°C; optical clarity (Haze <0.5%) maintained post-bonding
- Spectral purity validated by ratiometric background subtraction across 5-channel fluorescence array
Global Impact.
Global NGOs
Deploying POCT hardware in low-resource, high-need environments worldwide, targeting 12+ countries in the first pilot phase.
Research Hospitals
Co-developing clinical assays and verifying LIMS data fidelity at reference labs. ≤40% TAT reduction observed in early bench evaluations.
Biotech Accelerators
Pioneering synthetic biology methods for continuous-monitoring pathogen intelligence. 5 active co-development agreements in place.