ScienceDirect

Multimodal learning integrates data from different sensory modalities and plays a critical role in diagnosis of Alzheimer’s Disease. However, methodologically, the fusion of multimodal data often faces issues such as limited utilization of complementary information and modality imbalance. To address these problems, we propose a Spatial-Frequency domain fusion and Gradient Modulation Network (SFGM-Net) for multimodal Alzheimer’s Disease diagnosis, which consists of a Spatial-Frequency domain fusion (SFDF) module, and a Uncertainty-weighted Gradient Modulation (UwGM) module.
See full story at ScienceDirect
Sign up for our newsletter!
Get the latest information and inspirational stories for caregivers, delivered directly to your inbox.