FAQ
Find answers to the most asked questions right here. If you don’t find your answer, get in touch with us here.
- Devices & Workflow
Who typically uses RetinaLyze?
RetinaLyze is used by:
- Hospitals
- Eye clinics
- Optical chains
- Independent optometrists
- Diabetes clinics
- General Practitioners
Since 2013, more than 7 million screenings have been performed.
The system enables non-specialized staff (e.g., optometrists, nurses, GPs) to capture images while escalating only relevant cases to specialists.
- Devices & Workflow
Which cameras and OCT devices are compatible?
RetinaLyze is hardware-agnostic and supports over 60 fundus camera and OCT brands, including:
- Canon
- Huvitz
- Nidek
- Optomed
- Optopol
- Rodenstock
- Rodenstock Instruments
- Tomey
- Topcon
- Zeiss
- And many more
It works with both desktop and portable devices. See the full updated list here: https://retinalyze.atlassian.net/wiki/spaces/HC/pages/707559599/Input+data+and+device+compatibility
- Business Value & ROI
Where is the return on investment (ROI)?
RetinaLyze can generate ROI through:
- VAT returns (depending on country)
- Direct payment for retinal imaging/OCT exams
- Telemedicine services
- Increased patient traffic through health-focused branding
- Annual recall programs
- Improved anamnesis leading to additional optical sales
- Even one additional patient per day can significantly impact revenue
Additionally, AI triage and telemedicine workflows can reduce manual workload by up to 60%, freeing specialists for higher-value tasks.
- Devices & Workflow
What should be done if the system shows a yellow or red alert?
If RetinaLyze flags an image:
- Yellow alert: Further clinical evaluation is recommended.
- Red alert: Referral to an ophthalmologist is recommended as soon as possible.
RetinaLyze supports referral workflows and telemedicine review to ensure appropriate follow-up.
- Pricing
What if a customer has multiple devices?
Subscription setup depends on the situation.
Best practice for customers with fewer than 50 locations is typically separate accounts.
Enterprise and chain solutions are available with flexible volume alignment.
- Tech & Clinical Performance
What data was used to train the AI?
The AI was trained on large, diverse datasets of retinal images collected from hospitals, research institutions, and screening programs.
- Images were labeled by experienced ophthalmologists and optometrists
- Data includes multiple ethnicities, age groups, and disease severities
- Most data originates from Scandinavian and European sources
This ensures strong generalizability while maintaining high annotation quality.
- Tech & Clinical Performance
What AI technology does RetinaLyze use?
RetinaLyze uses a combination of Deep Learning, Convolutional Neural Networks (CNNs), and rule-based algorithms to analyze retinal images for early signs of eye disease.
This hybrid approach enables automated detection of retinal abnormalities with high accuracy and efficiency.
- Devices & Workflow
Is RetinaLyze redundant if a clinic already has advanced OCT?
No. RetinaLyze complements advanced OCT systems by offering:
- Vendor-neutral, multi-modal analysis (fundus + OCT)
- Unified triage and reporting across mixed device fleets
- Telemedicine integration
- Progression tracking with perimetry integration
- Centralized cloud-based workflow
It enhances operational efficiency and supports delegation and telemedicine at scale.
- Tech & Clinical Performance
Is RetinaLyze optimized for specific diseases or populations?
RetinaLyze is optimized to detect black/white lesions and changes in the optic nerve head. These changes can indicate:
- Diabetic Retinopathy (DR) – especially for diabetic screening programs
- Glaucoma – including at-risk individuals (age, family history, risk factors)
- Age-related Macular Degeneration (AMD) – particularly in elderly populations
The AI is trained on multi-ethnic datasets to support diverse patient populations.
- Regulatory & Compliance
Is RetinaLyze certified?
RetinaLyze is currently transitioning from CE Class I (MDD) to CE Class IIa (MDR).
RetinaLyze is ISO 13485 certified
The product is not FDA certified.
- Regulatory & Compliance
How long are images stored?
Images are stored as long as the customer remains active, in accordance with the Data Processing Agreement.
Storage enables longitudinal comparison and progression analysis.
- Regulatory & Compliance
How is patient consent handled?
The clinic/store (data controller) is responsible for obtaining patient consent.
- Consent is collected prior to retinal imaging
- Typically documented digitally in the patient journal
- Covers storage, analysis, and sharing of medical data
- Data shared: retinal images, OCT scans, and necessary metadata such as patient name and patient identifier
- Patient name and patient identifier can be anonymized or pseudonymized
- Governed by a Data Processing Agreement (DPA)
Find the latest DPA template here:
https://retinalyze.com/dpa
- Pricing
How does the points system work?
RetinaLyze operates on a monthly subscription with included points.
- Points are used per analysis
- Unused points do not accumulate
- Points expire at the end of the month
- Customers can adjust subscription levels if usage changes
Minor monthly overuse (e.g., 421 instead of 400 points) can be balanced across months. Persistent overuse requires subscription adjustment.
- Regulatory & Compliance
How does RetinaLyze comply with GDPR and data protection regulations?
RetinaLyze follows strict GDPR and medical data protection standards, including:
- End-to-end encryption
- Anonymization and pseudonymization options
- Secure EU-based cloud storage
- Strict access control
- Regular audits and compliance checks
RetinaLyze acts as a data processor, while the clinic or retail chain acts as the data controller.
RetinaLyze is also compliant with LGPD (Brazil) and FADP (Switzerland).
- Tech & Clinical Performance
Does RetinaLyze use the ICDR, ETDRS, or AREDS grading scales?
No. RetinaLyze does not directly apply the ICDR (International Clinical Diabetic Retinopathy), ETDRS (Early Treatment Diabetic Retinopathy Study), or AREDS (Age-Related Eye Disease Study) grading scales.
RetinaLyze is designed to detect signs of retinal disease using AI-based image analysis rather than to replicate manual clinical grading scales. The system identifies possible pathological features and provides risk-based triage output to support clinical decision-making and referral workflows.
Clinical diagnosis and formal disease staging according to ICDR, ETDRS, AREDS, or other classification systems remain the responsibility of the examining eye care professional.
- Tech & Clinical Performance
Does RetinaLyze diagnose diseases?
No. RetinaLyze does not provide a medical diagnosis.
The system detects signs of retinal diseases and supports clinical decision-making. It combines AI analysis with telemedicine workflows, enabling professional over-reads when required.
- Pricing
Can customers pay only per use without a points package?
Yes, for larger customers (24,000+ points/year) a license model is available.
RetinaLyze regularly tests business models in individual markets.
- Tech & Clinical Performance
Are there scientific validation studies?
Yes. RetinaLyze has been validated through multiple clinical and real-world studies.
Diabetic Retinopathy (2023)
A study from Aalborg University Hospital and Steno Diabetes Center demonstrated comparable performance to routine clinical grading, with overlapping confidence intervals and an AUC of 93.4%.
Glaucoma (2023)
A published study (DOI: 10.3390/jcm12175485) showed that RetinaLyze Glaucoma performed at the level of five clinical experts, with high agreement across 552 fundus images and an AUC of 95.3%.
More research references are available at:
https://www.retinalyze.com/research