Privacy Policy

Overview

ODyn (Ocular Dynamics LLC) builds a correlation engine for glaucoma patients that ingests product ingredient data via computer vision and correlates it against user-logged IOP data to surface hidden dietary and environmental triggers. This page explains what personal data we collect, why, how we protect it, and the rights you have over it.

For consumer health data specifically, we also publish a Consumer Health Data Privacy Policy that covers Washington State's My Health My Data Act requirements.

What we collect

• Identity: email, first name, last name, optional screen name, avatar initials.
• Clinical: glaucoma diagnosis, tonometer, baseline IOP, prescribed drops, family history, related conditions.
• Demographics (optional): birthdate, gender, race or ethnicity, country of birth, country of residence, city, state, height, weight, smoking status, caffeine consumption.
• Health measurements: IOP readings, SpO2 (finger and wrist), perfusion index, altitude, sleep quality, stress level.
• Product scans: label images, OCR output, parsed ingredients.
• User observations: visual anomaly notes, diary entries, spike event logs, drops applications.
• Billing: Stripe customer ID and subscription state (Stripe holds your payment method information; we do not).

Why we collect it

• To operate the platform: authentication, profile display, device personalization, subscription billing.
• To power the correlation engine: aggregated, anonymized analysis across users surfaces trigger patterns no individual data point could.
• To stratify insights: demographic context (where provided) may be used in aggregated, de-identified analysis to identify population-level patterns. It is not used for advertising or profiling outside the ODyn platform.

Data minimization: we collect only the data necessary to operate the platform and generate meaningful insights. Optional fields are clearly indicated and are not required for basic use.

We do not sell your personal data, and we do not use it for advertising. If our approach to identifiable data ever changes, we will ask for your explicit, revocable opt-in first — consent will not be inferred from continued use or passive acceptance. Anonymized aggregate data (see retention below) is a different matter: once stripped of all identifiers and individual traceability, it is no longer your personal data and may inform research partnerships. Your subscription supports the continued operation and development of the platform.

How we protect it

• Pseudonymized architecture: your personal identifiers (email, name) live in a separate table from your clinical and measurement data. Health data references a randomly-generated pseudonym ID with no direct link to your identity.
• Encryption: data is encrypted at rest (AES-256) and in transit (TLS).
• Row-level security: database policies enforce that users can only read and modify their own rows.
• Audit logging: queries against sensitive tables are logged for review.
• Narrow service role access: only specific server-side code paths can bypass RLS, and those paths are limited to operations like user provisioning and webhook handling.
• Incident response: we maintain incident response procedures designed to detect, contain, and remediate unauthorized access.

Your rights

You have the following rights over your personal data regardless of where you live:

• Access: Download a complete copy of your records in JSON or CSV format from your Account page any time.
• Correction: Edit any field via your Profile page.
• Deletion: Permanently delete your personal records from your Account page. Your email, name, clinical details, and demographic information are removed. Anonymized aggregate contributions to the community trigger database remain — they are severed from your identity and cannot be reverse-attributed.
• Portability: The export formats above are standard and portable to other tools.
• Withdraw consent: Cancel your subscription or delete your account at any time.

EU residents have additional rights under GDPR, including the right to object to processing, restrict processing, and file a complaint with a supervisory authority.

Data retention

We retain your personal data for as long as your account is active. After account deletion, identifying data is removed immediately. Anonymized aggregate data may persist indefinitely as part of the community dataset — it has no connection to your identity.

Third parties

We use the following categories of service providers:

• Managed PostgreSQL database and authentication
• Application hosting (cloud edge + CDN)
• Object storage for product label images (private, presigned-URL access)
• Payment processing (PCI-compliant; we do not store payment methods)
• Multimodal processing systems for ingredient parsing (data is processed transiently and is not used for model training)
• An AI language provider that powers the optional assistant — it processes a de-identifiedslice of your health context under a business-associate agreement, transiently and not for model training (see “AI assistant” below)
• Public biomedical ontology references (ChEBI, FoodOn, PubChem — these are open public datasets, not processors of your data)
• Error monitoring (aggregated application telemetry)

Data sent to processing systems is handled under strict contractual limitations and is not retained beyond what is necessary to perform the requested operation. A current vendor list is available under a signed data processing agreement.

AI assistant

ODyn offers an optional conversational assistant. It is opt-in — nothing is sent to it unless you turn it on and send a message. When you use it, we assemble a de-identified slice of your data (recent readings, spikes, health metrics, experiments, and your conversation) and send it to a third-party AI language provider to generate a response.

• Your identity — name, email, and account identifiers — is removed before anything leaves ODyn, and free-text is scrubbed of identifying details.
• The provider processes this context under a business-associate agreement, transiently, and does not use it to train models.
• The assistant is informational and educational — it is not medical advice and does not diagnose or change your treatment.

Contact

For privacy questions, deletion requests, or other inquiries, use our privacy request form.