Related Experiment Video
Updated: Jul 14, 2025
Hybrid PET/MRI Imaging of Alzheimer's Disease Based on 18F-AV-1451
Published on: April 18, 2025
MRI-based Deep Learning Assessment of Amyloid, Tau, and Neurodegeneration Biomarker Status across the Alzheimer
Christopher O Lew1, Longfei Zhou1, Maciej A Mazurowski1
1From the Department of Radiology, Division of Neuroradiology, Alzheimer Disease Imaging Research Laboratory (C.O.L., J.R.P.), and Neurocognitive Disorders Program, Departments of Psychiatry and Medicine (P.M.D.), Duke University Medical Center, DUMC-Box 3808, Durham, NC 27710-3808; and Duke Institute for Brain Sciences (P.M.D.) and Department of Electrical and Computer Engineering, Department of Computer Science, Department of Biostatistics and Bioinformatics (L.Z., M.A.M.), Duke University, Durham, NC.
Abstract:
Background PET can be used for amyloid-tau-neurodegeneration (ATN) classification in Alzheimer disease, but incurs considerable cost and exposure to ionizing radiation. MRI currently has limited use in characterizing ATN status. Deep learning techniques can detect complex patterns in MRI data and have potential for noninvasive characterization of ATN status. Purpose To use deep learning to predict PET-determined ATN biomarker status using MRI and readily available diagnostic data. Materials and Methods MRI and PET data were retrospectively collected from the Alzheimer's Disease Imaging Initiative. PET scans were paired with MRI scans acquired within 30 days, from August 2005 to September 2020. Pairs were randomly split into subsets as follows: 70% for training, 10% for validation, and 20% for final testing. A bimodal Gaussian mixture model was used to threshold PET scans into positive and negative labels. MRI data were fed into a convolutional neural network to generate imaging features. These features were combined in a logistic regression model with patient demographics, APOE gene status, cognitive scores, hippocampal volumes, and clinical diagnoses to classify each ATN biomarker component as positive or negative. Area under the receiver operating characteristic curve (AUC) analysis was used for model evaluation. Feature importance was derived from model coefficients and gradients. Results There were 2099 amyloid (mean patient age, 75 years ± 10 [SD]; 1110 male), 557 tau (mean patient age, 75 years ± 7; 280 male), and 2768 FDG PET (mean patient age, 75 years ± 7; 1645 male) and MRI pairs. Model AUCs for the test set were as follows: amyloid, 0.79 (95% CI: 0.74, 0.83); tau, 0.73 (95% CI: 0.58, 0.86); and neurodegeneration, 0.86 (95% CI: 0.83, 0.89). Within the networks, high gradients were present in key temporal, parietal, frontal, and occipital cortical regions. Model coefficients for cognitive scores, hippocampal volumes, and APOE status were highest. Conclusion A deep learning algorithm predicted each component of PET-determined ATN status with acceptable to excellent efficacy using MRI and other available diagnostic data. © RSNA, 2023 Supplemental material is available for this article.
Insights
Deep learning models can predict Alzheimer's disease biomarkers using MRI scans and patient data, offering a non-invasive alternative to PET scans. This approach aids in amyloid, tau, and neurodegeneration classification.
Area of Science:
- Neuroimaging and Artificial Intelligence
- Biomarker Discovery for Alzheimer's Disease
Background:
- Positron Emission Tomography (PET) is crucial for Alzheimer's disease (AD) amyloid-tau-neurodegeneration (ATN) classification but is costly and involves radiation.
- Magnetic Resonance Imaging (MRI) has limited current utility in characterizing ATN status.
- Deep learning (DL) excels at identifying complex patterns in MRI, offering potential for noninvasive ATN assessment.
Purpose of the Study:
- To develop and validate a DL algorithm for predicting PET-determined ATN biomarker status.
- To utilize MRI and routinely available diagnostic data for noninvasive ATN classification.
Main Methods:
- Retrospective analysis of MRI and PET data from the Alzheimer's Disease Imaging Initiative.
- A convolutional neural network (CNN) processed MRI data to extract imaging features.
- Features were integrated with demographics, APOE status, cognitive scores, hippocampal volumes, and clinical diagnoses in a logistic regression model for ATN classification.
Main Results:
- The DL model achieved AUCs of 0.79 for amyloid, 0.73 for tau, and 0.86 for neurodegeneration in the test set.
- Key brain regions identified by the model included temporal, parietal, frontal, and occipital cortices.
- Cognitive scores, hippocampal volumes, and APOE status were the most influential predictors.
Conclusions:
- A DL algorithm effectively predicts individual ATN biomarker components using MRI and standard diagnostic data.
- This approach shows promise for noninvasive and accessible ATN status classification in Alzheimer's disease.
