About Me
I am a Ph.D. candidate in Industrial Design at Eindhoven University of Technology (TU/e), specializing in AI for Health, Multimodal Learning, and Trustworthy Systems.
My research focuses on developing ECG-language models with cross-domain adaptation, multimodal fusion, and evidence-driven reasoning to achieve robust generalization under data scarcity, incomplete information, and distribution shifts in real-world clinical settings.
Research Interests
- AI for Health: ECG analysis, clinical decision support, medical AI systems
- Multimodal Learning: ECG-language models, cross-modal alignment, knowledge fusion
- Trustworthy AI: Interpretable diagnosis, evidence-driven reasoning, robustness under distribution shifts
- Few-shot & Zero-shot Learning: Meta-learning, retrieval-augmented generation
News
- 2026-01-02: Paper accepted at npj Cardiovascular Health - Interpretable multimodal zero-shot ECG diagnosis via structured clinical knowledge alignment
- 2026-01-03: Paper accepted at ICASSP 2026 - UniPACT: A multimodal framework for prognostic question answering on raw ECG and structured EHR
- 2025-01-01: Paper at CHIL 2025 - Electrocardiogram-language model for few-shot question answering with meta learning
- 2025-01-02: Paper at ICASSP 2025 - Electrocardiogram report generation and question answering via retrieval-augmented self-supervised modeling
- 2025-01-03: Paper at ECAI 2025 - Q-Heart: ECG question answering via knowledge-informed multimodal LLMs
Contact
Feel free to reach out via email at j.tang@tue.nl or connect with me on LinkedIn and GitHub.