Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 117 of the 120 references it cites were found in a registry or an index, and 2 are not checked yet.

Interpretation and visualization techniques for deep learning models in medical imaging

Open access
Huff, Daniel T, Weisman, Amy J, Jeraj, Robert · 2021 · Physics in Medicine & Biology

Abstract

Deep learning (DL) approaches to medical image analysis tasks have recently become popular; however, they suffer from a lack of human interpretability critical for both increasing understanding of the methods’ operation and enabling clinical translation. This review summarizes currently available methods for performing image model interpretation and critically evaluates published uses of these methods for medical imaging applications. We divide model interpretation in two categories: (1) understanding model structure and function and (2) understanding model output. Understanding model structure and function summarizes ways to inspect the learned features of the model and how those features act on an image. We discuss techniques for reducing the dimensionality of high-dimensional data and cover autoencoders, both of which can also be leveraged for model interpretation. Understanding model output covers attribution-based methods, such as saliency maps and class activation maps, which produce heatmaps describing the importance of different parts of an image to the model prediction. We describe the mathematics behind these methods, give examples of their use in medical imaging, and compare them against one another. We summarize several published toolkits for model interpretation specific to medical imaging applications, cover limitations of current model interpretation methods, provide recommendations for DL practitioners looking to incorporate model interpretation into their task, and offer general discussion on the importance of model interpretation in medical imaging contexts.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1088/1361-6560/abcd17

Cites

117 of 119 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1: Model interpretation techniques can focus either (1) on increasing understanding of internal model structure and function, or (2) on increasing understanding of model output. These two approaches to model interpretation are covered in review Section 2 and Section 3, respectively.
Figure 1: Model interpretation techniques can focus either (1) on increasing understanding of internal model structure and function, or (2) on increasing understanding of model output. These two approaches to model interpretation are covered in review Section 2 and Section 3, respectively.CC BY-NC-ND · PubMed Central
Figure 2: Overview of the model development process (bottom), overlaid with aspects of model interpretation best suited to each phase (top), and the sections of the review in which they are covered. Different approaches to model interpretation are applicable to all steps of the model development process, from initial data collection, model training, and through to model deployment.
Figure 2: Overview of the model development process (bottom), overlaid with aspects of model interpretation best suited to each phase (top), and the sections of the review in which they are covered. Different approaches to model interpretation are applicable to all steps of the model development process, from initial data collection, model training, and through to model deployment.CC BY-NC-ND · PubMed Central
Figure 3: The parts of a convolutional layer are shown to make clear the distinction between a convolutional filter and a feature map (A). The input layer is convolved with a set of filters and an activation function is applied to generate a feature map. Comparison of learned model filters from (Yu et al., 2018) (B) and (Shin et al., 2016) (C). Both examples show filters learned via random initialization (RI) of filters and after transfer learning (TL). Filters in (B) are of size 3×3, and filters in (C) range from size 5×5 to 11×11.
Figure 3: The parts of a convolutional layer are shown to make clear the distinction between a convolutional filter and a feature map (A). The input layer is convolved with a set of filters and an activation function is applied to generate a feature map. Comparison of learned model filters from (Yu et al., 2018) (B) and (Shin et al., 2016) (C). Both examples show filters learned via random initialization (RI) of filters and after transfer learning (TL). Filters in (B) are of size 3×3, and filters in (C) range from size 5×5 to 11×11.CC BY-NC-ND · PubMed Central
Figure 4: Examples of visualizing high-dimensional deep learning features in a 2D projection. (A) tSNE used to classify regions of histopathology images in (Faust et al., 2018). (B) Constraint-based embedding used to visualize brain MRI images from healthy controls and patients with a varying severity of Huntington disease (Plis et al., 2014).
Figure 4: Examples of visualizing high-dimensional deep learning features in a 2D projection. (A) tSNE used to classify regions of histopathology images in (Faust et al., 2018). (B) Constraint-based embedding used to visualize brain MRI images from healthy controls and patients with a varying severity of Huntington disease (Plis et al., 2014).CC BY-NC-ND · PubMed Central
Figure 5: Example uses of perturbation-based attribution methods for model interpretability. In (A), (Kermany et al., 2018) employed occlusion in order to visualize CNN-based diagnosis of retinal pathologies in optical coherence tomography images. (B) compares several approaches to interpretation for identifying congestive heart failure on chest x-ray (Seah et al., 2018). In (C), (Sayres et al., 2019) uses integrated gradients to visualize evidence of diabetic retinopathy on retinal fundus images.
Figure 5: Example uses of perturbation-based attribution methods for model interpretability. In (A), (Kermany et al., 2018) employed occlusion in order to visualize CNN-based diagnosis of retinal pathologies in optical coherence tomography images. (B) compares several approaches to interpretation for identifying congestive heart failure on chest x-ray (Seah et al., 2018). In (C), (Sayres et al., 2019) uses integrated gradients to visualize evidence of diabetic retinopathy on retinal fundus images.CC BY-NC-ND · PubMed Central
Figure 6: Backpropagation-based approaches to model interpretation include: (A) class maximization visualization of malignant and benign breast masses on mammogram (Yi et al., 2017), (B) weakly supervised detection of extra perivascular spaces on brain MR via saliency mapping (Dubost et al., 2019), and (C) class activation map visualization for classifying breast masses on mammograms (Kim et al., 2018).
Figure 6: Backpropagation-based approaches to model interpretation include: (A) class maximization visualization of malignant and benign breast masses on mammogram (Yi et al., 2017), (B) weakly supervised detection of extra perivascular spaces on brain MR via saliency mapping (Dubost et al., 2019), and (C) class activation map visualization for classifying breast masses on mammograms (Kim et al., 2018).CC BY-NC-ND · PubMed Central
Figure 7: Layerwise relevance propagation for model interpretation in (A) diagnosing multiple sclerosis on brain MRI (Eitel et al., 2019), and (B) visualizing evidence for Alzheimer’s disease on brain MRI (Böhle et al., 2019b).
Figure 7: Layerwise relevance propagation for model interpretation in (A) diagnosing multiple sclerosis on brain MRI (Eitel et al., 2019), and (B) visualizing evidence for Alzheimer’s disease on brain MRI (Böhle et al., 2019b).CC BY-NC-ND · PubMed Central
Figure 8: Family tree of attribution-based methods for model interpretation. Dates correspond to when the method was first published. Arrows indicate methods that are developmends or refinements of previous methods.
Figure 8: Family tree of attribution-based methods for model interpretation. Dates correspond to when the method was first published. Arrows indicate methods that are developmends or refinements of previous methods.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 120 references

Cited by

See all 205

Metadata

Authors
Huff, D. T., Weisman, A. J., & Jeraj, R.
Affiliations
University of Wisconsin–Madison (US)
Corresponding author
Jeraj, Robert
Venue
Physics in Medicine & Biology
Published
2021-02-01
Volume
66
Issue
4
Pages
04TR01
ISSN
  • 0031-9155 (ISSN-L)
  • 1361-6560
OpenAlex
W3108932871
Cited by
172
References
120
Citations by year
2026: 29 · 2025: 40 · 2024: 50 · 2023: 24 · 2022: 21 · 2021: 8
FWCI
0.46
Type
journal article
Language
en
Open access
Licence
Locations
  • Physics in Medicine and Biology (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion, cc-by-nc-nd)
Funders
  • National Institutes of Health (T32CA009206)
  • University of Wisconsin Carbone Cancer Center (P30CA014520)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
Keywords
deep learning interpretability, medical image analysis, model visualization, saliency maps, class activation maps, autoencoder, dimensionality reduction, feature visualization, attribution methods, explainable AI, clinical translation
MeSH terms
Deep Learning, Diagnostic Imaging, Humans, Image Processing, Computer-Assisted

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Interpretation and visualization techniques for deep learning models in medical imaging. https://app.cochat.ai/papers/10.1088/1361-6560/abcd17
The paper
Huff, D. T., Weisman, A. J., & Jeraj, R. (2021). Interpretation and visualization techniques for deep learning models in medical imaging. Physics in Medicine & Biology. https://doi.org/10.1088/1361-6560/abcd17

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.