Advanced Finite Element Modeling of the Human Head for Monitoring Repeated Sub-Concussive Blast, Impact, and Inertial Exposures

Funding opportunity

This funding opportunity involves the investigation of the potential use of advanced finite element model of the head and brain for the monitoring of repeated sub-concussive exposures to blast, impact, and inertial loads.

Anticipated timeline and budget

Background

Repeated sub-concussive exposures (RSCE) refer to low-magnitude mechanical loads transmitted to the brain that, while insufficient to cause immediate clinical symptoms or diagnosable injury, may accumulate over time and contribute to adverse neurological outcomes. These exposures are prevalent in both civilian and military contexts, including occupational activities involving high-powered tools or demolition, as well as contact sports such as football and soccer. In military training environments, RSCE may arise from exposure to blast waves, weapon system recoil, or high-intensity maneuvers. Despite growing concern, the precise relationship between RSCE frequency, intensity, and long-term cognitive effects remains poorly characterized, necessitating systematic investigation and monitoring.

Understanding the potential health implications of RSCE requires accurate quantification of exposure dosage and longitudinal tracking of associated neurological outcomes. The military context presents a particularly complex challenge due to the diversity of RSCE sources, each with distinct biomechanical characteristics and modes of interaction with neural tissue. For instance, blast-induced overpressure and inertial forces from parachute landings engage different physiological pathways. Finite Element Models (FEM) have been widely employed to simulate brain responses to such mechanical loads; however, current FEMs lack the versatility and computational efficiency to model both blast and impact scenarios in a timely manner, limiting their utility for real-time or long-term exposure assessment.

Recent advancements in computational modeling offer promising avenues for overcoming these limitations. Emerging approaches involve training machine learning algorithms, such as convolutional neural networks, on high-fidelity FEM simulations to enable rapid estimation of brain strain distributions. These surrogate models, when integrated with wearable sensor data, could facilitate near real-time monitoring of RSCE in individual subjects. Furthermore, progress in morphing generic FEMs to subject-specific anatomies suggests the feasibility of scalable, personalized brain models. A flexible, high-resolution FEM capable of simulating diverse RSCE scenarios would represent a critical tool for elucidating dose-response relationships and informing advanced monitoring strategies, enhancing the safety and cognitive health of military personnel and athletes alike.

Scope and Research objectives

The overarching goal of this funding opportunity is to investigate the use of detailed finite element models of the head and brain for the longitudinal monitoring of repeated sub-concussive exposures on military members. This will include:

1. Studying the feasibility of creating a detailed 3D finite element model of the head capable of handling blast, impact, and inertial loading conditions.
2. Investigating the potential of FEM-trained machine learning algorithms to predict brain strain and stresses given a blast, impact, and inertial loading conditions.
3. Investigating scalable approaches to morph generic head FEM to individual-specific head/brain anatomy obtained from medical imaging.

Phases of progress

Phase	Description and tentative schedule for desired outputs
3D FEM for blast, impact, and inertial loads	Assessment and, if achievable, demonstration of the feasibility of a 3D FEM of the head and brain capable of handling blast, impact, and inertial loads, within 18 months of project initiation.
FEM-trained machine learning algorithms	Assessment and identification, if achievable, of machine learning algorithms suitable for training using head/brain FEM input and outputs, within 30 months of project initiation.
Scalable morphing approaches	Assessment and identification, if achievable, of scalable approaches to morph a detailed 3D FEM of the head and brain to a specific head-brain anatomy, within 36 months of project initiation.

3D FEM for blast, impact, and inertial loads

• Identification of the competing requirements for a FEM model of the head and brain to be capable of predicting local brain stresses and strains from multiple military-relevant loading conditions, including blast waves, low energy impacts and high-rate head kinematics.
• Assessment of the achievable level of details of such FEM model.
• Assessment of the potential computational cost for running such FEM model with various types of inputs.
• Assessment of the potential performance of such FEM models against published validation datasets.

FEM-trained machine learning algorithms

• Investigate dataset requirements for using simulation results to train machine learning (ML) algorithm to predict the model output directly from various military-relevant input loads.
• Assessment of potential computing requirements for the training and running such algorithms.
• Assessment of the potential generalization and level of accuracy of such algorithms.
• Assessment of the potential deployability of such algorithms

Scalable morphing approaches

• Investigate imaging data requirement for morphing a detailed head model to a head-brain anatomy extracted from medical imagery data or from generic brain atlas.
• Assessment of the potential generalization and level of accuracy of selected morphing approaches.
• Assessment of the potential scalability of selected morphing approaches.

Desired outputs

The project’s desired outputs comprise of the following:

• Feasibility assessment/data on the use of FEM of the head and brain with various military-relevant loading conditions, including blast, impact, and inertial loading scenarios in the form of a summary report.
• Assessment of the potential of different machine learning algorithms for training using head-brain FEM datasets and for predicting head-brain FEM outputs, in the form of a summary report and/or peer-reviewed scientific publications.
• Assessment of the feasibility of a scalable/generalizable morphing approach for head/brain FEM in the form of a summary report and/or peer-reviewed scientific publications.
• Any data sets emerging from the aforementioned feasibility studies must be made available to the sponsor.

Applicant qualifications and requirements for selection

• Proposals must be led by a senior investigator with a PhD in mechanical engineering or a relevant field, specialized in biomechanics/injury biomechanics and human body modelling.
• Extensive open-literature publication record of original research in the development of human body model for injury prediction, including at least one publication on the prediction of brain injuries and one publication on blast injuries.
• Previous experience and extensive knowledge of the development of human body models, specifically head models using finite element methods, including the development of constitutive models for biological tissues.
• Previous experience researching the response of biological tissues to high rates of loading, including blast injuries to the brain.
• Previous experience working with the modelling of blast waves and other military-relevant high-rate loading conditions.
• Previous experience handling medical imagery anatomical data.
• Access to state-of-the-art computational FEM code/software/solver and computing resources.

Application deadline

Please download and submit the Research Funding Application form.

Enquiries

Questions about this funding opportunity can be sent to the VAC Research office.