Comparison of Convolutional neural network training parameters for detecting Alzheimers disease and effect on visualization
It addresses the need for better interpretability in CNN-based disease detection, but is incremental as it applies existing tools to a specific domain.
This thesis systematically evaluates the influence of CNN hyperparameters on model accuracy for detecting Alzheimer's disease from brain MRI data and compares various visualization methods to assess their quality in highlighting relevant image regions.
Convolutional neural networks (CNN) have become a powerful tool for detecting patterns in image data. Recent papers report promising results in the domain of disease detection using brain MRI data. Despite the high accuracy obtained from CNN models for MRI data so far, almost no papers provided information on the features or image regions driving this accuracy as adequate methods were missing or challenging to apply. Recently, the toolbox iNNvestigate has become available, implementing various state of the art methods for deep learning visualizations. Currently, there is a great demand for a comparison of visualization algorithms to provide an overview of the practical usefulness and capability of these algorithms. Therefore, this thesis has two goals: 1. To systematically evaluate the influence of CNN hyper-parameters on model accuracy. 2. To compare various visualization methods with respect to the quality (i.e. randomness/focus, soundness).