Related Experiment Video
Updated: Jul 28, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
A multi-modal and multi-stage region of interest-based fusion network convolutional neural network model to
Zhenpeng Chen1, Beier Qi1,2, Bin Jing1
1School of Biomedical Engineering, Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Capital Medical university, Beijing, Beijing, China.
Abstract:
BackgroundAccurately differentiating stable mild cognitive impairment (sMCI) from progressive MCI (pMCI) is clinically relevant, and identification of pMCI is crucial for timely treatment before it evolves into Alzheimer's disease (AD).ObjectiveTo construct a convolutional neural network (CNN) model to differentiate pMCI from sMCI integrating features from structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) images.MethodsWe proposed a multi-modal and multi-stage region of interest (ROI)-based fusion network (m2ROI-FN) CNN model to differentiate pMCI from sMCI, adopting a multi-stage fusion strategy to integrate deep semantic features and multiple morphological metrics derived from ROIs of sMRI and PET images. Specifically, ten AD-related ROIs of each modality images were selected as patches inputting into 3D hierarchical CNNs. The deep semantic features extracted by the CNNs were fused through the multi-modal integration module and further combined with the multiple morphological metrics extracted by FreeSurfer. Finally, the multilayer perceptron classifier was utilized for subject-level MCI recognition.ResultsThe proposed model achieved accuracy of 77.4% to differentiate pMCI from sMCI with 5-fold cross validation on the entire ADNI database. Further, ADNI-1&2 were formed into an independent sample for model training and validation, and ADNI-3&GO were formed into another independent sample for multi-center testing. The model achieved 73.2% accuracy in distinguishing pMCI and sMCI on ADNI-1&2 and 75% accuracy on ADNI-3&GO.ConclusionsAn effective m2ROI-FN model to distinguish pMCI from sMCI was proposed, which was capable of capturing distinctive features in ROIs of sMRI and PET images. The experimental results demonstrated that the model has the potential to differentiate pMCI from sMCI.
Insights
A new deep learning model effectively distinguishes progressive mild cognitive impairment (pMCI) from stable MCI (sMCI) using brain imaging. This advancement aids early Alzheimer's disease (AD) detection and treatment.
Area of Science:
- Neuroimaging
- Artificial Intelligence in Medicine
- Cognitive Neurology
Background:
- Distinguishing stable mild cognitive impairment (sMCI) from progressive mild cognitive impairment (pMCI) is critical for timely intervention before progression to Alzheimer's disease (AD).
- Early identification of pMCI is essential for initiating treatments that may slow disease progression.
Purpose of the Study:
- To develop and validate a convolutional neural network (CNN) model for differentiating pMCI from sMCI.
- To integrate multi-modal imaging features from structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) for improved diagnostic accuracy.
Main Methods:
- Proposed a multi-modal and multi-stage region of interest (ROI)-based fusion network (m2ROI-FN) CNN.
- Integrated deep semantic features from 3D hierarchical CNNs with morphological metrics from FreeSurfer.
- Utilized ten AD-related ROIs from sMRI and PET images as input, with a multilayer perceptron classifier for final recognition.
Main Results:
- The m2ROI-FN model achieved 77.4% accuracy in differentiating pMCI from sMCI via 5-fold cross-validation on the ADNI database.
- Independent testing on ADNI-1&2 yielded 73.2% accuracy, and on ADNI-3&GO yielded 75% accuracy, demonstrating multi-center generalizability.
Conclusions:
- The proposed m2ROI-FN model effectively distinguishes pMCI from sMCI by capturing distinctive features from sMRI and PET ROIs.
- This deep learning approach shows significant potential for clinical application in early MCI diagnosis and management.
