Abstract: Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive and functional decline. Early detection is critical to help patients and caregivers better manage symptoms and plan future care. However, current clinical evaluations alone cannot reliably identify AD at pre-dementia stages. Multi-modal neuroimaging provides complementary biomarkers that may aid more accurate machine learning-based diagnosis. This review discusses machine learning methodologies for developing an early AD diagnosis system using integrated data from multiple neuroimaging modalities. Feature extraction, selection, scaling and fusion techniques are described to synergistically combine correlated characteristics from different modalities. Challenges in designing such a system are also outlined. A thematic analysis compares machine learning workflows and their potential for computer-assisted diagnostic solutions. The report aims to advance the field by highlighting strategies that leverage multi-modal neuroimaging data through machine learning for improved early Alzheimer's detection. Automated tools incorporating biomarkers across modalities may help identify candidates for disease- modifying interventions prior to symptom onset.

Keywords: Alzheimer's disease, before symptoms, machine learning, extracted, fused features, multiple neuroimaging modalities

Cite:
Lakshmana Phaneendra Maguluri, Koneru Mahendra Krishna, Yerra Brunda, Alla Poojan Reddy, Chava Kavya Sree, "Leveraging Multi-Modal Neuroimaging Data and Machine Learning for Early Detection of Alzheimer's Disease", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13397.


PDF | DOI: 10.17148/IJARCCE.2024.13397

Open chat
Chat with IJARCCE