This research introduces a novel clinical decision support system that improves mental health diagnosis accuracy while maintaining interpretability.

• Leverages LLMs to translate complex diagnostic manuals into structured logic programs
• Employs constraint logic programming to ensure diagnostic decisions follow clinical guidelines
• Creates an interpretable system where healthcare providers can trace reasoning paths
• Addresses the critical challenge of diagnostic errors in mental health care

This approach has significant implications for healthcare by providing mental health professionals with reliable, transparent diagnostic assistance that can reduce errors while maintaining human oversight in clinical decisions.

Large Language Models for Interpretable Mental Health Diagnosis