Related Experiment Video
Updated: Aug 14, 2026
12:21
Probing the Brain in Autism Using fMRI and Diffusion Tensor Imaging
Published on: September 12, 2011
25.6K
Using diffusion tensor imaging to detect cortical changes in fronto-temporal dementia subtypes
M Torso1,2, M Bozzali3,4, M Cercignani5
1Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK. neuropsycom@gmail.com.
Scientific Reports
|July 10, 2020
Summary
Diffusion Tensor Imaging (DTI) measures show promise in distinguishing fronto-temporal dementia (FTD) subtypes. These novel DTI features can aid in accurate diagnosis and patient stratification for targeted treatments.
Area of Science:
- Neuroimaging
- Neurodegenerative Diseases
- Machine Learning in Medicine
Background:
- Fronto-temporal dementia (FTD) is a heterogeneous presenile dementia with distinct subtypes requiring accurate diagnosis for effective treatment.
- Current diagnostic methods for FTD subtypes can be challenging due to overlapping clinical presentations.
- Developing advanced imaging techniques is crucial for improving diagnostic accuracy and patient stratification in FTD.
Purpose of the Study:
- To evaluate the diagnostic performance of novel cortical Diffusion Tensor Imaging (DTI) measures in differentiating FTD subtypes.
- To assess the ability of machine learning classifiers utilizing DTI features to distinguish between healthy subjects (HS) and FTD patients, including specific subtypes.
- To explore the potential of DTI in supporting clinical differential diagnosis and patient selection for therapeutic interventions.
Main Methods:
- Inclusion of 96 FTD patients and 84 healthy subjects (HS).
- Utilized a multi-cohort approach (selection, training, and test cohorts) for feature selection and classifier validation.
- Employed machine learning models to analyze novel DTI measures, cortical grey matter fraction, and Mini-Mental State Examination (MMSE) scores.
Main Results:
- A single novel DTI feature achieved 85% accuracy in binary classification (HS vs. FTD).
- Combining DTI features with grey matter fraction and MMSE yielded an 88% accuracy for HS vs. FTD classification.
- The DTI features demonstrated 76% accuracy in distinguishing between HS and FTD subgroups.
Conclusions:
- Novel DTI measures show significant potential to aid in the differential diagnosis of FTD subtypes.
- These imaging biomarkers could enhance patient selection and stratification for clinical trials and targeted drug treatments.
- DTI analysis represents a promising tool for improving diagnostic precision and therapeutic strategies in FTD management.
