Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Classification of Bones01:18

Classification of Bones

8.6K
The bones of the human skeletal system are of varied shapes, sizes, and functions. They can be classified based on their shape and function into four major classes: long bones, short bones, flat bones, and irregular bones. Some classifications include a fifth type, the sesamoid bones, as a separate class, whereas others categorize them under short bones.
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The...
8.6K
Bone Remodeling01:40

Bone Remodeling

39.0K
Bone remodeling is a continuous and balanced process of bone resorption by osteoclasts and bone formation by osteoblasts. In adults, it helps maintain bone mass and calcium homeostasis. While mechanical stress can stimulate turnover as part of the normal maintenance and reparative process, several hormones also regulate bone remodeling.
39.0K

You might also read

Related Articles

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

Sort by
Same author

Visualization of small vibrations inside an MRI scanner using video motion amplification.

Medical physics·2026
Same author

Development of an augmented reality system for tracheal intubation guidance of airway management.

Proceedings of SPIE--the International Society for Optical Engineering·2024
Same author

A cross-sectional natural history study of aspartylglucosaminuria.

JIMD reports·2022
Same author

Anterior Pituitary Volume in Patients with Transfusion Dependent Anemias: Volumetric Approaches and Relation to Pituitary MRI‑R2.

Clinical neuroradiology·2021
Same author

Measurement and evaluation of specific absorption rate and temperature elevation caused by an artificial hip joint during MRI scanning.

Scientific reports·2021
Same author

Pituitary iron and factors predictive of fertility status in transfusion dependent thalassemia.

Haematologica·2020

Related Experiment Video

Updated: Nov 5, 2025

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.9K

Probing an AI regression model for hand bone age determination using gradient-based saliency mapping.

Zhiyue J Wang1

  • 1Department of Radiology, Children's Health and University of Texas Southwestern Medical Center, 1935 Medical District Drive, F1-02, Dallas, TX, 75235, USA. zhiyue.wang@utsouthwestern.edu.

Scientific Reports
|May 20, 2021
PubMed
Summary
This summary is machine-generated.

Gradient-based saliency mapping reveals how artificial intelligence (AI) models determine hand bone age from X-rays. The AI uses a holistic approach, with wrist importance decreasing as age increases.

More Related Videos

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

1.1K
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

1.8K

Related Experiment Videos

Last Updated: Nov 5, 2025

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.9K
Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

1.1K
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

1.8K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Understanding AI decision-making is crucial for clinical applications.
  • AI regression models are increasingly used for medical image analysis, including bone age determination.
  • Gradient-based saliency mapping offers a method to interpret "black box" AI models.

Purpose of the Study:

  • To apply gradient-based saliency mapping to an AI regression model for hand bone age determination.
  • To identify which anatomical regions of X-ray images are most influential in the AI's age predictions.
  • To assess the AI model's decision-making process and identify potential vulnerabilities.

Main Methods:

  • Utilized gradient-based saliency mapping on an AI regression model analyzing hand X-ray radiographs.
  • Calculated partial derivatives (PD) of inferred age with respect to pixel intensity as saliency markers.
  • Evaluated 100 test images, calculating the mean of absolute PD values across five anatomical regions of interest.

Main Results:

  • Saliency maps indicated a holistic AI approach to bone age determination.
  • The wrist area was most important for early bone ages, with decreasing importance for older ages.
  • The middle metacarpal region was least important; a muscular region showed high PD but no age information, indicating a vulnerability.

Conclusions:

  • Gradient-based saliency maps provide valuable insights into the decision-making processes of AI regression models in medical imaging.
  • The AI model's reliance on specific regions changes with age, highlighting the complexity of bone age assessment.
  • Identifying vulnerable regions in AI models is essential for improving accuracy and reliability in clinical practice.