The model was trained on OpenBHB dataset.
OpenBHB quasi-raw T1w images were undersampled, leading to final dimensions of 91 × 109 × 91 voxels.
Subsequently, z-score normalization was applied to standardize the intensity values
across the dataset. This step involved calculating the mean and standard deviation of
the voxel intensities and transforming each voxel value to its corresponding z-score.
How to Get Started with the Model
After downloading the files and installing torch, load the model as follows.
from densenet import *
model = load_preTrained_DenseNet(device = torch.device("cpu"), path = "./");
Use device to load the model on cpu or gpu. Use path to specify the path to the DenseNet_params.pt file.
Citation
If you find our work helpful, feel free to give us a cite.
BibTeX:
@article{de2024explainable,
title={Explainable brain age prediction: a comparative evaluation of morphometric and deep learning pipelines},
author={De Bonis, Maria Luigia Natalia and Fasano, Giuseppe and Lombardi, Angela and Ardito, Carmelo and Ferrara, Antonio and Di Sciascio, Eugenio and Di Noia, Tommaso},
journal={Brain Informatics},
volume={11},
number={1},
pages={33},
year={2024},
publisher={Springer}
}
APA:
De Bonis, M. L. N., Fasano, G., Lombardi, A., Ardito, C., Ferrara, A., Di Sciascio, E., & Di Noia, T. (2024). Explainable brain age prediction: a comparative evaluation of morphometric and deep learning pipelines. Brain Informatics, 11(1), 33.
