Multi-modal fusion network for early Alzheimer's MRI diagnosis outperforms existing methods

A deep cross-branch multimodal fusion network (DCMFNet) using sMRI and rs-fMRI outperformed traditional ML and state-of-the-art DL models in six binary Alzheimer's disease subtype classification tasks, potentially aiding early diagnosis.

• The model used a Logit Adjustment Cross-Entropy loss to mitigate class imbalance, improving recognition of minority subtypes.
• Evaluation was performed on a private clinical dataset; no external validation was reported.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.