Mycelix-Health 🏥

Decentralized Healthcare Infrastructure for Patient-Controlled Medical Records

Mycelix-Health is a Holochain-based healthcare application that puts patients in control of their medical data while enabling seamless, privacy-preserving sharing with healthcare providers, researchers, and insurers.

Vision

Healthcare data belongs to patients, not institutions. Mycelix-Health enables:

• Patient Sovereignty: You own and control your complete medical history
• Selective Sharing: Granular consent for who sees what data
• Provider Verification: MATL trust scoring for healthcare providers
• Clinical Trial Integration: Connect patients with research opportunities
• Cross-hApp Federation: Share reputation across the Mycelix ecosystem
• AI Health Advocacy: Your personal AI that knows your health history and advocates for you
• Zero-Knowledge Health Proofs: Prove health status without revealing private data
• Digital Health Twins: Personalized physiological models for "what if" scenarios
• Data Dividends: Share in the value created when your data advances medicine

Features

Core Capabilities

Feature	Description
Patient Profiles	Demographics, allergies, medications, conditions
Provider Management	Credentials, licenses, certifications with verification
Medical Records	Encounters, diagnoses, procedures, lab results, imaging
Prescriptions	E-prescribing, refills, drug interactions, adherence
Consent Management	Granular, time-bound, revocable consent directives
Clinical Trials	Study enrollment, data collection, adverse events
Insurance	Plans, claims, prior authorizations, EOBs
Mycelix Bridge	Cross-hApp data federation and reputation

Revolutionary Features (Phase 2) 🚀

Feature	Description
AI Health Advocate	Personal AI that knows your health history, prepares you for appointments, tracks provider outcomes, and advocates on your behalf
ZK Health Proofs	Zero-knowledge proofs to verify health status (vaccinations, drug tests, physicals) without revealing underlying health data
Health Twin	Digital physiological model of your body for predictions, simulations, and "what if" scenarios
Data Dividends	When your data contributes to drug discoveries or AI models, you receive fair compensation through transparent attribution chains

Compliance & Standards

• HIPAA Alignment: Privacy, security, and audit requirements
• HL7 FHIR: Standard medical terminology and data models
• FDA 21 CFR Part 11: Clinical trial electronic records
• ICD-10/CPT/LOINC: Standard medical coding systems
• RxNorm/NDC: Medication identification standards

Architecture

mycelix-health/
├── dna/
│   └── dna.yaml           # DNA manifest
├── happ.yaml              # hApp manifest
├── zomes/
│   ├── patient/           # Patient identity & demographics
│   │   ├── integrity/     # Entry types & validation
│   │   └── coordinator/   # Extern functions
│   ├── provider/          # Provider credentials & licensing
│   ├── records/           # Medical records & results
│   ├── prescriptions/     # Medication management
│   ├── consent/           # Access authorization
│   ├── trials/            # Clinical research
│   ├── insurance/         # Claims & coverage
│   ├── bridge/            # Mycelix federation
│   │
│   │ # Revolutionary Features (Phase 2)
│   ├── advocate/          # AI Health Advocate system
│   ├── zkhealth/          # Zero-knowledge health proofs
│   ├── twin/              # Digital health twins
│   ├── dividends/         # Data dividend distribution
│   │
│   │ # Clinical Integration (Phase 3)
│   ├── fhir_mapping/      # FHIR R4 resource mapping
│   ├── cds/               # Clinical Decision Support
│   ├── provider_directory/ # Provider directory & NPI
│   ├── telehealth/        # Telehealth sessions
│   │
│   │ # Equity & Access (Phase 4)
│   ├── sdoh/              # Social Determinants of Health
│   ├── mental_health/     # Mental health pathways
│   ├── chronic_care/      # Chronic disease management
│   ├── pediatric/         # Pediatric lifecycle care
│   │
│   │ # Advanced Research (Phase 5)
│   ├── research_commons/  # De-identified data sharing
│   ├── trial_matching/    # Patient-trial matching
│   ├── irb/               # Decentralized IRB
│   ├── federated_learning/ # Privacy-preserving ML
│   ├── population_health/ # Population analytics
│   │
│   │ # Global Scale (Phase 6)
│   ├── ips/               # International Patient Summary
│   ├── i18n/              # Multi-language support
│   ├── disaster_response/ # Emergency healthcare operations
│   ├── verifiable_credentials/ # W3C VC for health credentials
│   └── mobile_support/    # Mobile sync & offline support
│
├── services/
│   └── ehr-gateway/       # EHR integration service (TypeScript)
│       ├── auth/          # SMART on FHIR OAuth2
│       ├── adapters/      # Epic, Cerner, generic FHIR
│       └── sync/          # Pull/push & conflict resolution
│
├── sdk/                   # TypeScript SDK (@mycelix/health-sdk)
├── docs/                  # Documentation
└── tests/                 # Comprehensive tests

Quick Start

Prerequisites

• NixOS or Linux with Nix
• Holochain 0.6.x
• HDK 0.6 / HDI 0.7

Build

# Enter development environment
nix develop

# Build all zomes
cargo build --release --target wasm32-unknown-unknown

# Package DNA
hc dna pack dna/

# Package hApp
hc app pack .

Development

# Run tests
cargo test

# Run conductor tests
hc sandbox generate --num-sandboxes 2
hc sandbox run 0

Zome Overview

Patient Zome

Manages patient profiles with:

• Demographics and contact information
• Allergies with severity tracking
• Medical conditions and medications
• Mycelix identity linking
• MATL trust scoring

Provider Zome

Handles healthcare provider credentialing:

• NPI and DEA registration
• Medical licenses with expiration tracking
• Board certifications
• Practice locations
• Epistemic classification for claims

Records Zome

Comprehensive medical records:

• Encounters with diagnoses and procedures
• Lab results with LOINC codes
• Imaging studies with DICOM links
• Vital signs with range validation
• Critical result alerting

Prescriptions Zome

Full prescription lifecycle:

• E-prescribing with RxNorm
• Controlled substance tracking
• Pharmacy management
• Drug interaction alerts
• Medication adherence

Consent Zome

Granular access control:

• Data category permissions (14 HIPAA categories)
• Time-bound consents with automatic expiration
• Emergency access (break-glass) with mandatory audit
• Comprehensive audit logging
• HIPAA authorization documents

NEW: Consent Delegation System

• Healthcare proxy with legal documentation
• Caregiver access for family members
• Temporary delegation (travel, recovery)
• Legal guardian support for minors
• Sub-delegation capabilities

NEW: Patient Notification System

• Real-time alerts for data access
• Plain-language summaries ("Dr. Smith viewed your medications")
• Configurable priorities (immediate, daily, weekly, silent)
• Trusted provider silent lists
• Email, push, and SMS delivery options

NEW: Care Team Templates

• One-click consent for common scenarios
• 8 pre-built system templates:
  • Primary Care Team
  • Specialist Referral
  • Hospital Admission
  • Emergency Department
  • Mental Health Provider
  • Pharmacy Access
  • Insurance & Billing
  • Telehealth Visit
• Custom organization templates
• Automatic expiration management

Trials Zome

Clinical research support:

• Trial registration (NCT numbers)
• Participant enrollment
• Visit scheduling and data collection
• Protocol deviation tracking
• Adverse event reporting (MedWatch)

Insurance Zome

Claims and coverage:

• Plan management with coordination
• Claims submission and tracking
• Prior authorization workflows
• Eligibility verification
• Explanation of Benefits

Bridge Zome

Mycelix ecosystem integration:

• Cross-hApp data queries
• Provider verification requests
• Epistemic claim federation
• Reputation aggregation
• Trust score integration

AI Health Advocate Zome 🆕

Your personal health ally:

• Appointment Preparation: AI analyzes your history, suggests questions, identifies medications to discuss
• Health Insights: Pattern recognition across your data with evidence-based recommendations
• Provider Reviews: Track outcomes with specific providers, see aggregate ratings
• Health Alerts: Drug interaction warnings, abnormal results, preventive care reminders
• Second Opinions: Facilitate getting additional expert perspectives
• Medication Checks: Drug-drug interactions, duplicate therapy detection, contraindications

ZK Health Proofs Zome 🆕

Privacy-preserving health verification using zkSTARK proofs:

• 15 Proof Types: Vaccination status, drug screen clear, employment physical, BMI range, etc.
• Zero Knowledge: Prove claims without revealing underlying health data
• Post-Quantum Security: 128+ bit security with Kyber-1024 compatible proofs
• Trusted Attestors: Verified healthcare providers attest to health facts
• Verifiable Credentials: Employers, schools, insurers can verify without seeing PHI
• Selective Disclosure: Prove exactly what's needed, nothing more

Health Twin Zome 🆕

Digital physiological models for personalized medicine:

• Physiological Modeling: Cardiovascular, metabolic, respiratory, renal systems
• Continuous Learning: Integrates data from wearables, labs, visits
• Simulations: "What if I exercise 30 min daily?" with projected outcomes
• Predictions: Risk assessments, metric forecasts with confidence intervals
• Trajectories: Track trends over time with projections
• Model Confidence: Always shows uncertainty - never overconfident

Data Dividends Zome 🆕

Fair compensation when your data creates value:

• Contribution Tracking: Detailed records of what data you share
• Usage Monitoring: See exactly how your data is used
• Attribution Chains: Immutable lineage from contribution to discovery
• Revenue Events: When research produces value, patients share
• Fair Calculation: Weighted by data quality, quantity, uniqueness
• Patient Control: Specify permitted uses, exclude prohibited purposes
• Transparency: Full visibility into dividend calculations

SDOH Screening Zome 🆕

Social determinants of health assessment and intervention:

• Screening Instruments: PRAPARE, AHC-HRSN, WeCare, custom tools
• 5 SDOH Domains: Economic stability, education, healthcare, neighborhood, social/community
• Risk Assessment: No risk through urgent, with domain-specific scoring
• Community Resources: Searchable directory of local services
• Intervention Tracking: Referrals, follow-ups, outcome monitoring
• Patient Summaries: Aggregate view of social needs and support

Mental Health Pathway Zome 🆕

Comprehensive mental health support with regulatory compliance:

• Screening Instruments: PHQ-9, PHQ-2, GAD-7, C-SSRS, AUDIT, DAST-10, PCL-5, MDQ, EPDS, PSC-17
• Severity Tracking: None through severe with crisis level monitoring
• Safety Plans: Warning signs, coping strategies, crisis contacts, reasons for living
• 42 CFR Part 2 Consent: Substance abuse record protection with proper consent management
• Treatment Planning: Goals, medications, modalities, therapy notes
• Crisis Management: Event logging, emergency contacts, intervention tracking

Chronic Care Zome 🆕

Specialized management for chronic conditions:

• Condition Enrollment: Diabetes, heart failure, COPD, CKD with type-specific data
• Care Plans: Goals, medications, self-management tasks, review scheduling
• Condition Metrics: HbA1c/glucose (diabetes), LVEF/BNP (heart failure), FEV1 (COPD), eGFR (CKD)
• Medication Adherence: Tracking with adherence rate calculations
• Clinical Alerts: Info through critical severity with acknowledgment
• Exacerbation Events: Track disease flare-ups with severity and triggers
• Outcome Tracking: Patient-reported outcomes for quality of life

Pediatric Zome 🆕

Complete pediatric lifecycle management:

• Growth Tracking: Weight, height, head circumference with WHO/CDC percentile calculations
• Immunizations: Full CDC schedule, lot tracking, VIS documentation, catch-up scheduling
• Developmental Milestones: ASQ-3 based domains (motor, language, cognitive, social-emotional)
• Well-Child Visits: Comprehensive visit documentation with screening results
• Pediatric Conditions: Age-specific condition tracking with onset/resolution
• School Health Records: Sports physicals, accommodations, emergency contacts
• Adolescent Health: HEADSS assessment, reproductive health, mental health screening
• Newborn Records: Birth details, APGAR, feeding, circumcision, hearing screen

Research Commons Zome 🆕

Privacy-preserving research data sharing:

• De-identification Methods: HIPAA Safe Harbor, Expert Determination, k-anonymity, differential privacy
• Dataset Management: Create, search, and manage de-identified research datasets
• Access Agreements: Formal agreements with approved uses and restrictions
• Contribution Tracking: Record and audit all data contributions
• Data Quality: Automated quality reports with completeness and consistency metrics
• Usage Auditing: Complete audit trail of all data access

Trial Matching Zome 🆕

Intelligent patient-trial matching:

• Eligibility Criteria: Structured inclusion/exclusion criteria with comparison operators
• Matching Profiles: Patient data mapped to matchable attributes (diagnoses, labs, vitals)
• Patient Preferences: Travel distance, visit frequency, placebo acceptance, language preferences
• Match Results: Scored matches with criteria met/not met/indeterminate
• Notifications: Patient-controlled notifications for new matching opportunities
• Provider Review: Workflow for provider recommendation before patient contact

Decentralized IRB Zome 🆕

Distributed ethics review for research:

• Protocol Submissions: Full protocol submission with risk assessment and consent documents
• IRB Members: Reviewer profiles with roles, credentials, and expertise
• Review Workflow: Individual reviews with votes, comments, required modifications
• Meeting Management: Quorum tracking, attendance, protocol agenda
• Decision Recording: Vote tallies, conditions, approval duration, chair signature
• Continuing Review: Annual progress reports with enrollment and adverse event tracking

Federated Learning Zome 🆕

Privacy-preserving distributed machine learning:

• Learning Projects: Define tasks with model architecture and aggregation strategy
• Participant Management: Join projects with sample counts and public keys
• Training Rounds: Coordinate distributed training with learning parameters
• Model Updates: Encrypted gradient submissions with noise for differential privacy
• Aggregation: FedAvg, FedProx, Secure Aggregation, Byzantine-resilient options
• Privacy Budgets: Track epsilon consumption across training rounds
• Model Evaluation: Validation metrics with confusion matrices and confidence intervals

Population Health Zome 🆕

Aggregate health analytics with privacy:

• Population Statistics: Prevalence, incidence, mortality with confidence intervals
• Health Indicators: Composite scores for geographic regions with benchmarks
• Disease Surveillance: Case counts vs expected with automatic anomaly detection
• Public Health Alerts: Severity-graded alerts with recommendations and acknowledgment
• Disparity Analysis: Stratified analysis by race, income, education with trend tracking
• Quality Indicators: HEDIS-style measures with star ratings and percentiles
• Differential Privacy: All statistics protected with configurable epsilon

International Patient Summary (IPS) Zome 🆕

Cross-border healthcare data exchange following HL7 IPS standard:

• IPS Document: Complete patient summary with FHIR Bundle export
• 8 Clinical Sections: Allergies, medications, problems, immunizations, procedures, medical devices, results, advance directives
• Coding Systems: SNOMED CT, RxNorm, LOINC, ICD-10, ATC, CVX
• Cross-Border Sharing: Track IPS shares with purpose and jurisdiction
• Translation Support: Automatic translation of summaries
• Validation: Ensure compliance with HL7 IPS implementation guide

Internationalization (i18n) Zome 🆕

Multi-language support for global healthcare:

• Locale Management: BCP 47 tags with country/language/script variants
• Translation System: Source strings, translations, plural forms
• Medical Terminology: Domain-specific medical term translations
• User Preferences: Per-user locale settings with fallback chains
• Translation Memory: Fuzzy matching for translation reuse
• Glossary: Authoritative medical terminology with sources
• RTL Support: Right-to-left language rendering

Disaster Response Zome 🆕

Emergency healthcare operations for mass casualty events:

• Disaster Declaration: Multi-level severity (1-5), affected area, emergency contacts
• START Triage: Immediate/Delayed/Minor/Expectant/Deceased categorization
• Resource Management: Personnel, supplies, equipment, beds, transport tracking
• Patient Tracking: Missing, displaced, evacuated, at shelter/facility status
• Emergency Access: Break-glass access with consent waiver and audit
• Shelter Health: Capacity, medical staff, special needs, health alerts
• Resource Requests: Priority-based resource request and fulfillment

Verifiable Credentials Zome 🆕

W3C Verifiable Credentials standard for health credentials:

• 15+ Credential Types: Vaccination, lab results, medical license, insurance, patient identity
• W3C VC Data Model: JSON-LD context, proof types, credential status
• Issuer Management: Trusted issuer registry with key verification
• Holder Wallet: Personal credential storage with favorites and categories
• Presentations: Selective disclosure, presentation requests, verifier flow
• Revocation: Batch revocation registries, status checks
• Trust Registry: Hierarchical trust chains for issuer verification
• Health-Specific Claims: Vaccination records, lab results, medical licenses

Mobile Support Zome 🆕

Mobile-optimized healthcare operations:

• Device Management: Multi-device registration, platform-specific handling
• Offline Sync: Checkpoint-based resumable sync with priority queuing
• Conflict Resolution: Server/client wins, last-write-wins, manual merge
• Delta Sync: Incremental changes for bandwidth optimization
• Push Notifications: FCM/APNS support with quiet hours and type filtering
• QR Codes: Device pairing, record sharing, emergency access
• Biometric Auth: Fingerprint/face authentication integration
• Emergency Snapshot: Offline-available critical health info (allergies, medications, emergency contacts)
• Bandwidth Tracking: Usage monitoring for data-constrained environments

Ecosystem Integration

Mycelix-Health integrates with other Mycelix hApps:

hApp	Integration
Mycelix-Identity	Patient and provider identity verification
Mycelix-DeSci	Research publication linking
Mycelix-Finance	Healthcare payment processing
Mycelix-Justice	Medical malpractice dispute resolution
Mycelix-Marketplace	Medical equipment and supplies

Trust & Reputation

Mycelix-Health uses the Multi-Agent Trust Layer (MATL) for:

• Provider Trust: Based on outcomes, feedback, credentials
• Data Quality: Epistemic classification (E0-E3)
• Claims Verification: Multi-party attestation
• Byzantine Fault Tolerance: Up to 45% malicious actors

Epistemic Levels

Level	Description	Example
E0	Patient-reported	"I feel dizzy"
E1	Provider observation	Physical exam finding
E2	Test confirmed	Lab result
E3	Multi-provider consensus	Tumor board decision

Security

• Agent-centric: Data stored on patient's device
• End-to-end encryption: For sensitive data at rest
• Audit trails: Immutable access logs
• Break-glass: Emergency access with mandatory review
• Consent enforcement: Cryptographic access control

Field-Level Encryption

Sensitive PHI fields are encrypted individually:

• SSN and financial data
• Mental health and substance abuse notes
• Sexual health and genetic data
• Biometric identifiers

Key Management

• Secure key generation and derivation
• Key wrapping for storage
• Automatic rotation policies
• Multi-version support for backwards compatibility

Access Control Enforcement

All data access requires:

1. Patient self-access verification, OR
2. Valid active consent with matching scope, OR
3. Emergency override with mandatory justification
4. Automatic audit logging for all access attempts

Roadmap

Phase 1: Foundation ✅

• ✅ Core zome implementation (8 zomes)
• ✅ Basic patient/provider workflows
• ✅ Consent management
• ✅ Access control enforcement
• ✅ Field-level encryption
• ✅ Key management
• ✅ 292 comprehensive tests

Phase 1.5: Patient Experience ✅

• ✅ Consent delegation system
• ✅ Patient notification system
• ✅ Care team templates
• 🚧 SDK integration

Phase 2: Revolutionary Features ✅

• ✅ AI Health Advocate: Personal AI for appointment prep, insights, alerts
• ✅ ZK Health Proofs: Zero-knowledge proofs with 15 proof types
• ✅ Health Twin MVP: Digital physiological modeling and simulations
• ✅ Data Dividends: Fair compensation system with attribution chains
• ✅ 441 comprehensive tests (149 new for revolutionary features)

Phase 3: Clinical Integration ✅

• ✅ FHIR R4 Mapping: Patient bundles, observations, conditions, medications with LOINC/SNOMED/ICD-10 terminology
• ✅ EHR Gateway: SMART on FHIR authentication, Epic/Cerner adapters, bidirectional sync
• ✅ Clinical Decision Support: Drug interactions (RxNorm), allergy checks, clinical alerts, guidelines
• ✅ Provider Directory: NPI verification, provider search, affiliations, accepting patients
• ✅ Telehealth: Session scheduling, waiting room, documentation, provider availability

Phase 4: Equity & Access ✅

• ✅ SDOH Screening: PRAPARE/AHC-HRSN instruments, 5 domains, community resources, intervention tracking
• ✅ Mental Health Pathway: PHQ-9/GAD-7/C-SSRS screenings, safety plans, 42 CFR Part 2 consent
• ✅ Chronic Disease Management: Diabetes, heart failure, COPD, CKD with condition-specific metrics
• ✅ Pediatric Lifecycle: Growth tracking, CDC immunizations, developmental milestones, well-child visits
• ✅ Accessibility (WCAG 2.1): ARIA labels, screen reader support, contrast checking, reading level adaptation

Phase 5: Advanced Research ✅

• ✅ Research Commons: De-identified data sharing with HIPAA Safe Harbor/Expert Determination, k-anonymity, differential privacy
• ✅ Trial Matching: Eligibility criteria matching, patient preferences, automatic notifications
• ✅ Decentralized IRB: Protocol submissions, reviewer management, voting, continuing reviews
• ✅ Federated Learning: Privacy-preserving ML with FedAvg/SecureAggregation, differential privacy budgets
• ✅ Population Health: Aggregate statistics, surveillance reports, public health alerts, disparity analyses

Phase 6: Global Scale ✅

• ✅ International Patient Summary (IPS): HL7 IPS standard with FHIR Bundle export, 8 clinical sections, cross-border sharing
• ✅ Internationalization (i18n): BCP 47 locales, translation memory, medical terminology, RTL support
• ✅ Disaster Response: Multi-severity declarations, START triage, resource management, emergency access
• ✅ Verifiable Credentials: W3C VC standard with 15+ credential types, issuer/holder/verifier workflows, revocation
• ✅ Mobile Support: Offline sync, conflict resolution, push notifications, QR codes, emergency snapshots

Contributing

See CONTRIBUTING.md for guidelines.

License

Apache 2.0 - See LICENSE

Contact

• Website: mycelix.net
• Email: health@mycelix.net
• GitHub: Luminous-Dynamics/mycelix-health

---

"Returning healthcare data sovereignty to patients through decentralized trust."