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

You might also read

Related Articles

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

Sort by
Same authorSame journal

A novel framework for integrating post-mortem images with 3D body models for automatic injury medical documentation.

Journal of forensic and legal medicine·2026
Same author

[Out-of-hospital death confirmation algorithm].

Archiwum medycyny sadowej i kryminologii·2026
Same author

A strategy for distant kinship analysis using historical human remains: First identifications of Stalinist terror victims in Adjara, Georgia.

Forensic science international. Genetics·2026
Same author

Deciphering barley's stress response: metabolomic strategies and phenotypic implications under multiple abiotic stresses.

Metabolomics : Official journal of the Metabolomic Society·2026
Same author

MASHt: Software for statistical analysis of transcriptomes' qualitative features in factorial experiments.

Bioinformation·2025
Same author

Effect of developmental dynamics on WRKY expression in barley with varying phenologies and trichome micromorphologies.

BMC plant biology·2025

Related Experiment Video

Updated: Jan 16, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.4K

A machine learning approach for automated injuries classification on postmortem images.

Kamila Barbara Kalinowska1, Dorota Zawieska1, Sebastian Puchała2

  • 1Department of Photogrammetry, Remote Sensing and Spatial Information Systems, Warsaw University of Technology, Politechniki 1 Square, 00-661, Warsaw, Poland.

Journal of Forensic and Legal Medicine
|September 28, 2025
PubMed
Summary

Artificial Intelligence (AI) aids forensic science by automating postmortem injury detection, specifically bruises and abrasions. AI models achieved high accuracy, improving objective analysis of forensic images.

Keywords:
ClassificationDeep learningForensic medicineInjuries

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

43.4K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.6K

Related Experiment Videos

Last Updated: Jan 16, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.4K
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

43.4K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.6K

Area of Science:

  • Forensic Science
  • Medical Imaging
  • Computer Vision

Background:

  • Forensic science requires objective analysis of postmortem injuries.
  • Automated detection of injuries like bruises and abrasions can improve accuracy and efficiency.
  • Artificial Intelligence (AI) presents novel opportunities for forensic image analysis.

Purpose of the Study:

  • To develop and evaluate AI models for semantic segmentation of postmortem bruises and abrasions.
  • To investigate the performance of different deep learning architectures for forensic injury detection.
  • To optimize AI models for enhanced accuracy in postmortem image analysis.

Main Methods:

  • Collected and preprocessed a dataset of postmortem injury images.
  • Implemented and compared three deep learning architectures: U-Net, FPN, and LinkNet, with EfficientNetB3 and ResNet50 backbones.
  • Utilized a custom loss function, image transformation, and class balancing for model optimization.

Main Results:

  • The best-performing AI model achieved high sensitivity (92.7%) and specificity (98.9%) in detecting postmortem injuries.
  • Demonstrated the effectiveness of AI-driven semantic segmentation for distinguishing between bruises and abrasions.
  • AI models showed significant potential for automated and objective postmortem image analysis.

Conclusions:

  • AI-driven methods offer a promising approach for automated and objective analysis of postmortem images in injury detection.
  • The developed AI models can assist forensic experts in identifying and classifying postmortem injuries.
  • Further research can build upon these findings to advance AI applications in forensic pathology.