Related Experiment Video
Updated: Mar 23, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Medical Image Retrieval Using Multi-graph Learning for MCI Diagnostic Assistance
Yue Gao1, Ehsan Adeli-M1, Minjeong Kim1
1Department of Radiology and BRIC, University of North Carolina at Chapel Hill, NC 27599, USA.
Abstract:
Alzheimer's disease (AD) is an irreversible neurodegenerative disorder that can lead to progressive memory loss and cognition impairment. Therefore, diagnosing AD during the risk stage, a.k.a. Mild Cognitive Impairment (MCI), has attracted ever increasing interest. Besides the automated diagnosis of MCI, it is important to provide physicians with related MCI cases with visually similar imaging data for case-based reasoning or evidence-based medicine in clinical practices. To this end, we propose a multi-graph learning based medical image retrieval technique for MCI diagnostic assistance. Our method is comprised of two stages, the query category prediction and ranking. In the first stage, the query is formulated into a multi-graph structure with a set of selected subjects in the database to learn the relevance between the query subject and the existing subject categories through learning the multi-graph combination weights. This predicts the category that the query belongs to, based on which a set of subjects in the database are selected as candidate retrieval results. In the second stage, the relationship between these candidates and the query is further learned with a new multi-graph, which is used to rank the candidates. The returned subjects can be demonstrated to physicians as reference cases for MCI diagnosing. We evaluated the proposed method on a cohort of 60 consecutive MCI subjects and 350 normal controls with MRI data under three imaging parameters: T1 weighted imaging (T1), Diffusion Tensor Imaging (DTI) and Arterial Spin Labeling (ASL). The proposed method can achieve average 3.45 relevant samples in top 5 returned results, which significantly outperforms the baseline methods compared.
Insights
This study introduces a novel multi-graph learning method for retrieving similar medical images to aid in diagnosing Mild Cognitive Impairment (MCI), a risk stage for Alzheimer's disease (AD). The technique effectively assists physicians by providing relevant case examples for better diagnostic decisions.
Area of Science:
- Neuroimaging
- Artificial Intelligence in Medicine
- Medical Informatics
Background:
- Alzheimer's disease (AD) diagnosis is critical at the Mild Cognitive Impairment (MCI) stage.
- Physicians require similar case examples for evidence-based diagnosis and clinical reasoning.
- Existing automated diagnosis methods lack robust case retrieval capabilities.
Purpose of the Study:
- To develop a multi-graph learning based medical image retrieval technique for MCI diagnostic assistance.
- To enable physicians to access visually similar MCI cases for clinical decision support.
- To improve the accuracy and efficiency of MCI case retrieval.
Main Methods:
- A two-stage approach involving query category prediction and ranking using multi-graph learning.
- Formulating query and database subjects into multi-graph structures to learn relevance.
- Utilizing MRI data (T1, DTI, ASL) from MCI subjects and normal controls for evaluation.
Main Results:
- The proposed method achieved an average of 3.45 relevant samples within the top 5 retrieved results.
- Demonstrated significant outperformance compared to baseline retrieval methods.
- Successfully provided physicians with relevant reference cases for MCI diagnosis.
Conclusions:
- The multi-graph learning approach offers a powerful tool for medical image retrieval in MCI diagnosis.
- This technique enhances diagnostic assistance by providing clinically relevant case examples.
- The method shows promise for improving diagnostic accuracy and supporting evidence-based medicine in neurology.
