iHealth - Clinical Intelligence
Our technology · Pillar 05

Security and Governance

An LGPD-first architecture with source anonymization, irreversible hashing, and a complete data catalog.

With access control by clinical domain and full data lineage — from capture to analytical consumption.

Our approach

Governance isn't a feature. It's the foundation of everything we do.

iHealth handles the most sensitive data that exists: patients' clinical information. This demands — and receives — the highest level of protection, governance, and regulatory compliance a health tech company can offer.

Our security and governance architecture was built with an LGPD-first principle from the outset — not bolted on afterward as a compliance layer. Every technical decision, every integration, every new use case goes through a data protection impact assessment before being implemented.

Governance isn't what we do when we're being audited. It's what we do when no one is watching.

The iHealth Team
Ver as soluções
Clinical data security and governance at iHealth
Security pillars

How we protect data at every layer

Anonymization at the source

The removal of personally identifiable information (PII) happens at the source — before data leaves the hospital institution. No identified data travels between the institution and our Data Lake. Name, national ID (CPF/RG), full date of birth, address, and other identifiers are removed or transformed before transmission. Free-text anonymization is performed by iHealth's NLP, which identifies and replaces identifying entities in clinical text.

The removal of personally identifiable information (PII) happens at the source — before data leaves the hospital institution. No identified data travels between the institution and our Data Lake. Name, national ID (CPF/RG), full date of birth, address, and other identifiers are removed or transformed before transmission. Free-text anonymization is performed by iHealth's NLP, which identifies and replaces identifying entities in clinical text.

Irreversible hashing

Identifiers that must be retained for longitudinal follow-up — such as the medical record number — are replaced with irreversible cryptographic hashes. This allows tracking a given patient's clinical journey over time, without being able to reverse the hash to discover the original identity. The hash function is cryptographically secure and documented in the data catalog.

Identifiers that must be retained for longitudinal follow-up — such as the medical record number — are replaced with irreversible cryptographic hashes. This allows tracking a given patient's clinical journey over time, without being able to reverse the hash to discover the original identity. The hash function is cryptographically secure and documented in the data catalog.

Structured data catalog

All datasets on the platform are cataloged with complete metadata: origin, format, transformations applied, authorized purpose of use, domain owner, and expiration date. The catalog is the single source of truth about what exists on the platform — and is consulted before any new use of the data. No data is used for an uncataloged or unauthorized purpose.

All datasets on the platform are cataloged with complete metadata: origin, format, transformations applied, authorized purpose of use, domain owner, and expiration date. The catalog is the single source of truth about what exists on the platform — and is consulted before any new use of the data. No data is used for an uncataloged or unauthorized purpose.

Access control by clinical domain

No user has unrestricted access to the database. Our system follows the principle of least privilege: each user or system accesses only the data needed for their role — segmented by clinical domain (oncology, cardiology, neurology, etc.), originating institution, purpose of use, and hierarchical level. Every access is authenticated with multi-factor authentication (MFA) and logged immutably.

No user has unrestricted access to the database. Our system follows the principle of least privilege: each user or system accesses only the data needed for their role — segmented by clinical domain (oncology, cardiology, neurology, etc.), originating institution, purpose of use, and hierarchical level. Every access is authenticated with multi-factor authentication (MFA) and logged immutably.

Full data lineage

Every piece of data on the platform has traceable lineage from start to finish: which institution it came from, what transformations it underwent, which models processed it, who accessed it, when, for what purpose — and when it expires. This traceability is what allows use in regulatory submissions and ensures any audit finds answers, not gaps.

Every piece of data on the platform has traceable lineage from start to finish: which institution it came from, what transformations it underwent, which models processed it, who accessed it, when, for what purpose — and when it expires. This traceability is what allows use in regulatory submissions and ensures any audit finds answers, not gaps.

LGPD compliance

Our approach to Brazil's General Data Protection Law

Regulatory standards our platform is compatible with.

LGPDANVISACONITECFDA 21 CFR Part 11EMAICH E6 (GCP)ISO 27001LGPDANVISACONITECFDA 21 CFR Part 11EMAICH E6 (GCP)ISO 27001

Documented legal bases

Every data processing activity has an explicit legal basis — health protection, scientific research, or consent — documented in the Record of Processing Activities (ROPA), reviewed by the DPO for each new use case.

Every data processing activity has an explicit legal basis — health protection, scientific research, or consent — documented in the Record of Processing Activities (ROPA), reviewed by the DPO for each new use case.

Dedicated DPO

iHealth has a dedicated Data Protection Officer — an official channel with the ANPD (Brazil's data protection authority) and a point of contact for data subjects wishing to exercise their rights.

iHealth has a dedicated Data Protection Officer — an official channel with the ANPD (Brazil's data protection authority) and a point of contact for data subjects wishing to exercise their rights.

Data Protection Impact Assessment (DPIA)

For every new use case involving clinical data, we conduct a Data Protection Impact Assessment — identifying risks and defining mitigation measures before implementation.

For every new use case involving clinical data, we conduct a Data Protection Impact Assessment — identifying risks and defining mitigation measures before implementation.

Continuous auditing

Our compliance team conducts periodic audits of all processing flows — verifying compliance with documented purposes and the functioning of access controls.

Our compliance team conducts periodic audits of all processing flows — verifying compliance with documented purposes and the functioning of access controls.

Our technology · iHealth

Want to see this technology applied to your context?

Talk to our team and understand how iHealth's technology pillars translate into real results — for clinical research, market access, hospital management, or data auditing.

WhatsApp