Enhancing the Extraction of Interpretable Information for Ischemic Stroke Imaging from Deep Neural Networks
This work addresses interpretability issues in medical imaging for ischemic stroke diagnosis, but it is incremental as it builds on existing LRP methods.
The authors tackled the problem of noisy saliency maps in deep neural networks for ischemic stroke imaging by modifying Layer-wise Relevance Propagation (LRP) on a 3D U-Net trained on the ISLES 2017 dataset, resulting in more sensible visual explanations and linking signal amplitude to information content.
We implement a visual interpretability method Layer-wise Relevance Propagation (LRP) on top of 3D U-Net trained to perform lesion segmentation on the small dataset of multi-modal images provided by ISLES 2017 competition. We demonstrate that LRP modifications could provide more sensible visual explanations to an otherwise highly noise-skewed saliency map. We also link amplitude of modified signals to useful information content. High amplitude localized signals appear to constitute the noise that undermines the interpretability capacity of LRP. Furthermore, mathematical framework for possible analysis of function approximation is developed by analogy.