Layer-Wise Relevance Propagation for Deep Neural Network Architectures

Binder, Alexander; Bach, Sebastian; Montavon, Grégoire; Müller, Klaus‐Robert; Samek, Wojciech

doi:10.1007/978-981-10-0557-2_87

Cited by 134 publications

(88 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An important improvement in relation to clinical acceptance would be to implement supporting explanation methods in the predictions, such as layerwise relevance propagation, deep Taylor decomposition, pattern attribution, or other DL explanation approaches [34,35]. It is easy to imagine that a model that is more interpretable and supported by explanations would be more easily accepted in the clinic.…”

Section: Black Boxmentioning

confidence: 99%

Early detection of sepsis utilizing deep learning on electronic health record event sequences

Lauritsen

Kalør²,

Kongsgaard³

et al. 2020

Artificial Intelligence in Medicine

124

View full text Add to dashboard Cite

Background: The timeliness of detection of a sepsis incidence in progress is a crucial factor in the outcome for the patient. Machine learning models built from data in electronic health records can be used as an effective tool for improving this timeliness, but so far the potential for clinical implementations has been largely limited to studies in intensive care units. This study will employ a richer data set that will expand the applicability of these models beyond intensive care units. Furthermore, we will circumvent several important limitations that have been found in the literature: 1) Models are evaluated shortly before sepsis onset without considering interventions already initiated. 2) Machine learning models are built on a restricted set of clinical parameters, which are not necessarily measured in all departments. 3) Model performance is limited by current knowledge of sepsis, as feature interactions and time dependencies are hard-coded into the model. Methods: In this study, we present a model to overcome these shortcomings using a deep learning approach on a diverse multicenter data set. We used retrospective data from multiple Danish hospitals over a seven-year period. Our sepsis detection system is constructed as a combination of a convolutional neural network and a long short-term memory network. We suggest a retrospective assessment of interventions by looking at intravenous antibiotics and blood cultures preceding the prediction time. Results: Results show performance ranging from AUROC 0.856 (3 hours before sepsis onset) to AUROC 0.756 (24 hours before sepsis onset). Evaluating the clinical utility of the model, we find that a large proportion of septic patients did not receive antibiotic treatment or blood culture at the time of the sepsis prediction, and the model could therefore facilitate such interventions at an earlier point in time. Conclusion: We present a deep learning system for early detection of sepsis that is able to learn characteristics of the key factors and interactions from the raw event sequence data itself, without relying on a labor-intensive feature extraction work. Our system outperforms baseline models, such as gradient boosting, which rely on specific data elements and therefore suffer from many missing values in our dataset.

show abstract

Section: Black Boxmentioning

confidence: 99%

Early detection of sepsis utilizing deep learning on electronic health record event sequences

Lauritsen

Kalør²,

Kongsgaard³

et al. 2020

Artificial Intelligence in Medicine

124

View full text Add to dashboard Cite

show abstract

“…Attribution, a term introduced by Ancona et al (2018), also referred to as relevance (Bach et al, 2015;Binder et al, 2016;Zintgraf et al, 2017;Robnik-Šikonja and Kononenko, 2008), contribution (Shrikumar et al, 2017), class saliency (Simonyan et al, 2013) or influence (Kindermans et al, 2016;Adler et al, 2016;Koh and Liang, 2017), aims to reveal components of high importance in the input to the DNN and their effect as the input is propagated through the network. Because of this property we can categorize the following methods to the attribution category: occlusion (Guçlütürk et al, 2017), erasure (Li et al, 2016), perturbation (Fong andVedaldi, 2017), adversarial examples (Papernot et al, 2017) and prediction difference analysis (Zintgraf et al, 2017).…”

Section: Attribution Methodsmentioning

confidence: 99%

Explanation Methods in Deep Learning: Users, Values, Concerns and Challenges

Ras¹,

Gerven²,

Haselager³

2018

The Springer Series on Challenges in Machine Learning

156

107

View full text Add to dashboard Cite

show abstract

“…Analysis of the DNN-based background subtraction is needed for discussing the characteristics and the issues. Visualization methods for analyzing DNNs are proposed [28][29][30]. The authors visualized features contributing to classification by DNNs.…”

Section: Related Workmentioning

confidence: 99%

Analytics of Deep Neural Network-Based Background Subtraction

et al. 2018

View full text Add to dashboard Cite

Deep neural network-based (DNN-based) background subtraction has demonstrated excellent performance for moving object detection. The DNN-based background subtraction automatically learns the background features from training images and outperforms conventional background modeling based on handcraft features. However, previous works fail to detail why DNNs work well for change detection. This discussion helps to understand the potential of DNNs in background subtraction and to improve DNNs. In this paper, we observe feature maps in all layers of a DNN used in our investigation directly. The DNN provides feature maps with the same resolution as that of the input image. These feature maps help to analyze DNN behaviors because feature maps and the input image can be simultaneously compared. Furthermore, we analyzed important filters for the detection accuracy by removing specific filters from the trained DNN. From the experiments, we found that the DNN consists of subtraction operations in convolutional layers and thresholding operations in bias layers and scene-specific filters are generated to suppress false positives from dynamic backgrounds. In addition, we discuss the characteristics and issues of the DNN based on our observation.

show abstract

Layer-Wise Relevance Propagation for Deep Neural Network Architectures

Cited by 134 publications

References 12 publications

Early detection of sepsis utilizing deep learning on electronic health record event sequences

Early detection of sepsis utilizing deep learning on electronic health record event sequences

Explanation Methods in Deep Learning: Users, Values, Concerns and Challenges

Analytics of Deep Neural Network-Based Background Subtraction

Contact Info

Product

Resources

About