Functional MRI And Radiomic Analysis Of Alzheimer’s Disease And Amnestic Mild Cognitive Impairment

Objective:(1)Our study aimed to identify the hippocampus radiomics biomarkers for building classification models in amnestic mild cognitive impairment(a MCI)diagnosis.(2)Moreover,to investigate the correlation between hippocampal functional connectivity(FC)and MRI radiomic features in Alzheimer’s disease(AD).Subjects and Methods:(1)We recruited 42 subjects with aMCI and 44 normal controls(NC).The right and left hippocampus were segmented respectively for each subject using an efficient learning-based method.Then,the radiomic analysis was applied to calculate and select the radiomic features.Finally,two logistic regression models were built based on the selected features obtained from the right and left hippocampus correspondingly.(2)In addition,a total of 67 AD patients and 44 NC were enrolled in this study.Using resting-state functional MRI(rs-f MRI),the whole-brain FC with bilateral hippocampus as seed was performed,and the FC values were extracted from the bilateral hippocampus respectively.Bilateral hippocampal radiomics features were calculated and selected respectively using A.K software.Pearson correlation analysis was implemented between the bilateral hippocampal radiomics features and hippocampus FC values.Results:(1)There were 385 features derived after calculation and 4 features remained after feature selection from each group of data.The area under the receiver operating characteristic curve,specificity,sensitivity,positive predictive value,negative predictive value,precision,recall,and F-Score of the classification evaluation index of the right hippocampus logistic regression model were 0.76,0.71,0.69,0.69,0.71,0.69,0.69,and 0.69,and those of the left hippocampus model were 0.79,0.71,0.54,0.64,0.63,0.64,0.54,and 0.58,respectively.(2)We observed that AD patients demonstrated disruptive FC in some brain regions in the left hippocampal functional network,including right gyrus rectus,right anterior cingulate and paracingulate gyri,bilateral precuneus,bilateral angular gyrus,and bilateral middle occipital gyrus.In addition,decreased FC was witnessed in some brain regions in the right hippocampal functional network,including bilateral anterior cingulate and paracingulate gyri,right dorsolateral superior frontal gyrus,and right precentral gyrus.Pearson correlation analysis suggested that two gray level runlength matrix(RLM)radiomic features and one gray level co-occurrence matrix(GLCM)radiomic features weakly associated with FC values in the left hippocampus.However,there were no significant correlations between radiomic features and FC values in the right hippocampus.Conclusion: Results demonstrate the hippocampal radiomic biomarkers are potential valid for the a MCI diagnosis.The MRI-based radiomic analysis,with further improvement and validation,can be used to identify patients with a MCI and guide the individual treatment.In addition,this is a prospective study revealed the correlation between the MRI radiomic features and the rs-f MRI in AD patients.There was a weak correlation between several radiomic features and functional connectivity in left hippocampus.