Manasarthi: AI-Powered Mental Wellness Platforms with Holistic Healing and Emotional Support

Abstract

Abstract - This survey examines AI-powered mental wellness platforms through systematic analysis of seventeen research works (2023-2025), investigating architectural paradigms, AI integration methodologies, and clinical effectiveness. We analyze the Manasarthi platform featuring TypeScript architecture with React frontend, Express backend, Prisma ORM, and multi-provider LLM integration (Gemini, OpenAI, Anthropic, Hugging Face, Ollama). The platform demonstrates conversation memory management, dual-layer crisis detection, validated clinical assessments (GAD-7, PHQ-9, PSS-10, PCL-5, PTQ, TEIQue, Mini-IPIP), and integration of Western clinical psychology with Eastern spiritual traditions. Findings reveal a paradigm shift from rule-based chatbots toward sophisticated LLM implementations with personalized wellness planning. Key challenges include limited clinical validation, engagement sustainability, cultural competence gaps, and ethical considerations. Future enhancements propose parental companion applications with computer vision-based posture analysis for trauma detection, mobile applications, and clinical integration pathways. Keywords: Artificial Intelligence, Mental Health Support, Large Language Models, Conversational AI, Crisis Detection, Personalized Wellness, Multi-Provider Integration, Holistic Healing, Clinical Assessment, Digital Therapeutics

Ähnliche Arbeiten