From a hunch in the field to testable research and implementation.

• Patient-facing & personal health AI
• Medical RAG / F-RAG
• Decision support for healthcare professionals
• AI agents for healthcare and elder care
• Medical education using AI, XR, and avatars
• Missing information and responsible decisions in medical AI

• Learning-log × LLM-dialogue analytics
• AI mentors and personalized learning
• Dropout prevention, social isolation, and career support
• Learning assessment in the generative-AI era
• The boundary between AI dependence and learner agency

• 1.5-person personal AI
• AI with long-term memory
• Memory ownership and portability
• AI-human connections that reduce loneliness and isolation
• AI and well-being

• Omission-aware evaluation
• Informational health
• Making unreferenced information visible
• Stopping and escalation design for AI decisions
• Human-in-the-loop AI governance

• Disaster medical information agents
• Multi-agent disaster response
• Infrastructure balancing data protection and responsible use
• Synthetic data and digital twins

• Automorphic studies / automorphic engineering
• Automorphic intelligence
• New requirements engineering for the AI era
• Artifacts and organizations that keep adapting
• Integrating futures studies with system development

1. 01

   Framing

   Separating and aligning the field problem and the research question.

2. 02

   Landscape review

   Reviewing prior research and the current technical landscape.

3. 03

   Design

   Hypotheses, metrics, data, ethics, consent, and governance.

4. 04

   Build

   Developing a prototype or research system.

5. 05

   Validation

   Testing both research novelty and effectiveness in the field.

6. 06

   Extension

   Connecting the work to papers, reports, IP, and commercialization.

The goal is neither “a paper only” nor “a system only,” but research that combines novelty with real-world effectiveness. Early conversations about research funding proposals are welcome.