Classification of Alzheimer's Disease stages from Magnetic Resonance Images using Deep Learning

Authors

Alejandro Mora-Rubio¹, Mario Alejandro Bravo-Ortiz¹, Sebastián Quiñones-Arredondo¹, Jose Manuel Saborit-Torres², Gonzalo A. Ruz^3,4,5, and Reinel Tabares-Soto^1,3,6

[1] Department of Electronics and Automation, Universidad Autónoma de Manizales, Manizales 170001, Colombia

[2] Unidad Mixta de Imagen Biomédica FISABIO-CIPF, Fundación para el Fomento de la Investigación Sanitario y Biomédica de la Comunidad Valenciana, Valencia 46020, Spain

[3] Universidad Adolfo Ibáñez, Facultad de Ingeniería Ciencias, Santiago, 7941169, Chile

[4] Center of Applied Ecology and Sustainability (CAPES), Santiago, 8331150, Chile

[5] Data Observatory Foundation, Santiago, 7941169, Chile

[6] University of Caldas, 27985, Department of Systems and Informatics, Manizales, Colombia

Materials and Methods

The implementation of DL models and training was done using MONAI framework, it facilitates the development with the included architectures and preprocessing operations, as well as, allowing the use of custom models such as the Siamese3D used in this work.

The FreeSurfer software was utilized in the study and the command line used for processing the data is provided below:

Recon-all -s $SUBJECT_NAME -i $INPUT_PATH -sd $PATH_RESULTS -all -cw256 -ba-labels

The arguments included in the command line are explained as follows:

• -s $SUBJECT_NAME: specifies the ID of the different subjects.
• -i $INPUT_PATH: denotes the path where the images of the subjects are located.
• -sd $PATH_RESULTS: defines the path where the results will be saved.
• -all: applies all of the steps available in FreeSurfer.
• -cw256: reduces the size of the magnetic resonance imaging (MRI) image by cropping it to 256 pixels, which is done using the mri_convert command.
• -ba-labels: includes the volumes of the Brodmann areas in the analysis.

Data

Data from ADNI and OASIS can be accesed by request in their corresponding websites. The search parameters for ADNI database are presented in the IDASearch.pdf file; furthermore, only the 3 Tesla, T1 weighted, sagittal plane images with slice thickness between 1 and 1.5 mm were used. The tsv files on the partition_tables folder contain 10 different train, validation and test partitions maintaining a correct distribution of the subjects among the sets.

Environment

Use anaconda and the provided YAML file to replicate the programming environment conda env create -f pytorch_monai.yml.