Three purpose-built AI products that unify clinical trial governance, data integrity, and regulatory compliance — from study start to CSR submission — in real time.
The operating system for the clinical trial lifecycle. A protocol-native, AI-powered governance layer that unifies every connected system into one continuously validated, inspection-ready compliance environment.
Aurelyn Trial | OS™ is the industry's first closed-loop clinical trial operating system — purpose-built for sponsors, CROs, and life sciences organizations that cannot afford the cost of clinical entropy. By treating the study protocol as executable logic and anchoring every decision to pre-loaded global regulatory standards, it transforms compliance from a retrospective audit into a proactive, real-time governance function. Built for clinical operations leaders who demand a single, unimpeachable source of truth across every trial system, site, and data point — from Day One through NDA filing.
Each feature is engineered around a direct operational benefit — eliminating the gap between what clinical regulations require and what trials actually deliver.
| Specification | Detail | Standard / Reference | Status |
|---|---|---|---|
| Ingestion Model | Read-only, API-based, real-time. No write access to source systems. | Non-Destructive Observer | Core |
| Data Normalization | Canonical Data Model normalized to CDISC and eTMF Reference Model on every ingestion event | CDISC CDASH / SDTM / ADaM | Core |
| Protocol Parsing | ICH/FDA template-anchored structural parsing. Extracts visit schedule, time-events matrix, reporting windows, endpoint definitions, eligibility criteria | ICH E6(R3) / FDA Protocol Templates | Core |
| Rule Activation | All derived protocol rules require human expert review and 21 CFR Part 11 e-signature before activation | 21 CFR Part 11 | Mandatory |
| Deviation Classification | 4-tier severity model: Critical / Major / Moderate / Minor — derived from protocol definitions and applicable regulations | ICH GCP / Protocol-Defined | Core |
| SAE Reporting Enforcement | 7-day fatal/life-threatening SUSAR and 15-day all other unexpected SUSAR windows enforced with progressive re-notification | ICH E2A / 21 CFR 312.32 | Critical |
| Audit Trail | Immutable, timestamped log of all system events. Non-modifiable by any user. Exportable for regulatory inspection. | 21 CFR Part 11.10(e) | Core |
| Authentication | MFA mandatory. SSO via SAML 2.0 and OAuth 2.0. Unique user identification per 21 CFR Part 11 | NIST SP 800-63B | Core |
| Encryption | AES-256 at rest. TLS 1.3 minimum in transit. End-to-end encryption across all API connector channels | NIST FIPS 140-2 | Core |
| Validation | GAMP 5 Category 4/5 approach. Full IQ/OQ/PQ documentation package. Change control required for all modifications | GAMP 5 / 21 CFR Part 11 | Required |
| Uptime SLA | 99.9% uptime. Active-active geographic redundancy. RTO < 4 hours, RPO < 1 hour | GxP Data Integrity | Core |
| Regulatory Knowledge Base | Pre-loaded with ICH, FDA CFR, EMA, PMDA, CDISC, HIPAA, GDPR, EU AI Act. Controlled versioning. User-immutable. | Global Standards | Core |
| Deployment Model | Enterprise SaaS. Trial-isolated instances. Per-trial protocol-native configuration. | Cloud-Native | Core |
| EU AI Act Classification | High-risk AI system per Annex III. Full human oversight mandate. Transparency documentation included. | EU AI Act 2024/1689 | Compliant |
| Data Privacy | Pseudonymized/de-identified subject data only. GDPR Article 89 and HIPAA Safe Harbor compliant. DPA required with all source system providers. | HIPAA / GDPR | Core |
The core AI intelligence layer that turns fragmented multi-system clinical data into a single, validated, audit-defensible stream of truth — detecting every discrepancy at the moment it occurs, with the regulatory evidence to resolve it.
The Clinical Evidence & Consistency Engine™ is Aurelyn AI Clinical's proprietary deterministic validation core — the intelligence that sits between your disparate trial systems and the decisions your teams must make. Unlike conventional data management tools that flag errors retrospectively, the Engine operates as a continuous, real-time compliance pulse: ingesting evidence from every source system, comparing it against the protocol blueprint and pre-loaded regulatory standards, and surfacing every discrepancy with the specific citation and prescriptive remediation path already attached. Built for Data Management, Clinical Operations, and Regulatory Affairs professionals who cannot wait until a monitoring visit to discover a data integrity failure.
| Specification | Detail | Standard / Reference | Classification |
|---|---|---|---|
| Validation Logic Model | Deterministic — derived exclusively from protocol and regulatory standards. No probabilistic inference on clinical outcomes. | Protocol-Native | Core Design |
| Ingestion Frequency | Real-time via webhook or API polling. Configurable minimum polling interval per system criticality. | Real-Time | Core |
| Deviation Types Detected | Temporal, data integrity, eligibility, documentation, procedural, safety reporting — 6 primary deviation categories | ICH GCP Deviation Framework | Core |
| Severity Tiers | Critical / Major / Moderate / Minor — derived from protocol definitions and applicable regulatory requirements | Protocol-Defined | Core |
| Regulatory Knowledge Base | ICH GCP E6(R3), 21 CFR 312/50/54/11, EMA GCP Directive, PMDA, ICH E2A/E3/E9, MedDRA, CDISC CDASH/SDTM/ADaM, HIPAA, GDPR, EU AI Act | 17+ Standards Pre-Loaded | Immutable |
| Citation Output | Every deviation includes: protocol section, regulatory clause, evidence comparison (expected vs. actual), remediation guidance | ALCOA+ Principles | Core |
| SAE Window Enforcement | 7-day (fatal/life-threatening SUSAR), 15-day (all other unexpected SUSAR), protocol-defined timelines for non-expedited events | ICH E2A / 21 CFR 312.32 | Critical |
| Action Assignment | Responsible party auto-identified based on role mapping and protocol delegation at detection. Escalation chain protocol-defined. | ICH GCP Delegation | Core |
| Action Resolution | Auto-close triggered when correction confirmed in source system via real-time ingestion. Resolution timestamp logged. | Automated | Core |
| Protocol Epoch Versioning | Every amendment creates a new logical epoch. Prior epoch data preserved with full lineage — not retroactively altered. | Data Integrity / 21 CFR 11 | Required |
| Data Integrity Standard | ALCOA+: Attributable, Legible, Contemporaneous, Original, Accurate. Data hashing at ingestion to detect tampering. | FDA Data Integrity Guidance 2018 | Core |
The first operational intelligence layer built specifically for the electronic Trial Master File — transforming the eTMF from a passive document archive into a real-time, continuously monitored compliance system.
The eTMF Intelligence OS™ solves one of the most persistent failures in clinical trial conduct: the eTMF as a static database that teams fill reactively and regulators critique retroactively. Aurelyn AI Clinical's eTMF Intelligence OS™ maps every expected document artifact against the eTMF Reference Model, monitors completeness in real time, tracks expirations and missing signatures across every site and zone, and enforces version currency — ensuring that every site, at every moment, is operating on the correct approved documents. Built for Regulatory Affairs, Clinical Operations, and Quality teams who need their eTMF to be inspection-ready not at close-out, but always.
| Specification | Detail | Reference Standard | Classification |
|---|---|---|---|
| eTMF Standard | TMF Reference Model v3.x — complete zone and artifact mapping for all expected trial documents | eTMF Reference Model | Core |
| Monitoring Mode | Real-time continuous monitoring via read-only eTMF platform API. No document modification capability. | Non-Destructive Observer | Core |
| Completeness Scope | Zone-level, section-level, and site-level completeness tracking. Expected artifacts per eTMF Reference Model mapped per trial type. | eTMF Ref Model Zones 1–16 | Core |
| Signature Monitoring | Investigator, sponsor, CRO, and IRB signatures tracked. Missing/pending signatures flagged with responsible party and GCP citation. | ICH GCP E6(R3) / 21 CFR 312 | Core |
| Expiration Alert Horizons | Configurable: 90-day, 60-day, 30-day, and at-expiration alerts. Document-type-specific alert logic. | ICH GCP / Protocol-Defined | Core |
| Document Types Monitored | Investigator CVs, GCP certifications, financial disclosures, regulatory approvals, IRB/IEC approvals, informed consents, delegation logs, monitoring reports, laboratory certifications, and all protocol/amendment documents | eTMF Reference Model | Core |
| Version Enforcement | Each site's filed documents continuously compared against the current approved version. Amendment-triggered re-verification across all sites. | ICH GCP / 21 CFR 312.62 | Core |
| ICF Version Tracking | Consent form version monitored per site and per subject cohort. Flags any site or subject on a superseded ICF post-amendment. | 21 CFR Part 50 / ICH GCP | Critical |
| Inspection Export | On-demand inspection-readiness report: completeness status, gap log, remediation history, audit trail — mapped to eTMF Reference Model structure. | eTMF Reference Model / GCP | Feature |
| Platform Integrations | Compatible with leading eTMF platforms via read-only API. Platform-agnostic design — no proprietary system lock-in. | API-Native | Core |
| Audit Trail | Immutable log of all gap detections, alert transmissions, acknowledgments, and resolutions. Timestamped to source system ingestion. | 21 CFR Part 11.10(e) | Core |
| Role-Based Access | CRA, Regulatory Affairs, Clinical Lead, QA, and Site Coordinator views — each scoped to relevant eTMF domains and sites | RBAC / 21 CFR Part 11 | Core |
Every Aurelyn AI Clinical enterprise deployment is delivered as a production-grade, validated environment — complete with the regulatory infrastructure, integration support, and human expertise your trial demands from Day One.
From study startup through NDA submission — Aurelyn AI Clinical ensures your clinical evidence is continuously validated, inspection-ready, and structurally sound.