Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/task-force-for-ai-agents-in-healthcare/haarf

⚕️ the world's first unified, lifecycle-centric regulatory framework specifically designed for autonomous AI agents in medical environments (270+ requirements, 3 risk categories).
https://github.com/task-force-for-ai-agents-in-healthcare/haarf

ai ai-agents ai-regulation ai-safety aisvs artificial-intelligence awesome cybersecurity digital-health fda fda-compliance framework healthcare-ai healthcare-compliance iso42001 machine-learning medical-ai open-source regulatory-framework

Last synced: 5 days ago
JSON representation

⚕️ the world's first unified, lifecycle-centric regulatory framework specifically designed for autonomous AI agents in medical environments (270+ requirements, 3 risk categories).

Awesome Lists containing this project

README 

          
# HAARF: Healthcare AI Agents Regulatory Framework



> 🏥 **The world's first comprehensive regulatory compliance framework for autonomous AI agents in healthcare environments**



[![Framework Version](https://img.shields.io/badge/HAARF-v1.0-blue.svg)]()

[![Categories](https://img.shields.io/badge/categories-8-green.svg)]()

[![Requirements](https://img.shields.io/badge/requirements-279-orange.svg)]()

[![License](https://img.shields.io/badge/License-CC%20BY--SA%204.0-blue.svg)](https://creativecommons.org/licenses/by-sa/4.0/)



## 🎯 What is HAARF?



The **Task Force for AI Agents in Healthcare** presents the world's first **unified, lifecycle-centric regulatory framework** specifically designed for autonomous AI agents in medical environments. Our framework synthesizes the most effective elements from global regulatory bodies—FDA, EU AI Act, Health Canada, UK MHRA, NIST, and WHO—into a comprehensive standard that addresses the unique risks and opportunities of agentic AI systems.



**Unlike traditional AI/ML models that provide predictions, AI agents take autonomous actions.** This fundamental difference requires a new regulatory paradigm focused on continuous governance, traceability, and human oversight.



## 🤖 Understanding AI Agents in Healthcare



AI agents are **autonomous, goal-directed systems** that:



### Core Characteristics

- **🔄 Autonomous Decision-Making**: Make independent choices within defined parameters

- **🎯 Goal-Oriented Behavior**: Plan and execute sequences of actions to achieve specific objectives  

- **🔌 Active Integration**: Direct interaction with healthcare systems (EHRs, diagnostic tools, medical devices)

- **📈 Adaptive Learning**: Continuous learning and evolution from real-world clinical data

- **🤝 Human-AI Collaboration**: Augment human capabilities while maintaining human oversight



### Critical Distinction

**Traditional AI**: Input → Prediction → Human Decision  

**AI Agents**: Input → Autonomous Planning → Direct Action → Continuous Learning



![Traditional AI Pipeline vs AI Agent Pipeline](HAARF/1.0/images/1.png)



## 🏛️ Eight Core Verification Categories



Based on comprehensive analysis of global regulatory approaches and insights from our Healthcare AI Agents Working Group meetings with FDA industry committee stakeholders, our framework establishes **eight comprehensive verification categories** for AI agent governance in healthcare, with recent FDA priority enhancements addressing tool-enabled agent regulatory pathways:



### 🔐 **[C1: Unified Risk & Lifecycle Assessment](./HAARF/1.0/en/0x10-C01-Risk-Lifecycle-Assessment.md)**

**Enhanced FDA-EU Model with SaMD Classification & 510(k) Pathway**



- **SaMD Risk Mapping**: Four-tier classification system (Class I-IV) based on healthcare decision state and situation severity

- **FDA 510(k) Predicate Analysis**: Comprehensive substantial equivalence determination for clearance pathway

- **Enhanced PCCP**: FDA-style Predetermined Change Control Plans with clinical evidence requirements

- **Clinical Evidence Scaling**: Risk-based evidence requirements from analytical validation to randomized controlled trials

- **Continuous Monitoring**: Real-time surveillance for model drift and performance degradation

- **Regulatory Sandboxes**: UK MHRA-style testing environments for real-world evidence generation



### 📋 **[C2: Model Passport & Traceability](./HAARF/1.0/en/0x10-C02-Model-Passport-Traceability.md)**

**Complete Lifecycle Documentation & Accountability**



- **Data Lineage**: Full provenance tracking with Health Canada SGBA+ diversity analysis

- **Model Lineage**: Comprehensive documentation of algorithms, training, and updates

- **Decision Lineage**: Explainable AI audit trails for every agent output

- **Machine-Interpretable**: Automated compliance verification and regulatory reporting



### 🛡️ **[C3: Proactive Cybersecurity Framework](./HAARF/1.0/en/0x10-C03-Cybersecurity-Framework.md)**

**[OWASP AISVS](https://github.com/OWASP/AISVS)-Aligned AI-Specific Threat Protection**



- **AISVS Primary Reference**: Organizations align with [OWASP Artificial Intelligence Security Verification Standard (AISVS)](https://github.com/OWASP/AISVS)

- **Adversarial Robustness**: Mandatory testing against prompt injection and evasion attacks

- **Supply Chain Integrity**: Full vetting of pre-trained models and dependencies

- **Real-Time Threat Monitoring**: Continuous anomaly detection for agent behavior

- **Healthcare-Specific Adaptations**: AISVS requirements adapted for clinical environments



### 👤 **[C4: Human Oversight & Integration](./HAARF/1.0/en/0x10-C04-Human-Oversight.md)**

**Clinical Augmentation with Accountability**



- **Clear Accountability**: Defined human responsibility chains throughout agent lifecycle

- **Clinical Oversight**: Human validation required for all critical healthcare decisions

- **Human Factors Engineering**: Mandatory usability testing to prevent automation bias

- **Transparency Requirements**: Agent capabilities and limitations clearly communicated to users



### 🆔 **[C5: Agent Registration & Identity](./HAARF/1.0/en/0x10-C05-Agent-Registration-Identity.md)**

**Comprehensive Agent Visibility & Control**



- **Healthcare Agent Cataloging**: Centralized registries for all AI agents in clinical environments

- **Identity Verification**: Cryptographic authentication and healthcare IAM integration

- **Lifecycle Management**: Complete tracking from deployment to decommissioning

- **Behavioral Monitoring**: Anomaly detection and security event correlation



### 🎛️ **[C6: Autonomy Governance & Control](./HAARF/1.0/en/0x10-C06-Autonomy-Governance.md)**

**Progressive Autonomy with Safety Boundaries**



- **Autonomy Classification**: Standardized levels from co-pilot to full autonomous operation

- **Progressive Implementation**: Safety gates and clinical validation for autonomy advancement

- **Human-Agent Authority**: Clear decision boundaries and emergency override capabilities

- **Multi-Agent Coordination**: Governance for coordinated autonomous agent systems



### ⚖️ **[C7: Bias Mitigation & Population Equity](./HAARF/1.0/en/0x10-C07-Bias-Equity-Population.md)**

**Fairness Across All Patient Populations**



- **Demographic Representativeness**: Comprehensive training data diversity requirements

- **Algorithmic Bias Detection**: Multi-metric fairness assessment and intersectional analysis

- **Vulnerable Population Protection**: Enhanced safeguards for at-risk patient groups

- **Global Health Equity**: Considerations for resource-limited and diverse healthcare settings



### 🔧 **[C8: Tool Use & Integration Security](./HAARF/1.0/en/0x10-C08-Tool-Use-Integration.md)** ⭐ *FDA Priority Enhancement*

**Regulatory Compliance for Tool-Enabled Agents**



- **Tool Authorization Controls**: Role-based access with clinical appropriateness validation

- **Medical Device Integration**: FDA-compliant integration without altering device classification

- **Clinical Tool Selection**: Evidence-based tool selection with contraindication checking

- **Cascading Failure Prevention**: Circuit breakers and isolation mechanisms

- **Regulatory Pathway Analysis**: Tool combination impact on FDA submission requirements

- **Quality System Integration**: Manufacturing controls for tool-enabled agent development



## 🎨 Framework Architecture



![HAARF Framework Overview](HAARF/1.0/images/2.png)



## 🌍 Regulatory Framework Mappings



HAARF provides **comprehensive mappings** to major regulatory frameworks:



| Framework | Coverage | Key Benefits |

|-----------|----------|--------------|

| **[FDA Digital Health](https://www.fda.gov/regulatory-information/search-fda-guidance-documents/marketing-submission-recommendations-predetermined-change-control-plan-artificial-intelligence)** | 84% | 510(k) pathway, PCCP templates, breakthrough device guidance |

| **[EU AI Act](https://artificialintelligenceact.eu/)** | 71% | High-risk classification, human oversight, transparency |

| **[Health Canada SGBA+](https://cihr-irsc.gc.ca/e/50968.html)** | 67% | Comprehensive equity analysis, vulnerable populations |

| **[UK MHRA](https://www.gov.uk/government/organisations/medicines-and-healthcare-products-regulatory-agency)** | 60% | AI Airlock integration, real-world evidence |

| **[NIST AI RMF](https://www.nist.gov/itl/ai-risk-management-framework)** | 88% | Complete 4-function mapping (Govern/Map/Measure/Manage) |

| **[OWASP AISVS](https://github.com/OWASP/AISVS)** | 56% | Primary cybersecurity reference with healthcare adaptations |

| **[WHO GI-AI4H](https://www.who.int/initiatives/global-initiative-on-ai-for-health)** | 48% | Global health equity, ethics alignment |

| **[ISO/IEC 42001](https://www.iso.org/standard/42001)** | 71% | AI management system certification |

| **[IMDRF GMLP](https://www.imdrf.org/documents/good-machine-learning-practice-medical-device-development-guiding-principles)** | 72% | International harmonization foundation |



**Multi-Jurisdictional Value**: Single HAARF compliance supports regulatory submissions across multiple jurisdictions, reducing regulatory burden and enabling global deployment.



![Regulatory Framework Mappings](HAARF/1.0/images/3.png)



## 📁 Repository Structure



```

├── HAARF/

│   └── 1.0/

│       ├── en/                           # Complete Framework (English)

│       │   ├── 0x10-C01-Risk-Lifecycle-Assessment.md

│       │   ├── 0x10-C02-Model-Passport-Traceability.md

│       │   ├── 0x10-C03-Cybersecurity-Framework.md

│       │   ├── 0x10-C04-Human-Oversight.md

│       │   ├── 0x10-C05-Agent-Registration-Identity.md

│       │   ├── 0x10-C06-Autonomy-Governance.md

│       │   ├── 0x10-C07-Bias-Equity-Population.md

│       │   ├── 0x10-C08-Tool-Use-Integration.md     # FDA Priority ⭐

│       │   ├── 0x90-Appendix-A_Glossary.md

│       │   └── Categories.md

│       ├── mappings/                     # Regulatory Framework Mappings

│       │   ├── fda-mapping.json

│       │   ├── eu-ai-act-mapping.json

│       │   ├── health-canada-mapping.json

│       │   ├── uk-mhra-mapping.json

│       │   ├── nist-ai-rmf-mapping.json

│       │   ├── owasp-aisvs-mapping.json

│       │   ├── who-gi-ai4h-mapping.json

│       │   ├── iso-iec-42001-mapping.json

│       │   ├── imdrf-gmlp-mapping.json

│       │   └── README.md

│       └── tools/                        # Assessment and Validation Tools

```



## 🚀 Getting Started



![Getting Started with HAARF - User Paths](HAARF/1.0/images/4.png)



### **For Healthcare Organizations**



1. **Risk Assessment**: Start with [C1 Risk Classification](./HAARF/1.0/en/0x10-C01-Risk-Lifecycle-Assessment.md)

2. **Security Foundation**: Implement [C3 Cybersecurity Framework](./HAARF/1.0/en/0x10-C03-Cybersecurity-Framework.md) 

3. **Human Oversight**: Establish [C4 Clinical Integration](./HAARF/1.0/en/0x10-C04-Human-Oversight.md)

4. **Tool-Enabled Agents**: Apply [C8 Tool Use Security](./HAARF/1.0/en/0x10-C08-Tool-Use-Integration.md) for autonomous systems



### **For AI Developers**



1. **Framework Overview**: Read [Categories.md](./HAARF/1.0/en/Categories.md)

2. **Implementation Guide**: Follow [Using HAARF](./HAARF/1.0/en/0x03-Using-HAARF.md)

3. **Regulatory Mapping**: Review [appropriate jurisdiction mapping](./HAARF/1.0/mappings/)

4. **Compliance Verification**: Implement verification requirements by risk level



### **For Regulatory Affairs**



1. **Regulatory Mappings**: Explore comprehensive [framework mappings](./HAARF/1.0/mappings/)

2. **Submission Guidance**: Use mapping evidence for regulatory submissions

3. **Multi-Jurisdictional Strategy**: Leverage harmonized compliance approach

4. **FDA Priority**: Focus on [C8 Tool Use](./HAARF/1.0/en/0x10-C08-Tool-Use-Integration.md) for tool-enabled agents



## 🎯 FDA Priority: Tool-Enabled Agents



**C8: Tool Use & Integration Security** addresses FDA's highest priority feedback:



> *"The enhanced HAARF framework with C8 (Tool Use) represents the most comprehensive guidance available for medical device companies developing AI agents. It directly addresses the unique regulatory challenges posed by autonomous systems that can access and control multiple healthcare tools."*



**Regulatory Value**:

- **Breakthrough Device Pathway**: Guidance for novel tool-enabled capabilities

- **510(k) Submissions**: Tool integration without device classification alteration  

- **Quality Systems**: Manufacturing controls for tool-enabled agents

- **Post-Market Surveillance**: Tool usage monitoring and adverse event reporting



## 📊 Implementation Levels



HAARF provides **three implementation levels** scaled to risk:



- **Level 1** (85 requirements): Foundational controls for basic healthcare AI agents

- **Level 2** (144 requirements): Standard implementation for moderate-risk agents  

- **Level 3** (50 requirements): Advanced controls for high-risk autonomous agents



![HAARF Implementation Levels](HAARF/1.0/images/5.png)



**Risk-Based Approach**: Higher risk agents require higher verification levels, optimizing regulatory burden while ensuring safety.



## 🤝 Contributing



We welcome contributions from healthcare professionals, AI developers, regulatory experts, and researchers:



- **Issues**: Report bugs, suggest improvements, request clarifications

- **Pull Requests**: Submit enhancements, corrections, and new content

- **Discussions**: Engage in framework development and implementation discussions

- **Regulatory Feedback**: Provide insights from regulatory submissions and approvals



## 📞 Support & Contact



**All Inquiries**: [haarf@quome.site](mailto:haarf@quome.site)



*Technical support, regulatory guidance, security issues, and partnership inquiries*



## ⚖️ License



This work is licensed under the [Creative Commons Attribution-ShareAlike 4.0 International License](https://creativecommons.org/licenses/by-sa/4.0/). 



HAARF is designed for regulatory submissions and compliance across multiple jurisdictions while maintaining open source principles for global healthcare benefit.



---



## 🙏 Framework Foundation: OWASP AISVS



HAARF is built upon the excellent foundational work of the **[OWASP Artificial Intelligence Security Verification Standard (AISVS)](https://github.com/OWASP/AISVS)** project, led by Jim Manico and Russ Memisyazici. 



The OWASP AISVS team created a comprehensive **13-category verification structure** covering:

- Training Data Governance & Bias Management

- Model Lifecycle Management & Change Control  

- Infrastructure & Deployment Security

- **Autonomous Orchestration & Agentic Action Security**

- Adversarial Robustness & Attack Resistance

- Privacy Protection & Personal Data Management

- And 7 additional critical AI security categories



**HAARF's Specialized Adaptation**: We've adapted and extended the OWASP AISVS framework specifically for healthcare environments, adding regulatory compliance requirements, clinical safety protocols, and international harmonization mappings while maintaining the rigorous verification approach that makes AISVS so effective.



**Credit Where Due**: The structured documentation format, categorized verification requirements, and comprehensive security focus that define HAARF all stem from the excellent foundation provided by the [OWASP AISVS](https://github.com/OWASP/AISVS) community. We encourage healthcare AI teams to implement both frameworks - OWASP AISVS for general AI security and HAARF for healthcare-specific regulatory compliance.



---



## 🌟 Recognition



HAARF represents unprecedented collaboration between **40+ international experts** from FDA, EMA, Health Canada, UK MHRA, WHO GI-AI4H, NIST AI RMF, [OWASP AISVS](https://github.com/OWASP/AISVS), and ISO/IEC 42001 communities.



**⭐ Star this repository to support the world's first comprehensive healthcare AI agent regulatory framework!**



---



*Building trust in the AI-augmented future of healthcare through rigorous regulatory compliance and global harmonization.*