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: Mar 6, 2026

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

8.9K

A hybrid machine learning approach for predicting traffic accident collision severity.

Zhengping Tan1,2,3, Lei Cui1,3, Hao Xu1

  • 1School of Automotive and Transportation, Xihua University, Chengdu, China.

International Journal of Injury Control and Safety Promotion
|March 5, 2026
PubMed
Summary

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

Tripeptide and hexapeptide topical as adjunct to nonablative fractional resurfacing for photodamage: A randomized split-face trial.

Journal of cosmetic dermatology·2020
Same author

Rare case of a basal cell carcinoma with intravascular invasion.

International journal of women's dermatology·2020
Same author

The Complete Genome Sequence of a Bacterial Strain with High Alkalic Xylanase Activity Isolated from the Sludge Near a Papermill.

Current microbiology·2020
Same author

Assessment of treatment tolerance and parental perspective of outpatient pulsed-dye laser treatment for port wine birthmark without general anesthesia in infants and toddlers.

Journal of the American Academy of Dermatology·2020
Same author

Association Between Non-high-density Lipoprotein Cholesterol and 3-Month Prognosis in Patients With Spontaneous Intracerebral Hemorrhage.

Frontiers in neurology·2020
Same author

The immune response after noise damage in the cochlea is characterized by a heterogeneous mix of adaptive and innate immune cells.

Scientific reports·2020

This study introduces a novel neural network to accurately predict traffic accident severity, overcoming challenges like feature coupling and class imbalance for safer autonomous driving systems. The new model significantly improves prediction accuracy.

Area of Science:

  • Traffic safety research
  • Machine learning applications
  • Autonomous driving systems

Background:

  • Predicting traffic accident severity is crucial for autonomous driving safety but is hindered by feature coupling and class imbalance.
  • Existing methods struggle to reliably address these challenges, limiting their practical application.

Purpose of the Study:

  • To develop an effective deep learning model for accurate traffic accident severity prediction.
  • To address feature coupling and class imbalance issues in accident data analysis.
  • To enhance the reliability of autonomous driving safety systems.

Main Methods:

  • A Dynamic and Static Cross Entropy Integrated Neural Network (DSCE-INN) was proposed.
  • A Weighted Injury Coefficient and K-means clustering were used for severity reclassification.
Keywords:
Collision severityautonomous drivingmachine learningneural networksafety

Related Experiment Videos

Last Updated: Mar 6, 2026

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

8.9K
  • Feature decoupling, dynamic-static weighted cross-entropy loss, soft-hard voting, L1 regularization, and focal loss were employed.

    • Main Results:

    • The DSCE-INN model achieved prediction accuracies of 0.782, 0.729, and 0.801.
    • The proposed model significantly outperformed a baseline Artificial Neural Network (ANN).
    • The methods effectively mitigated feature coupling and class imbalance issues.

    Conclusions:

    • The DSCE-INN model demonstrates significant effectiveness and practical value for improving autonomous driving safety.
    • The study successfully addressed key challenges in traffic accident severity prediction.
    • The findings pave the way for more reliable safety systems in autonomous vehicles.