A proposed framework for multimodal diagnosis in Alzheimer’s disease

Alzheimer’s disease (AD) is a very prevalent neurodegenerative disorder, accounting for 60 to 70% of dementia cases. Identifying improved diagnostics for Alzheimer’s is vital because it would provide earlier diagnosis, and thus timely intervention, and possible delay of disease progression. The problem with current diagnosis methods, which utilize clinical symptoms and mental decline, is that they develop later in the course of AD, hence limiting the opportunity for early treatment. This proposal presents a new, minimally invasive diagnostic model for Alzheimer’s disease that integrates cognitive scores, tau PET imaging, blood-derived exosomal biomarkers, retinal imaging, diffusion tensor imaging, and machine learning. The suggested method demonstrated a higher diagnostic accuracy (88.18%) compared to current standards (79.57%), with improved sensitivity and specificity. This method can enable earlier detection of the disease, allow more reliable detection, and allow for proper intervention to take place. It can assist with improved inclusion of patients with Alzheimer’s in clinical trials and support stage-based management of treatment for patients with stage-based configurations. This framework will facilitate a better understanding of Alzheimer’s progression and allow for opportunities to develop more novel, and effective therapies.