Application of Machine Learning in the Diagnosis and Prognosis of Mild Traumatic Brain Injury Using Diffusion Tensor Imaging: A Systematic Review
Anwendung von Machine Learning in der Diagnose und Prognose von leichter traumatischer Hirnverletzung mithilfe der Diffusions-Tensor-Bildgebung: systematische Übersichtsarbeit
Journal of Magnetic Resonance Imaging · 2026-02 · 63 (2) · 328-345
DOI: 10.1002/jmri.70122 ↗Auf einen Blick
Die Übersichtsarbeit zu 36 Artikeln zeigt: Machine-Learning-Modelle mit DTI-Metriken konnten mTBI-Fälle von Kontrollen und Verlaufsmerkmale unterscheiden. Der klinische Einsatz bleibt laut Autor:innen jedoch entfernt.
Fragestellung
Welche Studien untersuchten Machine Learning mit DTI zur Diagnose und Prognose nach mTBI?
Methode
Systematische Übersichtsarbeit nach PRISMA; Bewertung mit dem Modified QualSyst Assessment Tool.
Ergebnisse
36 Artikel. ML mit DTI-Metriken allein oder kombiniert mit anderen Modalitäten konnte mTBI von Kontrollen klassifizieren und auch Erholung, Symptomschwere, kognitive Scores und Hirnveränderungen vorhersagen.
Einschränkungen
Kleine Stichproben und fehlende externe Validierung; Gefahr von Overfitting.
Schlussfolgerung der Autor:innen
Laut Autor:innen werden größere, extern validierte Studien benötigt, um robuste Modelle für Diagnose und Prognose der mTBI zu entwickeln.
Kernaussagen
- 36 Artikel wurden eingeschlossen
- ML mit DTI klassifizierte mTBI vs. Kontrollen
- Klinische Anwendung derzeit noch fern
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Systematische Übersicht in der Zeitschrift Journal of Magnetic Resonance Imaging, online erschienen am 30.09.2025. Verfasst von Christian John A Saludar und 8 weiteren Autor:innen. Beteiligte Einrichtung: Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand. Hauptthemen laut MEDLINE: Diffusion Tensor Imaging, Machine Learning, Brain Concussion, Brain Injuries, Traumatic. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 95 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Health Research Council of New Zealand, Te Titoki Mataora Research Acceleration Programme.
Abstract (Originalsprache)
Background
Traumatic Brain Injury (TBI) is a global health concern, with mild TBI (mTBI) being the most common form. Despite its prevalence, accurately diagnosing mTBI remains a significant challenge. While advanced neuroimaging techniques like diffusion tensor imaging (DTI) offer promise for more robust diagnosis, their clinical application is limited by inconsistent and heterogeneous post‐injury findings. Recently, machine learning (ML) techniques, utilizing DTI metrics as features, have shown increasing utility in mTBI research. This approach helps identify distinct between‐group features, paving the way for more precise and efficient diagnostic and prognostic tools.
Purpose
This review aims to analyze studies employing ML techniques to assess changes in DTI metrics after mTBI. Study Type Systematic review. Population or Subjects or Phantom or Specimen or Animal Model We conducted a systematic review, adhering to PRISMA guidelines, on the application of ML with DTI for mTBI diagnosis and prognosis on human subjects. This review identified 36 articles. Field Strength/Sequence N/A. Assessment Study quality was assessed using the Modified QualSyst Assessment Tool. Statistical Tests N/A.
Results
The review found ML techniques using DTI Metrics either alone or in combination with other modalities (i.e., structural MRI, functional MRI, clinical scores, or demographics) can effectively classify mTBI patients from controls. These approaches have also demonstrated potential in classifying mTBI patients according to the degree of recovery and symptom severity. In addition, these ML models showed strong predictive power toward cognitive scores and brain structural decline, as quantified by brain‐predicted age difference. Data Conclusion Larger, externally validated studies are needed to develop robust models for the diagnosis and prognosis of mTBI, using imaging biomarkers (including DTI) in conjunction with non‐imaging, on‐field, or clinical data. Despite the high predictive performance of ML algorithms, the clinical application remains distant, likely due to the small sample size of studies and lack of external validation, which raises concerns about overfitting. Evidence Level 5. Technical Efficacy Stage 1.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand.
- University of the Immaculate Conception, Davao City, Philippines.
- Mātai Medical Research Institute, Gisborne, New Zealand.
- Faculty of Medical and Health Sciences and Centre for Brain Research, University of Auckland, Auckland, New Zealand.
- TBI Network, Auckland University of Technology, Auckland, New Zealand.
- Vision Research Foundation, University of Auckland, Auckland, New Zealand.
ORCID-Profile: Christian John A Saludar
Förderung
- Health Research Council of New Zealand
- Te Titoki Mataora Research Acceleration Programme
Zitieren
APA
Saludar, C. J. A., Tayebi, M., Kwon, E., McGeown, J., Schierding, W., Wang, A., Fernandez, J., Holdsworth, S., & Shim, V. (2026). Application of Machine Learning in the Diagnosis and Prognosis of Mild Traumatic Brain Injury Using Diffusion Tensor Imaging: A Systematic Review. Journal of Magnetic Resonance Imaging, 63(2), 328-345. https://doi.org/10.1002/jmri.70122
Vancouver
Saludar CJA, Tayebi M, Kwon E, McGeown J, Schierding W, Wang A, et al. Application of Machine Learning in the Diagnosis and Prognosis of Mild Traumatic Brain Injury Using Diffusion Tensor Imaging: A Systematic Review. Journal of Magnetic Resonance Imaging. 2026;63(2):328-345. doi:10.1002/jmri.70122
BibTeX
@article{Saludar2026,
author = {Saludar, Christian John A and Tayebi, Maryam and Kwon, Eryn and McGeown, Joshua and Schierding, William and Wang, Alan and Fernandez, Justin and Holdsworth, Samantha and Shim, Vickie},
title = {Application of Machine Learning in the Diagnosis and Prognosis of Mild Traumatic Brain Injury Using Diffusion Tensor Imaging: A Systematic Review},
journal = {Journal of Magnetic Resonance Imaging},
year = {2026},
volume = {63},
number = {2},
pages = {328--345},
doi = {10.1002/jmri.70122}
}
EUVIMED