Characterizing Distal Tibia Fractures in Motor Vehicle Crashes: Assessing Occupant Risk, Vehicle  Crashworthiness, and Clinical Outcomes

Loading...
Thumbnail Image

TR Number

Date

2026-08-25

Journal Title

Journal ISSN

Volume Title

Publisher

Virginia Tech

Abstract

High-energy distal tibia fractures are a clinically complex injury associated with high rates of clinical complications, chronic pain, and functional limitations. One of the primary causes of severe distal tibia fractures are motor vehicle crashes. Road traffic injuries account for a significant proportion of the global burden of traumatic injuries; accordingly, a large and diverse community has formed with the goal of reducing serious crash injuries to zero. Distal tibia fractures caused by motor vehicle crashes are associated with complications in 20-50% of cases, and high-energy fractures often result in the development of post-traumatic osteoarthritis, costing millions and significantly impacting the quality of life of tens of thou sands of crash occupants in the United States annually. Despite the link between high-energy distal tibia fractures and motor vehicle crashes, a gap in knowledge exists in the current lit erature regarding the relationship between severe fracture loading mechanisms and occupant demographics, vehicle crashworthiness, and crash severity. This knowledge gap limits the capacity of vehicle safety systems to prevent loading conditions necessary for severe distal tibia fractures and hinders progress towards the goal of zero serious road traffic injuries. To address this gap, the goal of this dissertation was to characterize distal tibia fracture incidence and severity in real-world crash data using established statistical designs, novel machine learning algorithms, and physics-based computational modeling. A multivariate logistic regression was trained using data from the Crash Investigation Sampling System to identify distal tibia risk factors in a large-scale crash population. Next, a machine learning classifier was trained using data from the Crash Injury Research and Engineering Network to predict clinical fracture types following a motor vehicle crash without radiology. A SHapley Additive exPlanation interpretability test was used to explain the underlying relationships between predictions of the "black-box" model and details related to the CIREN crash sample. Finally, a novel quantitative computational model of injury severity was used to assess how factors of the crash event influence the loading mechanisms of severe distal tibia fractures.

Description

Keywords

Motor Vehicle Crash Injury Epidemiology, Clinical Fracture Classification, Quantitative Computational Modeling, Machine Learning, Distal Tibia Fracture

Citation