Saliency methods aim to explain the predictions of deep neural networks. These methods lack reliability when the explanation is sensitive to factors that do not contribute to the model prediction. We use a simple and common pre-processing step -adding a constant shift to the input data-to show that a transformation with no effect on the model can cause numerous methods to incorrectly attribute. In order to guarantee reliability, we posit that methods should fulfill input invariance, the requirement that a saliency method mirror the sensitivity of the model with respect to transformations of the input. We show, through several examples, that saliency methods that do not satisfy input invariance result in misleading attribution.
Saliency methods have emerged as a popular tool to highlight features in an input deemed relevant for the prediction of a learned model. Several saliency methods have been proposed, often guided by visual appeal on image data. In this work, we propose an actionable methodology to evaluate what kinds of explanations a given method can and cannot provide. We find that reliance, solely, on visual assessment can be misleading. Through extensive experiments we show that some existing saliency methods are independent both of the model and of the data generating process. Consequently, methods that fail the proposed tests are inadequate for tasks that are sensitive to either data or model, such as, finding outliers in the data, explaining the relationship between inputs and outputs that the model learned, and debugging the model. We interpret our findings through an analogy with edge detection in images, a technique that requires neither training data nor model. Theory in the case of a linear model and a single-layer convolutional neural network supports our experimental findings 2 . * Work done during the Google AI Residency Program. 2 All code to replicate our findings will be available here: https://goo.gl/hBmhDt 3 We refer here to the broad category of visualization and attribution methods aimed at interpreting trained models. These methods are often used for interpreting deep neural networks particularly on image data. 32nd Conference on Neural Information Processing Systems (NIPS 2018),
To evaluate the trustworthiness of saliency maps for abnormality localization in medical imaging. Materials and Methods:Using two large publicly available radiology datasets (SIIM-ACR Pneumothorax Segmentation and RSNA Pneumonia Detection), we quantified the performance of eight commonly used saliency map techniques in regards to their 1) localization utility (segmentation and detection), 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. We compared their performances versus baseline methods and localization network architectures, using area under the precision-recall curve (AUPRC) and structural similarity index (SSIM) as metrics.Results: All eight saliency map techniques fail at least one of the criteria and were inferior in performance compared to localization networks. For pneumothorax segmentation, the AUPRC ranged from 0.024-0.224, while a U-Net achieved a significantly superior AUPRC of 0.404 (p<0.005). For pneumonia detection, the AUPRC ranged from 0.160-0.519, while a RetinaNet achieved a significantly superior AUPRC of 0.596 (p<0.005). Five and two saliency methods (out of eight) failed the model randomization test on the segmentation and detection datasets, respectively, suggesting that these methods are not sensitive to changes in model parameters. The repeatability and reproducibility of the majority of the saliency methods were worse than localization networks for both the segmentation and detection datasets. Conclusion:We suggest that the use of saliency maps in the high-risk domain of medical imaging warrants additional scrutiny and recommend that detection or segmentation models be used if localization is the desired output of the network.
No abstract
Saliency maps have become a widely used method to make deep learning models more interpretable by providing post-hoc explanations of classifiers through identification of the most pertinent areas of the input medical image. They are increasingly being used in medical imaging to provide clinically plausible explanations for the decisions the neural network makes. However, the utility and robustness of these visualization maps has not yet been rigorously examined in the context of medical imaging. We posit that trustworthiness in this context requires 1) localization utility, 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. Using the localization information available in two large public radiology datasets, we quantify the performance of eight commonly used saliency map approaches for the above criteria using area under the precision-recall curves (AUPRC) and structural similarity index (SSIM), comparing their performance to various baseline measures. Using our framework to quantify the trustworthiness of saliency maps, we show that all eight saliency map techniques fail at least one of the criteria and are, in most cases, less trustworthy when compared to the baselines. We suggest that their usage in the high-risk domain of medical imaging warrants additional scrutiny and recommend that detection or segmentation models be used if localization is the desired output of the network.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.