RECORD
One Map Does Not Fit All: Evaluating Saliency Map Explanation on Multi-Modal Medical Images
Computer Vision and Pattern Recognition · Submitted 11 Jul 2021
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our evaluations on 16 commonly used saliency map methods, including a clinician user study, show that although most saliency map methods captured modality importance information in general, most of them failed to highlight modality-specific important features consistently and precisely.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Being able to explain the prediction to clinical end-users is a necessity to leverage the power of AI models for clinical decision support. For medical images, saliency maps are the most common form of explanation. The maps highlight important features for AI model's prediction. Although many saliency map methods have been proposed, it is unknown how well they perform on explaining decisions on multi-modal medical images, where each modality/channel carries distinct clinical meanings of the same underlying biomedical phenomenon. Understanding such modality-dependent features is essential for clinical users' interpretation of AI decisions. To tackle this clinically important but technically ignored problem, we propose the MSFI (Modality-Specific Feature Importance) metric to examine whether saliency maps can highlight modality-specific important features. MSFI encodes the clinical requirements on modality prioritization and modality-specific feature localization. Our evaluations on 16 commonly used saliency map methods, including a clinician user study, show that although most saliency map methods captured modality importance information in general, most of them failed to highlight modality-specific important features consistently and precisely. The evaluation results guide the choices of saliency map methods and provide insights to propose new ones targeting clinical applications.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Being able to explain the prediction to clinical end-users is a necessity to leverage the power of AI models for clinical decision support. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s1 | For medical images, saliency maps are the most common form of explanation. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s2 | The maps highlight important features for AI model's prediction. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s3 | Although many saliency map methods have been proposed, it is unknown how well they perform on explaining decisions on multi-modal medical images, where each modality/channel carries distinct clinical meanings of the same underlying biomedical phenomenon. | gap 1.00 | 0.06 | 0.04 | asserted 0.94 |
| s4 | Understanding such modality-dependent features is essential for clinical users' interpretation of AI decisions. | background 0.88 | 0.04 | 0.03 | asserted 1.00 |
| s5 | To tackle this clinically important but technically ignored problem, we propose the MSFI (Modality-Specific Feature Importance) metric to examine whether saliency maps can highlight modality-specific important features. | contribution 0.95 | 0.89 | 0.05 | asserted 0.98 |
| s6 | MSFI encodes the clinical requirements on modality prioritization and modality-specific feature localization. | method 0.87 | 0.81 | 0.04 | asserted 1.00 |
| s7 | Our evaluations on 16 commonly used saliency map methods, including a clinician user study, show that although most saliency map methods captured modality importance information in general, most of them failed to highlight modality-specific important features consistently and precisely. | result 1.00 | 0.96 | 0.85 | asserted 1.00 |
| s8 | The evaluation results guide the choices of saliency map methods and provide insights to propose new ones targeting clinical applications. | interpretation 0.96 | 0.78 | 0.20 | asserted 0.93 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Being able to explain the prediction to clinical end-users is a necessity to leverage the power of AI models for clinical decision support. For medical images, saliency maps are the most common form of explanation. The maps highlight important features for AI model's prediction. Although many saliency map methods have been proposed, it is unknown how well they perform on explaining decisions on multi-modal medical images, where each modality/channel carries distinct clinical meanings of the same underlying biomedical phenomenon. Understanding such modality-dependent features is essential for clinical users' interpretation of AI decisions. To tackle this clinically important but technically ignored problem, we propose the MSFI (Modality-Specific Feature Importance) metric to examine whether saliency maps can highlight modality-specific important features. MSFI encodes the clinical requirements on modality prioritization and modality-specific feature localization. Our evaluations on 16 commonly used saliency map methods, including a clinician user study, show that although most saliency map methods captured modality importance information in general, most of them failed to highlight modality-specific important features consistently and precisely. The evaluation results guide the choices of saliency map methods and provide insights to propose new ones targeting clinical applications.
Human review
References
How the literature uses it
3 arXiv papers cite this in 6 sentences
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.