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: Jul 5, 2025

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

42.6K

Predicting pedestrian-involved crash severity using inception-v3 deep learning model.

Md Nasim Khan1, Subasish Das2, Jinli Liu3

  • 1Senior Engineer, AtkinsRealis, 11801 Domain Blvd Suite 500, Austin, TX 78758, United States.

Accident; Analysis and Prevention
|January 14, 2024
PubMed
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

Porcine Deltacoronavirus M Protein Binds NLRP3 to Promote Inflammasome Assembly via Competition with TRIM31.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Joint Associations of Multidimensional Socioeconomic Status and Healthy Lifestyle with Prevalent Hypertension: A Large Population-Based Study in Northwest China.

Nutrients·2026
Same author

Behavioral and temporal dynamics of child pedestrian crash injury patterns: evidence from random parameter modeling.

Scientific reports·2026
Same author

Decoding pedestrian severity at crosswalks using hybrid clustering and random parameter models.

Scientific reports·2026
Same author

Discovery of YTB53, a Marine-Derived MNK-Active Anti-Acute Myeloid Leukemia Lead with Multi-Kinase Activity.

Journal of medicinal chemistry·2026
Same author

Global, regional, and national burden of type 1 diabetes mellitus in children and adolescents: trend and health inequality analyses based on multiple databases, 1990-2021.

Science bulletin·2026
Same journal

Modeling road-segment-level speeding risk of new energy vehicle taxis using a multistage framework with spatial spillover, endogeneity, and nonlinear effects.

Accident; analysis and prevention·2026
Same journal

Role of streetscape feature in pedestrian safety: A modified multi-level multiple membership model.

Accident; analysis and prevention·2026
Same journal

Assessing autonomous driving performance and environmental influencing factors using real-world operational trajectory data.

Accident; analysis and prevention·2026
Same journal

Multi-scale modeling of electric vehicle fatal crash risk: uncovering spatial heterogeneity and infrastructure-land use coupling mechanisms.

Accident; analysis and prevention·2026
Same journal

Differential sensitivity of self-reported driving and collision measures to aspects of shiftwork, sleep, and fatigue.

Accident; analysis and prevention·2026
Same journal

Delving into the visual attention of pedestrians during street crossing under time pressure: An eye-tracking approach.

Accident; analysis and prevention·2026
See all related articles

This study uses a deep learning model (Inception-v3) to predict pedestrian crash severity in Louisiana. Combining over and under-sampling techniques significantly improved prediction accuracy for fatal, injury, and no-injury crashes.

Area of Science:

  • Road safety and transportation science.
  • Artificial intelligence and machine learning applications.
  • Public health and injury prevention.

Background:

  • Pedestrian safety remains a critical concern, with numerous factors contributing to crash severity.
  • Predicting crash outcomes is essential for developing targeted safety interventions.
  • Existing models may not fully capture the complex interactions influencing pedestrian accidents.

Purpose of the Study:

  • To develop and evaluate a novel deep learning model for predicting pedestrian crash severity.
  • To identify key contributing factors to pedestrian crash severity using variable importance analysis.
  • To compare the performance of the deep learning model against traditional methods.

Main Methods:

  • Utilized a five-year dataset (2016-2021) from Louisiana, incorporating 40 variables.
Keywords:
Crash SeverityData BalancingDeepInsightInception-v3PedestrianSynthetic Minority Oversampling Technique

More Related Videos

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

556

Related Experiment Videos

Last Updated: Jul 5, 2025

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

42.6K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

556
  • Applied the Boruta algorithm for feature selection and the DeepInsight technique for data transformation.
  • Employed Random Under Sampling (RUS) and Synthetic Minority Oversampling Technique (SMOTE) to address data imbalance.
  • Developed and tested Inception-v3 deep learning models under various data scenarios.

    • Main Results:

    • The Inception-v3 model combining over and under-sampling demonstrated superior performance across multiple metrics (accuracy, sensitivity, precision, etc.).
    • Achieved high prediction accuracies: 93.5% for fatal, 77.5% for injury, and 85.9% for no-injury crashes.
    • The deep learning approach showed statistically significant improvements over traditional machine learning and statistical models.

    Conclusions:

    • The proposed deep learning model offers a powerful tool for accurate pedestrian crash severity prediction.
    • Key factors influencing severity include pedestrian/driver impairment, speed limits, alcohol, age, and visibility.
    • Findings can inform safety professionals, policymakers, and manufacturers to enhance road safety measures.