Related Experiment Video
Updated: Jul 26, 2025
05:17
Hybrid PET/MRI Imaging of Alzheimer's Disease Based on 18F-AV-1451
Published on: April 18, 2025
282
Do radiomics or diffusion-tensor images provide additional information to predict brain amyloid-beta positivity?
Sungyang Jo1, Hyunna Lee2, Hyung-Ji Kim3
1Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea.
Scientific Reports
|June 16, 2023
Summary
T1 MRI volume is the best predictor of amyloid-beta positivity in mild cognitive impairment patients, outperforming radiomics and diffusion-tensor imaging. This finding aids in early disease detection using standard MRI scans.
Area of Science:
- Neurology
- Radiology
- Medical Imaging
Background:
- Alzheimer's disease diagnosis relies on detecting amyloid-beta plaques.
- Early detection of amyloid-beta positivity in mild cognitive impairment (MCI) is crucial for timely intervention.
- Magnetic Resonance Imaging (MRI) offers various techniques for brain assessment.
Purpose of the Study:
- To predict amyloid-beta positivity in MCI patients using conventional T1-weighted MRI, radiomics, and diffusion-tensor imaging.
- To compare the predictive performance of different MRI-derived features for amyloid-beta status.
- To identify the optimal MRI-based predictors for amyloid-beta positivity.
Main Methods:
- A cohort of 186 MCI patients underwent Florbetaben positron emission tomography (PET), MRI (T1-weighted and diffusion-tensor images), and neuropsychological testing.
- A stepwise machine learning algorithm was developed using demographics, T1 MRI features (volume, cortical thickness, radiomics), and diffusion-tensor imaging.
- Performance was evaluated by comparing the area under the curve (AUC) of algorithms utilizing different MRI features.
Main Results:
- The machine learning algorithm using T1 volume achieved a higher AUC (0.73) than using clinical information alone (0.69).
- T1 volume outperformed cortical thickness (AUC 0.68) and radiomic texture features (AUC 0.71) in predicting amyloid-beta positivity.
- Incorporating diffusion-tensor imaging features (fractional anisotropy) with T1 volume did not improve predictive performance (AUC 0.73 vs. 0.73).
Conclusions:
- T1-weighted MRI volume is the most effective MRI-derived feature for predicting amyloid-beta PET positivity in MCI patients.
- Radiomics and diffusion-tensor imaging do not offer additional predictive benefits over T1 volume for this task.
- Conventional T1 MRI volume can serve as a valuable, accessible tool for predicting amyloid-beta status in MCI.
