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 Experiment Video

Updated: Dec 7, 2025

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.8K

Bruise dating using deep learning.

Jhonatan Tirado1, David Mauricio1

  • 1Department of Systems Engineering, Universidad Nacional Mayor de San Marcos, Lima, Peru.

Journal of Forensic Sciences
|September 29, 2020
PubMed
Summary

This study introduces a deep learning model for accurately dating bruises from images, significantly improving upon current medical expertise. The MnasNet model achieves high precision, sensitivity, and specificity in classifying bruise age ranges.

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

Challenges in the Classification of Cardiac Arrhythmias and Ischemia Using End-to-End Deep Learning and the Electrocardiogram: A Systematic Review.

Diagnostics (Basel, Switzerland)·2026
Same author

Maximizing Survival in Pediatric Congenital Cardiac Surgery Using Machine Learning, Explainability, and Simulation Techniques.

Journal of clinical medicine·2024
Same author

Machine Learning Approach for Analyzing 3-Year Outcomes of Patients with Brain Arteriovenous Malformation (AVM) after Stereotactic Radiosurgery (SRS).

Diagnostics (Basel, Switzerland)·2024
Same author

Fourteen years of manifestations and factors of health insurance fraud, 2006-2020: a scoping review.

Health & justice·2021
Same author

Smart water consumption measurement system for houses using IoT and cloud computing.

Environmental monitoring and assessment·2020

Area of Science:

  • Forensic Medicine
  • Computer Vision
  • Medical Imaging Analysis

Background:

  • Bruise dating is crucial for medicolegal cases, particularly in family violence and violence against women.
  • Current medical specialist accuracy in bruise age classification is approximately 50%, limited by image variability and bruise color.
  • Accurate bruise dating requires objective, reliable methods beyond subjective visual assessment.

Purpose of the Study:

  • To develop and evaluate a deep convolutional neural network (CNN) model for automated bruise dating using only photographic images.
  • To establish a robust dataset and methodology for training and validating bruise classification models.
  • To improve the accuracy and reliability of bruise age estimation compared to existing methods.

Main Methods:

Keywords:
MasNetbruise datingconvolutional neural networkdeep learning

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

3.2K

Related Experiment Videos

Last Updated: Dec 7, 2025

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.8K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

3.2K
  • Construction of a 2140-image experimental bruise dataset with a defined data capture protocol and preprocessing procedure.
  • Training of 20 classification models using Inception V3, Resnet50, MobileNet, and MnasNet architectures.
  • Application of transfer learning, cross-validation, and data augmentation techniques to optimize model performance.

    • Main Results:

    • The MnasNet-based classification models demonstrated superior performance, achieving 97.00% precision and sensitivity, and 99.50% specificity.
    • These results significantly exceed the reported 40% precision in the literature for bruise dating.
    • Model precision was observed to decrease as the age of the bruise increased.

    Conclusions:

    • Deep convolutional neural networks, particularly the MnasNet architecture, offer a highly accurate and objective method for bruise dating.
    • The developed model provides a significant advancement in forensic image analysis for medicolegal applications.
    • Further research should explore the impact of bruise age on model performance and potential refinements.