# Prediction of Progression to Alzheimer's disease with Deep InfoMax

## TL;DR

This study compares unsupervised Deep InfoMax with supervised CNNs for predicting Alzheimer's progression using brain imaging, showing promising results for DIM's potential in neuroimaging analysis.

## Contribution

It introduces the application of Deep InfoMax variants to Alzheimer's progression prediction and compares its performance with supervised CNNs on a large neuroimaging dataset.

## Key findings

- DIM shows promising results in classifying Alzheimer's progression.
- Unsupervised DIM performs comparably to supervised CNNs.
- The study highlights DIM's potential for future neuroimaging research.

## Abstract

Arguably, unsupervised learning plays a crucial role in the majority of algorithms for processing brain imaging. A recently introduced unsupervised approach Deep InfoMax (DIM) is a promising tool for exploring brain structure in a flexible non-linear way. In this paper, we investigate the use of variants of DIM in a setting of progression to Alzheimer's disease in comparison with supervised AlexNet and ResNet inspired convolutional neural networks. As a benchmark, we use a classification task between four groups: patients with stable, and progressive mild cognitive impairment (MCI), with Alzheimer's disease, and healthy controls. Our dataset is comprised of 828 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Our experiments highlight encouraging evidence of the high potential utility of DIM in future neuroimaging studies.