Related Experiment Video
Updated: Mar 6, 2026
Visualization of Amyloid β Deposits in the Human Brain with Matrix-assisted Laser Desorption/Ionization Imaging Mass Spectrometry
Published on: March 7, 2019
MRI In Vivo Detection of Amyloid-β Protein Deposition in Different Brain Regions of Patients with AD and MCI
Qingning Yang1, Zhongrui Wang1, Tie Deng1
1Department of Medical Imaging, Chongqing Emergency Medical Center, Chongqing University Central Hospital, School of Medicine, Chongqing University, No. 1 Jiankang Road, Yuzhong District, Chongqing, 400014, China.
Abstract:
To investigate a non-invasive magnetic resonance imaging (MRI)-based method for detecting amyloid-β (Aβ) protein deposition in different brain regions of patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD). This study included 80 patients with MCI and 62 patients with AD, who were randomly divided into training and testing sets at an 8:2 ratio. All participants underwent 18 F-florbetapir positron emission tomography (PET) imaging and three-dimensional T1-weighted MRI. The interval between MRI and PET examinations did not exceed 30 days. A deep learning-based three-dimensional VB-Net model was developed for brain region segmentation. All PET images were registered to the corresponding MRI images, and standardized uptake ratios for 109 brain regions were calculated and averaged. Following radiomics feature extraction and selection using multiple methods, six machine learning algorithms were applied to establish regression models. In addition, a lightweight transformer-based deep learning model was constructed by improving the original transformer architecture. A total of 1,409 features were extracted from each brain region in patients with MCI and AD. After feature selection, 46, 16, 47, 59, 17, and 72 features were retained for the construction of stochastic gradient regression (SGR), GBR, random forest regression (RFR), support vector regression (SVR), extreme gradient boosting (XGB), and k-nearest neighbor (KNN) models, respectively. Delong test analysis demonstrated that the RFR model achieved the best performance, with mean absolute error (MAE), mean squared error (MSE), R2 score (RS), and Pearson correlation coefficient (PCC) values of 0.13 ± 0.05, 0.03 ± 0.02, 0.77 ± 0.22, and 0.89 ± 0.05 in the training set, and 0.23 ± 0.10, 0.09 ± 0.08, 0.36 ± 0.12, and 0.65 ± 0.09 in the testing set, respectively. For the deep learning model, the MAE, MSE, RS, and PCC in the testing set were 0.41 ± 0.17, 0.25 ± 0.18, - 0.83 ± 0.42, and - 0.01 ± 0.17, respectively. An artificial intelligence-based approach was successfully developed to quantitatively detect Aβ protein accumulation in different brain regions of patients with AD and MCI using MRI. This method is convenient and non-invasive and does not require cerebrospinal fluid puncture or exposure to ionizing radiation.
Insights
This study developed a non-invasive AI method using MRI to detect amyloid-β (Aβ) protein in Alzheimer's disease (AD) and mild cognitive impairment (MCI) patients. The AI approach accurately quantifies Aβ deposition, offering a convenient diagnostic alternative.
Area of Science:
- Neuroimaging
- Artificial Intelligence
- Biomarker Detection
Background:
- Alzheimer's disease (AD) and mild cognitive impairment (MCI) are characterized by amyloid-β (Aβ) protein deposition.
- Current detection methods for Aβ can be invasive or require ionizing radiation.
Purpose of the Study:
- To develop and validate a non-invasive magnetic resonance imaging (MRI)-based artificial intelligence (AI) method for detecting Aβ protein deposition in the brains of patients with MCI and AD.
- To compare the performance of machine learning and deep learning models in quantifying Aβ protein levels.
Main Methods:
- A cohort of 142 patients (80 MCI, 62 AD) underwent MRI and 18F-florbetapir positron emission tomography (PET) imaging.
- A deep learning-based VB-Net model was used for brain region segmentation, followed by radiomics feature extraction.
- Six machine learning algorithms (SGR, GBR, RFR, SVR, XGB, KNN) and a transformer-based deep learning model were trained to predict Aβ deposition.
Main Results:
- The Random Forest Regression (RFR) model demonstrated superior performance, achieving an R² score of 0.77 ± 0.22 in the training set and 0.36 ± 0.12 in the testing set.
- The RFR model achieved a Pearson correlation coefficient (PCC) of 0.89 ± 0.05 in the training set and 0.65 ± 0.09 in the testing set.
- The developed AI approach successfully quantified Aβ protein accumulation using MRI, offering a non-invasive alternative to existing methods.
Conclusions:
- An AI-based, non-invasive MRI method can effectively detect and quantify amyloid-β protein deposition in patients with MCI and AD.
- This approach eliminates the need for cerebrospinal fluid puncture or exposure to ionizing radiation, providing a more convenient and accessible diagnostic tool.
- Further validation and clinical implementation of this AI-powered neuroimaging technique hold significant promise for early diagnosis and monitoring of Alzheimer's disease progression.
