Global

This study presents AlzFormer, a novel deep learning framework utilizing spatiotemporal self-attention to classify Alzheimer’s disease (AD), mild cognitive impairment (MCI), and cognitively normal (CN) individuals from structural MRI scans. By modeling MRI volumes as sequential slice-based inputs and fine-tuning a pre-trained TimeSformer model, AlzFormer achieved 94% accuracy and high class-wise F1-scores, while attention map analyses highlighted clinically relevant brain regions, demonstrating both robust performance and interpretability in multiclass AD diagnosis.

With an estimated 2 billion people worldwide suffering from iron deficiency — a condition linked to anemia, impaired childhood brain development, and higher infant mortality — researchers are racing to find more effective nutritional interventions. MIT scientists have developed a promising solution: tiny crystalline particles known as metal-organic frameworks (MOFs) that can be used to fortify everyday foods and beverages, mixed into foods such as bread or sprinkled directly into drinks like coffee or tea.