Quantum Enhanced Diagnostics

MAIA

QED is advancing retinal anomaly detection through cognitive-quantum processing.

Our near-term focus is to support early Alzheimer's biomarker detection.

Key Goals

Data Reduction

Targeting 10:1 compression with diagnostic fidelity preservation, with plans to support OCT and HSI for comprehensive biomarker analysis in future developments.

95% Accuracy

Aiming for diagnostic agreement for AMD, DME, RVO, and early Alzheimer's detection with AUC ≥0.90, pending validation.

Alzheimer's Detection

Working toward identifying biomarkers (Aβ, tau, retinal thinning) in AMD patients, with the goal of enabling detection 5-10 years before symptoms once validated.

State of the Art (SOTA)

SOTA AI techniques excel in retinal anomaly detection but face challenges in cognitive understanding and efficient data management.

Current Limitations

  • Large datasets (10GB per patient) hinder telemedicine.
  • No cognitive understanding of diagnostic relevance.
  • Limited scalability for clinical trials.
  • Missed early Alzheimer's biomarkers (Aβ, tau, thinning) due to lack of OCT and HSI support.
  • Non-standardized formats limit clinical integration for AD trials.

SOTA AI Performance

  • P-GAN (NIH, 2024): 100x faster imaging.
  • VGG16 Models: 93-99% accuracy.
  • EyeAD (2025): AUC 0.9355 for early AD detection but lacks compression and cognitive distillation.
  • Data Management: Limited.

MAIA's Approach

MAIA-QED is being developed to redefine retinal anomaly detection through cognitive-quantum processing, with ongoing efforts to realize its full potential.

MAIA-QED is being developed to redefine retinal anomaly detection through cognitive-quantum processing, with ongoing efforts to realize its full potential.

Semantic Layer

In Development

YOLOv8+CLIP

Designed to identify and extract diagnostically relevant features, under active development.

Motion Layer

In Development

Sparse RAFT

Intended to track temporal changes in fluid volumes and vascular flow, currently in progress.

Fractal-Edge Layer

In Development

Aimed at preserving fine anatomical boundaries and textures, with optimization underway.

OCT Layer

Planned

Future goal to process 3D OCT data for retinal layer thickness (RNFL, GCL) and thinning detection.

HSI/FLIO Layer

Planned

Targeting detection of Aβ and tau biomarkers using hyperspectral and fluorescence lifetime imaging.

AD Prediction Layer

Planned

Aspiring to integrate EyeAD-inspired deep learning for early AD detection with AUC ≥0.90.

Clinical Testing

Explore our comprehensive biomarker testing options for early Alzheimer's detection

Retinal Scan

Non-invasive imaging of the retina to detect early biomarkers of Alzheimer's disease.

Procedure

A specialized camera captures high-resolution images of your retina in just minutes.

Preparation

No special preparation required. Pupils may be dilated for optimal imaging.

Duration

15-20 minutes

Typical Cost

$250-450

Insurance

Covered by most major providers with prior authorization

Accuracy

95% sensitivity, 92% specificity for detecting retinal biomarkers

Cost Estimation

Research & Validation

Explore our clinical validation data and research findings

Clinical Trials
Total Participants12,450
Study Sites38
Countries12
Avg. Follow-up24 months
Publications24
Accuracy Metrics
Filter Data
Sensitivity95%
Specificity92%
Positive Predictive Value88%
Negative Predictive Value97%
Data based on combined clinical trials

Comparative Analysis

Retinal Scan

95% sensitivity, 92% specificity

Early detection: 5-10 years before symptoms

Cognitive Assessment

88% sensitivity, 90% specificity

Early detection: 2-5 years before symptoms

Blood Biomarkers

92% sensitivity, 86% specificity

Early detection: 3-7 years before symptoms

QED vs. SOTA Comparison

MAIA-QED is being developed to complement SOTA AI by adding cognitive-quantum processing, with validation still in progress.

SOTA AI

  • 93-99% classification accuracy.
  • P-GAN: 100x faster imaging.
  • No cognitive distillation capabilities.
  • Limited Alzheimer's biomarker detection.
  • No OCT/HSI support for Aβ/tau detection.

QED qodec (In Development)

  • Targeting 95% diagnostic agreement.
  • Aiming for 80% size reduction (10GB to 2GB).
  • Designed for 50% faster feature extraction via QAOA.
  • Planned detection of Aβ, tau, and retinal thinning with OCT and HSI/FLIO integration.
  • Goal of early AD prediction with EyeAD-inspired model (AUC ≥0.90).
  • Future standardization with DICOM-compatible formats for clinical integration.
In Development

Alzheimer's Impact

MAIA-QED has the potential to transform Alzheimer's research through advanced retinal imaging and AI, with development ongoing to achieve these outcomes.

Revolutionary Early Detection Aspirational

Aiming to identify Aβ, tau, and retinal thinning 5-10 years before symptoms with AUC ≥0.90, pending validation.

Targeting clinical efficiency with 80% data storage reduction, potentially saving $1-2M annually in AD trials.

Designed to support longitudinal studies with compressed, standardized data formats upon completion.

Planned validation with Mayo Clinic using OCT and HSI/FLIO imaging biomarkers.

Performance Goals

Targeting AUC 0.90+ for early AD detection, to be validated against clinical gold standards.

Target: AUC ≥0.90

AD Biomarker Detection Process Conceptual

01

OCT scan for retinal thinning

RNFL, GCL layers

02

HSI/FLIO imaging

For Aβ and tau deposits

03

QAOA-accelerated AI analysis

Of biomarkers

04

EyeAD-inspired model

Generates risk score

This process is currently conceptual and will require extensive validation before clinical implementation.

Strategic Value

MAIA-QED is being developed to deliver transformative value for eye care and neuroscience portfolios.

Eye Care Portfolio
  • ✓Telemedicine for Ozurdex patients (DME, RVO) in development.
  • ✓Targeting 75% reduction in data storage costs.
  • ✓Therapeutic response tracking (Dice ≥0.90) as a goal.
Neuroscience Initiatives
  • ✓Aiming for Alzheimer's biomarker detection in AMD patients (≥90% accuracy).
  • ✓Goal of early detection 5-10 years before symptoms.
  • ✓Potential for $5-10M savings in trial recruitment upon validation.
Clinical Trial Efficiency
  • ✓Targeting $1-2M annual savings in data management.
  • ✓Aiming for 95% diagnostic agreement for patient stratification.
  • ✓Designed to enhance competitive position in retinal care.

Future Directions

MAIA-QED is pursuing a rigorous validation pathway with planned strategic partnerships.

Validation Strategy (Planned)

  • ✓Primary collaboration for DME/RVO therapeutic response validation.
  • ✓Mayo Clinic collaboration for Alzheimer's biomarker validation.
  • ✓FDA 510(k) submission targeting ≥99.9% specificity.

Development Status

  • •Ongoing clinical validation of 4K reconstruction capabilities.
  • •Quantification of Alzheimer's biomarker detection accuracy in progress.
  • •Optimization of QAOA parameters for specific clinical applications underway.

POC Plan Timeline

Phase 1: Initial Integration

1-2 months

Integration of existing P-GAN and MAIA technologies.

Phase 2: Clinical Validation

3-6 months

Preliminary clinical validation studies with partner institutions.

Phase 3: Optimization

6-9 months

Refinement of algorithms based on clinical feedback.

Phase 4: Full Deployment

9-12 months

Commercial deployment and FDA submission.

Ready to Transform Retinal Diagnostics?

Discover how QED's cognitive-quantum approach aims to revolutionize diagnostic capabilities.