MedChat: a fully offline multimodal AI system for privacy-preserving clinical anamnesis

Recent advances in large language models made it possible to achieve high conversational performance with substantially reduced computational demands, enabling practical on-site deployment in clinical environments. Such progress allows for local integration of AI systems that uphold strict data protection and patient privacy requirements, yet their secure implementation in medicine necessitates careful consideration of ethical, regulatory, and technical constraints. In this study, we introduce MedChat, a locally deployable virtual physician framework that integrates an LLM-based medical chatbot with a diffusion-driven avatar for automated and structured anamnesis. The chatbot was fine-tuned using a corpus of LLM-generated medical dialogues derived from publicly available symptom-disease datasets, enabling scalable and privacy-preserving training. A secure and isolated database interface was implemented to ensure complete separation between patient data and the model's inference process. The avatar component was realized through a conditional diffusion model operating in latent space, trained on researcher video datasets and synchronized with mel-frequency audio features for realistic speech and facial animation. We demonstrate that the complete multimodal pipeline can operate fully offline on consumer-grade hardware while maintaining interactive response times (average latency: 2.9 ± 0.3 s) and stable system performance. Preliminary evaluation of generated dialogue indicates high linguistic coherence, supporting its suitability for structured anamnesis tasks. MedChat provides a privacy-preserving, resource-efficient, and multimodal solution for clinical data collection. While clinical validation is ongoing, the presented framework establishes a foundation for secure, locally deployable AI-assisted anamnesis in real-world healthcare settings.