Document Type : Review Article

Authors

• Mohammad Danesh-Doust ^{1, 2}
• Farzaneh Nikparast ^{1, 2}

¹ Medical Physics Research Center, Basic Sciences Research Institute, Mashhad University of Medical Sciences, Mashhad, Iran

² Student research committee, Mashhad University of medical sciences, Mashhad, Iran

Abstract

Introduction:
Neurodegenerative diseases are progressive and irreversible brain pathologies that lead to the abnormal accumulation of iron, amyloid-beta plaques, and oxidative stress, before the onset of clinical symptoms. These changes are detectable on Magnetic Resonance Imaging (MRI), and Artificial Intelligence (AI) is revolutionising the rules in medical image analysis. This review discusses the joint use of MRI and artificial intelligence to diagnose and differentiate between neurodegenerative diseases.
Methods:
A literature search was performed using databases including PubMed, Scopus, Web Of Science, ScienceDirect, and Embase, in 2010-2023.Studies focusing on AI-driven MRI analysis for early diagnosis, disease differentiation, and biomarker discovery were included, while case reports and non-English articles were excluded.
Results:
The review demonstrated that compared to the traditional approach, convolutional neural networks (CNN) are much better at analyzing the results of structural and functional MRI. Advanced MRI scans, including Susceptibility-Weighted Imaging (SWI) and Diffusion Tensor Imaging (DTI), combined with AI, enable the detection of microstructural changes years before clinical onset. Furthermore, AI algorithms successfully differentiated between overlapping neurodegenerative phenotypes, while identifying novel radiomic biomarkers for tracking disease progression and evaluating therapeutic responses.
Conclusion:
Integrating advanced neuroimaging with AI transforms neurodegenerative disease management. This synergistic approach enhances early detection, enables precise phenotyping, and facilitates personalized interventions. Future studies should prioritize multi-center validation and interpretable AI models to translate these computational biomarkers into routine clinical practice.

Highlights

Mohammad Danesh-Doust (Google Scholar) (PubMed)

Farzaneh Nikparast (Google Scholar) (PubMed)

 

 

 

 

 

 

Keywords

• Artificial Intelligence
• Neuroimaging
• Cognitive Dysfunction
• Neurodegenerative Diseases
• and Magnetic Resonance Imaging

Main Subjects

• Medical physics