Alzheimer’s Disease Detection using 3D ResNet-18 on MRI
Updated 6 May 2021
This model detects Alzheimer’s Disease (AD) using the ResNet-18 model on Magnetic Resonance Imaging (MRI). In this model, we propose a method to utilise transfer learning in 3D CNNs, which allows the transfer of knowledge from 2D image datasets (ImageNet) to a 3D image dataset. To build 3D ResNet-18, 2D filters of 2D ResNet-18 were extended in the third dimension to have 3D filters. The remaining layers were adjusted according to the new filters. Then, the entire MRIs were used for training 3D ResNet-18 to make one decision per person.
Our results show that introducing transfer learning to a 3D CNN improves an AD detection system's accuracy. This approach achieved 96.88% accuracy, 100% sensitivity, and 93.75% specificity on our ADNI dataset.
There are currently some sample images in this folder. To have access to more images, you need to send your application to ADNI (http://adni.loni.usc.edu/data-samples/access-data/).
Before applying your MRI data, you should register MRI scans to the MNI space using the SPM12 toolbox.
Ebrahimi, Amir, et al. “Introducing Transfer Learning to 3D ResNet-18 for Alzheimer’s Disease Detection on MRI Images.” 2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ), IEEE, 2020, doi:10.1109/ivcnz51579.2020.9290616.
Ebrahimi, Amir, et al. “Convolutional Neural Networks for Alzheimer’s Disease Detection on MRI Images.” Journal of Medical Imaging, vol. 8, no. 02, SPIE-Intl Soc Optical Eng, Apr. 2021, doi:10.1117/1.jmi.8.2.024503.
MATLAB Release Compatibility
Platform CompatibilityWindows macOS Linux
Inspired by: Pre-trained 3D ResNet-18, Deep Learning Toolbox Model for ResNet-18 Network
Inspired: Alzheimer’s Disease Detection using multi-modal 3D data, Alzheimer’s Disease Detection using multi-modal 3D data, Training 3D CNN models
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!Start Hunting!
Discover Live Editor
Create scripts with code, output, and formatted text in a single executable document.