Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Nov 29, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Interpretation and visualization techniques for deep learning models in medical imaging.

Daniel T Huff¹, Amy J Weisman¹, Robert Jeraj^1,2

¹Department of Medical Physics, University of Wisconsin-Madison, Madison WI, United States of America.

Physics in Medicine and Biology

|November 23, 2020

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Illustration of transfer learning from breast cancer detection to risk prediction: adaptation to local data and local objectives.

Journal of medical imaging (Bellingham, Wash.)·2026

Same author

Repeatability of Semi-Quantitative and Volumetric Features from Artificial-Intelligence-Guided Lesion Segmentation on <sup>18</sup>F-DCFPyL PSMA-PET/CT Images: Results from a Test-Retest Cohort.

Tomography (Ann Arbor, Mich.)·2026

Same author

Integrating biomarker-derived individual treatment response assessment into Bayesian trial design for personalized cancer treatment.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026

Same author

Probabilistic clinical target definition with nearest neighbor correlation.

Physics in medicine and biology·2025

Same author

Early assessment of treatment response in AML using FLT PET/cT: a trial of the ECOG-ACRIN Cancer Research Group (EAI141).

Leukemia & lymphoma·2025

Same author

Deep learning-based PSMA PET segmentation repeatability: A post-hoc analysis of a single-center, prospective, test-retest trial.

La Radiologia medica·2025

Same journal

Deep learning-based dose prediction to enhance planning efficiency in cervical brachytherapy with hybrid applicators.

Physics in medicine and biology·2026

Same journal

Corrigendum: Referenceless MR thermometry-a comparison of five methods (2017<i>Phys. Med. Biol</i>.<b>62</b>1-16).

Physics in medicine and biology·2026

Same journal

Corrigendum: Measured and Monte Carlo simulated electron backscatter to the monitor chamber for the varian TrueBeam linac (2016<i>Phys. Med. Biol</i>.<b>61</b>8779).

Physics in medicine and biology·2026

Same journal

Corrigendum: 3D range-modulator for scanned particle therapy: development, Monte Carlo simulations and experimental evaluation (2017<i>Phys. Med. Biol</i>.<b>62</b>7075).

Physics in medicine and biology·2026

Same journal

Recent progress in applications of computing to radiotherapy (ICCR 2016).

Physics in medicine and biology·2026

Same journal

Novel TMS coils designed using an inverse boundary element method.

Physics in medicine and biology·2026

See all related articles

Deep learning (DL) models for medical imaging lack interpretability. This review explores methods to understand DL model structure, function, and output, aiding clinical translation and trust in AI diagnostics.

Area of Science:

Artificial Intelligence in Medicine
Medical Imaging Analysis
Machine Learning Interpretability

Background:

Deep learning (DL) models are increasingly used for medical image analysis.
A critical barrier to clinical adoption is the lack of human interpretability in these complex models.
Interpretability is essential for understanding model behavior and facilitating clinical translation.

Purpose of the Study:

To review and categorize current methods for interpreting medical image analysis models.
To critically evaluate the application of these interpretation methods in published medical imaging studies.
To provide guidance for DL practitioners and discuss the importance of interpretability in medical AI.

Main Methods:

Categorization of interpretation methods into understanding model structure/function and understanding model output.

More Related Videos

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Related Experiment Videos

Last Updated: Nov 29, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Discussion of techniques like dimensionality reduction and autoencoders for model inspection.

Explanation of attribution-based methods (saliency maps, class activation maps) for output interpretation.

Main Results:

Methods for understanding model structure/function include inspecting learned features and dimensionality reduction techniques.
Methods for understanding model output include attribution-based approaches generating heatmaps of feature importance.
Published toolkits and limitations of current interpretation methods in medical imaging are summarized.

Conclusions:

Effective model interpretation is crucial for the clinical translation of deep learning in medical imaging.
A comprehensive understanding of interpretation techniques can enhance trust and transparency in AI-driven diagnostics.
Further development and application of interpretation methods are needed to fully realize the potential of DL in healthcare.