Neuroimaging statistical approaches for determining neural correlates of Alzheimer's disease via positron emission tomography imaging

Assessing the metabolism of the olfactory circuit by use of 18F-FDG PET-CT imaging in patients suspected of suffering from Alzheimer's disease or frontotemporal dementia

PURPOSE

The loss of olfactory function is known to occur in patients suffering from (behavioral variant) frontotemporal dementia ((bv)FTD) and Alzheimer's disease (AD), although different pathophysiological mechanisms underpin this clinical symptom in both disorders. This study assessed whether brain metabolism of the olfactory circuit as assessed by positron emission tomography (PET) imaging with 2-[fluorine-18]fluoro-2-deoxy-d-glucose ([18F]-FDG) can distinguish these entities in different subsets of patients.

METHODS

Patients presenting with cognitive decline were included from a prospectively kept database: (1) bvFTD patients, (2) AD patients and (3) patients with logopenic primary progressive aphasia (PPA). Metabolic rates were calculated for different regions of the olfactory circuit for each subgroup and compared with a cohort of subjects with normal brain metabolism. Additionally, in patients with a logopenic PPA pattern on PET-imaging, statistical parametric mapping (SPM) analysis was performed.

RESULTS

The metabolism of subdivisions of the olfactory circuit as assessed by [18F]-FDG PET brain imaging to bvFTD and AD from control subjects resulted in sensitivity/specificity rates of 95/87.5% and 80/83.3%, respectively. A sensitivity/specificity rate of 100/87.5% was achieved when used to differentiate AD from bvFTD. In patients with the PPA pattern on imaging, the underlying cause (either FTD or AD) could be determined with a sensitivity/specificity rate of 88/82%. SPM analysis concurred that different regions of the olfactory circuit were affected in patients suffering from AD PPA or bvFTD PPA.

CONCLUSION

Metabolic dysfunction in the olfactory circuit is different in various neurodegenerative disorders. Further investigation of the correlations between the cerebral metabolism and the mechanisms which drive olfactory dysfunction is needed.

A hybrid multimodal machine learning model for Detecting Alzheimer's disease

Alzheimer's disease (AD) diagnosis utilizing single modality neuroimaging data has limitations. Multimodal fusion of complementary biomarkers may improve diagnostic performance. This study proposes a multimodal machine learning framework integrating magnetic resonance imaging (MRI), positron emission tomography (PET) and cerebrospinal fluid (CSF) assays for enhanced AD characterization. The model incorporates a hybrid algorithm combining enhanced Harris Hawks Optimization (HHO) algorithm referred to as ILHHO, with Kernel Extreme Learning Machine (KELM) classifier for simultaneous feature selection and classification. ILHHO enhances HHO's search efficiency by integrating iterative mapping (IM) to improve population diversity and local escaping operator (LEO) to balance exploration-exploitation. Comparative analysis with other improved HHO algorithms, classic meta-heuristic algorithms (MHAs), and state-of-the-art MHAs on IEEE CEC2014 benchmark functions indicates that ILHHO achieves superior optimization performance compared to other comparative algorithms. The synergistic ILHHO-KELM model is evaluated on 202 AD Neuroimaging Initiative (ADNI) subjects. Results demonstrate superior multimodal classification accuracy over single modalities, validating the importance of fusing heterogeneous biomarkers. MRI + PET + CSF achieves 99.2 % accuracy for AD vs. normal control (NC), outperforming conventional and proposed methods. Discriminative feature analysis provides further insights into differential AD-related neurodegeneration patterns detected by MRI and PET. The differential PET and MRI features demonstrate how the two modalities provide complementary biomarkers. The neuroanatomical relevance of selected features supports ILHHO-KELM's potential for extracting sensitive AD imaging signatures. Overall, the study showcases the advantages of capitalizing on complementary multimodal data through advanced feature learning techniques for improving AD diagnosis.