Imaging algorithm predicts Alzheimer's onset with 99% accuracy
2021
Vytautas Magnus University, Kaunas, Lithuania
One of the possible Alzheimer’s first signs is mild cognitive impairment (MCI), which is the stage between the expected cognitive decline of normal ageing and dementia. Functional magnetic resonance imaging (fMRI) can be used to identify the regions in the brain which can be associated with the onset of Alzheimer’s disease. The earliest stages of MCI often have almost no clear symptoms, but in quite a few cases can be detected by neuroimaging.
The researchers have developed a deep learning-based method that can predict the possible onset of Alzheimer’s disease from brain images.
For the model, a modification of well-known fine-tuned ResNet 18 (residual neural network) was used to classify functional MRI images obtained from 138 subjects. The images fell into six different categories: from healthy through the spectre of mild cognitive impairment (MCI) to Alzheimer’s disease. In total, 51,443 and 27,310 images from The Alzheimer’s Disease Neuroimaging Initiative fMRI dataset were selected for training and validation.
The model was able to effectively find the MCI features in the given dataset, achieving a classification accuracy of 99% for early MCI vs. AD, late MCI vs. AD, and MCI vs. early MCI, respectively.
Analysis of features of Alzheimer’s Disease: detection of early stage from functional brain changes in magnetic resonance images using a finetuned ResNet18 network
Robertas Damaševičius
Added on: 10-07-2021
[1] https://www.mdpi.com/2075-4418/11/6/1071/htm[2] https://www.technologynetworks.com/neuroscience/news/image-algorithm-predicts-alzheimers-onset-with-99-accuracy-353391#:~:text=Researchers%20have%20developed%20a%20deep,of%20over%2099%20per%20cent.
