We are happy to announce that our paper, "DiGAN: Diffusion-Guided Attention Network for Early Alzheimer's Disease Detection," received the Best Paper Presentation Award at the 1st AI for Healthy Aging and Longevity Workshop (AIAA), held at the 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026) in Singapore. In this work, we introduce DiGAN, a novel deep learning architecture that integrates diffusion models with attention mechanisms to capture subtle and complex neurodegenerative patterns associated with early-stage Alzheimer's disease. By leveraging diffusion-based representation learning to improve data quality and attention-driven feature selection to focus on the most diagnostically relevant biomarkers, the proposed framework improves sensitivity and robustness in early detection scenarios where changes are often minimal and difficult to identify. The award recognizes both the scientific contribution of the method and the clarity of its presentation, advancing AI-driven healthcare solutions that address critical challenges in aging and longevity.

Reference:

Rahman, M. R., Hammouda, M., & Maass, W. (2026). DiGAN: Diffusion-Guided Attention Network for Early Alzheimer's Disease Detection. In 1st AI for Healthy Aging and Longevity Workshop (AIAA): 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026), Singapore.

Contact:
Maxx Richard Rahman – maxx_richard.rahman@dfki.de