Keywords: 3D CNN, ADNI, skull-stripping, GradCAM, Guided Backpropagation
TL;DR: Sanity checks with 3DCNN model interpretation can help avoid overfitting and control learnable features for transfer learning
Abstract: In recent years, with the improvement of data collection and preprocessing, as well as the development of deep learning algorithms, there have been more opportunities for applying artificial intelligence to different areas, including neuroimaging. Various model learning pipelines are emerging to study the degree of cognitive impairment in diseases such as Alzheimer's disease (AD). In this study, we explore knowledge transfer for the stability of the 3D computer vision models (CNN) for the classification of AD on ADNI data. To assess the model performance, and the quality of learned patterns and examine the ways of models overfitting we utilize conventional 3DCNN interpretation methods and swap tests. We imply that skull-stripping and knowledge transfer strategies can significantly impact the robustness and reproducibility of learned patterns, and suggest applying swap tests to ensure the model stability.
Registration: I acknowledge that acceptance of this work at MIDL requires at least one of the authors to register and present the work during the conference.
Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.
Paper Type: novel methodological ideas without extensive validation
Primary Subject Area: Transfer Learning and Domain Adaptation
Secondary Subject Area: Learning with Noisy Labels and Limited Data
Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.
Code And Data: https://adni.loni.usc.edu/