Learning objectives
This paper aims to familiarise radiologists with basic anatomical, pathophysiological, and clinical notions of Alzheimer’s disease, the neuroimaging modalities, and specific findings of this condition. Furthermore, we highlight deep-learning-based algorithms with applications in earlier diagnosis, development of prediction models, and treatment research.
Background
According to WHO, dementia affects more than 55 million people worldwide, being the seventh global cause of death, with approximately 60-70% of cases attributed to Alzheimer’s disease (AD). It represents a significant burden on patients, informal caretakers, and global economies. [1] AD is a neurological disorder characterized by a progressive decline of cognitive abilities leading to severe impairment. It is triggered by the accumulation of amyloid-β (Aβ) and tau neurofibrillary tangles (NFTs) with a specific distribution pattern, causing disruption and loss of complex synaptic networks....
Findings and procedure details
Neuroimaging Modalities in ADStructural MRI can assess the degree of brain tissue loss or damage in key regions like the medial temporal lobe (MTL) in the early stages, and the neocortex as the disease progresses. It also helps differentiate AD from other forms of dementiaand has great utility in monitoring disease progression, allowing for therapy-response evaluation and adjustments.[3,4,5] [Fig 2] Brain atrophy is an inevitable consequence of neurodegeneration and an accurate marker of structural brain changes. The earliest MRI changes appear along the hippocampal pathway...
Conclusion
Neuroimaging plays a key role in diagnosing and monitoring AD, as it depicts neuropathological, structural, and functional changes even from the early stages. The recent development of AI and deep-learning-based software provide amazing automated, time-saving solutions that significantly aid neuroradiologists in catching the disease early, staging it accurately and monitoring the evolution and treatment response. AI-based solutions should rapidly find their way into AD routine clinical practice, as they will improve patient’s quality of life.
Personal information and conflict of interest
A. Fulga:
Nothing to disclose
A. C. Ciurescu:
Nothing to disclose
M. Calinescu:
Nothing to disclose
B. Popa:
Nothing to disclose
References
Abdulaziz Alorf, Muhammad Usman Ghani Khan, Multi-label classification of Alzheimer's disease stages from resting-state fMRI-based correlation connectivity data and deep learning, Computers in Biology and Medicine, Volume 151, Part A, 2022, 106240, ISSN 0010-4825, https://doi.org/10.1016/j.compbiomed.2022.106240.
Park, S.W., Yeo, N.Y., Kim, Y. et al. Deep learning application for the classification of Alzheimer’s disease using 18F-flortaucipir (AV-1451) tau positron emission tomography. Sci Rep 13, 8096 (2023). https://doi.org/10.1038/s41598-023-35389-w
Frisoni, G., Fox, N., Jack, C. et al. The clinical use of structural MRI in Alzheimer disease. Nat Rev Neurol...