Academic Journal
View in EDS

Rapid atlas-based predictions of regional brain strain simulations from wearable head kinematics using diffusion MRI-informed machine learning.

Bibliographic Details
Title: Rapid atlas-based predictions of regional brain strain simulations from wearable head kinematics using diffusion MRI-informed machine learning.
Authors: Mathew J; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand., Saludar CJA; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; University of the Immaculate Conception, Davao City, Philippines., Yeung T; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand., Holdsworth S; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand., Wang A; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand., Fernandez J; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand., Bussey M; School of Physical Education Sport and Exercise Sciences, University of Otago, Dunedin, New Zealand., McGeown JP; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand; Brain Health Research Institute, Auckland University of Technology, Auckland, New Zealand., Tayebi M; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand., Kwon E; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand., Shim V; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand. Electronic address: v.shim@auckland.ac.nz.
Source: Journal of biomechanics [J Biomech] 2026 Sep; Vol. 206, pp. 113497. Date of Electronic Publication: 2026 Jul 29.
Publication Type: Journal Article
Language: English
Journal Info: Publisher: Elsevier Science Country of Publication: United States NLM ID: 0157375 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1873-2380 (Electronic) Linking ISSN: 00219290 NLM ISO Abbreviation: J Biomech Subsets: MEDLINE
Imprint Name(s): Original Publication: New York ; Oxford : Elsevier Science
MeSH Terms: Brain*/diagnostic imaging , Brain Injuries, Traumatic*/diagnostic imaging , Brain Injuries, Traumatic*/physiopathology , Machine Learning* , Diffusion Magnetic Resonance Imaging* , Wearable Electronic Devices*, Football/injuries ; Humans ; Biomechanical Phenomena ; Finite Element Analysis ; Male ; Computer Simulation ; Adolescent ; Acceleration ; Head
Abstract: Sports-related traumatic brain injury (TBI) remains significantly underdiagnosed, with up to 50% of mild TBI cases in sport going undetected. While finite element (FE) simulations can predict brain deformation from head impacts, their computational cost limits clinical applicability for on-field assessments. This study shows a proof-of-principle in using a personalised machine learning framework capable of rapidly predicting regional brain strain simulations over time from wearable sensor kinematics, requiring as few as 150 head acceleration events (HAE) per athlete for training. Subject-specific anisotropic, viscoelastic FE brain models were constructed from T1-weighted and diffusion MRI for ten high school rugby players. Head kinematics were recorded via instrumented mouthguards and used to simulate the FE brain models. The resulting brain strains were parcellated to two commonly used atlases, including the Desikan-Killiany (supplemented with additional structures for whole-brain coverage, totalling 98 regions) and Automated Anatomical Labelling (116 regions) atlases, with the approach readily adaptable to other atlas schemes. Subject-specific multi-output Random Forest regression models were trained on 150 HAEs for every athlete using 70 time-series-extracted kinematic features (using the tsfresh python package) to predict time-dependent parcellated strain profiles. The ML models achieved an R2 from 75% to 87% and an average RMSE from 0.02 to 0.04 between target and predicted regional strains, with low inter-athlete variability. This framework captures the location, magnitude, and temporal behaviour of regional brain strain simulations, enabling rapid estimation directly from wearable devices and potentially enabling personalised, real-time concussion monitoring in sport.
(Copyright © 2026 The Authors. Published by Elsevier Ltd.. All rights reserved.)
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Contributed Indexing: Keywords: Feature engineering; Finite element modelling; Machine learning; Mild traumatic brain injury; Wearable sensors
Entry Date(s): Date Created: 20260801 Date Completed: 20260912 Latest Revision: 20260912
Update Code: 20260913
DOI: 10.1016/j.jbiomech.2026.113497
PMID: 42541993
Database: MEDLINE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://www.doi.org/10.1016/j.jbiomech.2026.113497?
    Name: ScienceDirect (all content) (s7799221)
    Category: fullText
    Text: View record from ScienceDirect
    MouseOverText: View record from ScienceDirect
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:cmedm&genre=article&issn=18732380&ISBN=&volume=206&issue=&date=20260901&spage=113497&pages=113497&title=Journal of biomechanics&atitle=Rapid%20atlas-based%20predictions%20of%20regional%20brain%20strain%20simulations%20from%20wearable%20head%20kinematics%20using%20diffusion%20MRI-informed%20machine%20learning.&aulast=Mathew%20J&id=DOI:10.1016/j.jbiomech.2026.113497
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: cmedm
DbLabel: MEDLINE
An: 42541993
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Rapid atlas-based predictions of regional brain strain simulations from wearable head kinematics using diffusion MRI-informed machine learning.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Mathew+J%22">Mathew J</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand.<br /><searchLink fieldCode="AU" term="%22Saludar+CJA%22">Saludar CJA</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; University of the Immaculate Conception, Davao City, Philippines.<br /><searchLink fieldCode="AU" term="%22Yeung+T%22">Yeung T</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand.<br /><searchLink fieldCode="AU" term="%22Holdsworth+S%22">Holdsworth S</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand.<br /><searchLink fieldCode="AU" term="%22Wang+A%22">Wang A</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand.<br /><searchLink fieldCode="AU" term="%22Fernandez+J%22">Fernandez J</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand.<br /><searchLink fieldCode="AU" term="%22Bussey+M%22">Bussey M</searchLink>; School of Physical Education Sport and Exercise Sciences, University of Otago, Dunedin, New Zealand.<br /><searchLink fieldCode="AU" term="%22McGeown+JP%22">McGeown JP</searchLink>; Mātai Medical Research Institute, Gisborne, New Zealand; Faculty of Medical and Health Sciences & Centre for Brain Research, University of Auckland, Auckland, New Zealand; Brain Health Research Institute, Auckland University of Technology, Auckland, New Zealand.<br /><searchLink fieldCode="AU" term="%22Tayebi+M%22">Tayebi M</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand.<br /><searchLink fieldCode="AU" term="%22Kwon+E%22">Kwon E</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand.<br /><searchLink fieldCode="AU" term="%22Shim+V%22">Shim V</searchLink>; Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand; Mātai Medical Research Institute, Gisborne, New Zealand. Electronic address: v.shim@auckland.ac.nz.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%220157375%22">Journal of biomechanics</searchLink> [J Biomech] 2026 Sep; Vol. 206, pp. 113497. <i>Date of Electronic Publication: </i>2026 Jul 29.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Elsevier+Science%22">Elsevier Science </searchLink><i>Country of Publication: </i>United States <i>NLM ID: </i>0157375 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1873-2380 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2200219290%22">00219290 </searchLink><i>NLM ISO Abbreviation: </i>J Biomech <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: New York ; Oxford : Elsevier Science
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Brain%22">Brain*</searchLink>/<searchLink fieldCode="MM" term="%22Brain+diagnostic+imaging%22">diagnostic imaging</searchLink> <br /><searchLink fieldCode="MM" term="%22Brain+Injuries%2C+Traumatic%22">Brain Injuries, Traumatic*</searchLink>/<searchLink fieldCode="MM" term="%22Brain+Injuries%2C+Traumatic+diagnostic+imaging%22">diagnostic imaging</searchLink> <br /><searchLink fieldCode="MM" term="%22Brain+Injuries%2C+Traumatic%22">Brain Injuries, Traumatic*</searchLink>/<searchLink fieldCode="MM" term="%22Brain+Injuries%2C+Traumatic+physiopathology%22">physiopathology</searchLink> <br /><searchLink fieldCode="MM" term="%22Machine+Learning%22">Machine Learning*</searchLink> <br /><searchLink fieldCode="MM" term="%22Diffusion+Magnetic+Resonance+Imaging%22">Diffusion Magnetic Resonance Imaging*</searchLink> <br /><searchLink fieldCode="MM" term="%22Wearable+Electronic+Devices%22">Wearable Electronic Devices*</searchLink><br /><searchLink fieldCode="MH" term="%22Football%22">Football</searchLink>/<searchLink fieldCode="MH" term="%22Football+injuries%22">injuries</searchLink> ; <searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Biomechanical+Phenomena%22">Biomechanical Phenomena</searchLink> ; <searchLink fieldCode="MH" term="%22Finite+Element+Analysis%22">Finite Element Analysis</searchLink> ; <searchLink fieldCode="MH" term="%22Male%22">Male</searchLink> ; <searchLink fieldCode="MH" term="%22Computer+Simulation%22">Computer Simulation</searchLink> ; <searchLink fieldCode="MH" term="%22Adolescent%22">Adolescent</searchLink> ; <searchLink fieldCode="MH" term="%22Acceleration%22">Acceleration</searchLink> ; <searchLink fieldCode="MH" term="%22Head%22">Head</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Sports-related traumatic brain injury (TBI) remains significantly underdiagnosed, with up to 50% of mild TBI cases in sport going undetected. While finite element (FE) simulations can predict brain deformation from head impacts, their computational cost limits clinical applicability for on-field assessments. This study shows a proof-of-principle in using a personalised machine learning framework capable of rapidly predicting regional brain strain simulations over time from wearable sensor kinematics, requiring as few as 150 head acceleration events (HAE) per athlete for training. Subject-specific anisotropic, viscoelastic FE brain models were constructed from T1-weighted and diffusion MRI for ten high school rugby players. Head kinematics were recorded via instrumented mouthguards and used to simulate the FE brain models. The resulting brain strains were parcellated to two commonly used atlases, including the Desikan-Killiany (supplemented with additional structures for whole-brain coverage, totalling 98 regions) and Automated Anatomical Labelling (116 regions) atlases, with the approach readily adaptable to other atlas schemes. Subject-specific multi-output Random Forest regression models were trained on 150 HAEs for every athlete using 70 time-series-extracted kinematic features (using the tsfresh python package) to predict time-dependent parcellated strain profiles. The ML models achieved an R<superscript>2</superscript> from 75% to 87% and an average RMSE from 0.02 to 0.04 between target and predicted regional strains, with low inter-athlete variability. This framework captures the location, magnitude, and temporal behaviour of regional brain strain simulations, enabling rapid estimation directly from wearable devices and potentially enabling personalised, real-time concussion monitoring in sport.<br /> (Copyright © 2026 The Authors. Published by Elsevier Ltd.. All rights reserved.)
– Name: Abstract
  Label: Competing Interests
  Group: Ab
  Data: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
– Name: SubjectMinor
  Label: Contributed Indexing
  Group:
  Data: <i>Keywords: </i>Feature engineering; Finite element modelling; Machine learning; Mild traumatic brain injury; Wearable sensors
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20260801 <i>Date Completed: </i>20260912 <i>Latest Revision: </i>20260912
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20260913
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1016/j.jbiomech.2026.113497
– Name: AN
  Label: PMID
  Group: ID
  Data: 42541993
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42541993
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.jbiomech.2026.113497
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        StartPage: 113497
    Subjects:
      – SubjectFull: Football injuries
        Type: general
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Biomechanical Phenomena
        Type: general
      – SubjectFull: Finite Element Analysis
        Type: general
      – SubjectFull: Male
        Type: general
      – SubjectFull: Computer Simulation
        Type: general
      – SubjectFull: Adolescent
        Type: general
      – SubjectFull: Acceleration
        Type: general
      – SubjectFull: Head
        Type: general
      – SubjectFull: Brain diagnostic imaging
        Type: general
      – SubjectFull: Brain Injuries, Traumatic diagnostic imaging
        Type: general
      – SubjectFull: Brain Injuries, Traumatic physiopathology
        Type: general
      – SubjectFull: Machine Learning
        Type: general
      – SubjectFull: Diffusion Magnetic Resonance Imaging
        Type: general
      – SubjectFull: Wearable Electronic Devices
        Type: general
    Titles:
      – TitleFull: Rapid atlas-based predictions of regional brain strain simulations from wearable head kinematics using diffusion MRI-informed machine learning.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Mathew J
      – PersonEntity:
          Name:
            NameFull: Saludar CJA
      – PersonEntity:
          Name:
            NameFull: Yeung T
      – PersonEntity:
          Name:
            NameFull: Holdsworth S
      – PersonEntity:
          Name:
            NameFull: Wang A
      – PersonEntity:
          Name:
            NameFull: Fernandez J
      – PersonEntity:
          Name:
            NameFull: Bussey M
      – PersonEntity:
          Name:
            NameFull: McGeown JP
      – PersonEntity:
          Name:
            NameFull: Tayebi M
      – PersonEntity:
          Name:
            NameFull: Kwon E
      – PersonEntity:
          Name:
            NameFull: Shim V
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 09
              Text: 2026 Sep
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1873-2380
          Numbering:
            – Type: volume
              Value: 206
          Titles:
            – TitleFull: Journal of biomechanics
              Type: main
ResultId 1