FedMentalCare introduces a novel framework that enables LLMs to analyze mental health data while maintaining patient privacy and regulatory compliance.

• Combines Federated Learning with Low-Rank Adaptation (LoRA) to fine-tune models without exposing sensitive mental health data
• Ensures compliance with healthcare regulations like HIPAA and GDPR
• Enables AI-powered mental health support tools without compromising patient confidentiality
• Demonstrates how advanced AI can be deployed in regulated healthcare environments

This research addresses critical barriers to AI adoption in mental healthcare by prioritizing both clinical utility and data privacy, potentially expanding access to mental health support for underserved populations.

FedMentalCare: Towards Privacy-Preserving Fine-Tuned LLMs to Analyze Mental Health Status Using Federated Learning Framework