Early Alzheimer’s disease diagnosis with the contrastive loss using paired structural MRIs

•We proposed a contrastive loss for Alzheimer’s disease diagnosis using sMRI with group categories comparative (G-CAT) information and subject MMSE ranking (S-MMSE) information.•The group based contrastive loss layer is applied to help the model learn the category information of the paired images so that the sMRIs with the same category can learn the closer feature representation.•S-MMSE ranking information was also added to construct the ranking layer between sMRIs with varying MMSE and assist the model to learn the subtle changes between individuals.•G-CAT and S-MMSE both reach remarkable performance in terms of classification sensitivity and specificity respectively. Alzheimer’s Disease (AD) is a chronic and fatal neurodegenerative disease with progressive impairment of memory. Brain structural magnetic resonance imaging (sMRI) has been widely applied as important biomarkers of AD. Various machine learning approaches, especially deep learning-based models, have been proposed for the early diagnosis of AD and monitoring the disease progression on sMRI data. However, the requirement for a large number of training images still hinders the extensive usage of AD diagnosis. In addition, due to the similarities in human whole-brain structure, finding the subtle brain changes is essential to extract discriminative features from limited sMRI data effectively. In this work, we proposed two types of contrastive losses with paired sMRIs to promote the diagnostic performance using group categories (G-CAT) and varying subject mini-mental state examination (S-MMSE) information, respectively. Specifically, G-CAT contrastive loss layer was used to learn the closer feature representation from sMRIs with the same categories, while ranking information from S-MMSE assists the model to explore subtle changes between individuals. The model was trained on ADNI-1. Comparison with baseline methods was performed on MIRIAD and ADNI-2. For the classification task on MIRIAD, S-MMSE achieves 93.5% of accuracy, 96.6% of sensitivity, and 94.9% of specificity, respectively. G-CAT and S-MMSE both reach remarkable performance in terms of classification sensitivity and specificity respectively. Comparing with state-of-the-art methods, we found this proposed method could achieve comparable results with other approaches. The proposed model could extract discriminative features under whole-brain similarity. Extensive experiments also support the accuracy of this model, i.e., it provides bette