Image Analysis for Alzheimer's and Parkinson's
Digital Poster
Analysis Methods
Tuesday, 12 May 2026
This session will present the development and validation of image analysis tools for the analysis of Alzheimer’s disease and Parkinson’s disease.
|
465-04-001.
Signed Network Analysis Reveals Compensatory Responses of Default Mode Network Functional Connectivity to Amyloid Deposition
Impact: This work demonstrates a compensatory response of the brain's default mode network functional connectivity to amyloid deposition. Using signed network metrics may improve early detection of pathological brain reorganization and guide future network-based therapeutic strategies.
|
||
|
465-04-002.
Improving Diagnostic Confidence in Neuroimaging with MR-Guided PET Reconstruction
Impact: MR-guided PET image reconstruction improves image quality and diagnostic confidence in focal or regional anomaly detection in epilepsy and dementia.
|
||
|
465-04-003.
Probing cellular microstructure via time-dependent diffusion MRS and machine learning based modeling in Alzheimer’s disease
Impact: This work provides a unique insight into the Td-dependency
of intracellular metabolites and water to probe the microstructural changes during
the early presymptomatic stages of AD, which helps revealing microstructural
and metabolite profile during AD progression pathogenesis.
|
||
|
465-04-004.
Progressive Glymphatic Impairment Across Alzheimer’s Disease Stages: A Multimodal MRI Study
Impact: This
study provides in vivo evidence of progressive glymphatic dysfunction across
Alzheimer’s disease stages. MRI-derived glymphatic metrics may serve as
potential biomarkers for disease staging and monitoring therapeutic efficacy.
|
||
|
465-04-005.
Characteristic changes of resting-state networks in early Alzheimer's disease patients: a multi-method brain network analysis
Impact: This study provides a more integrated view of network dysfunction in early AD. The identification of specific inter-network disconnection and altered structure-function coupling in the prodromal MCI stage offers evidence for developing neuroimaging biomarkers for early detection and differential diagnosis.
|
||
|
465-04-006.
NMR and x-ray Advanced 2D/3D techniques to investigate 23Na relaxation and distribution concentration changes in AD
Impact: This study establishes sodium (²³Na) MRI metrics as potential biomarkers of neuroinflammation in Alzheimer’s disease, enabling earlier and non-invasive detection of neurodegenerative changes, informing clinical research, and advancing neuroimaging tools for improved understanding and monitoring of AD pathophysiology.
|
||
|
465-04-007.
Integrating AI-Derived Brain Volumetrics with Clinical Measures Enhances Early Alzheimer’s Disease Detection
Impact: AI-enabled brain volumetrics from MRI scans can modestly enhance
diagnostic accuracy for AD beyond traditional clinical measures. Scalable,
automated AI tool could support diagnostic assessment and early intervention in
routine clinical workflows, accelerating personalised approaches to dementia
diagnosis and care
|
||
|
465-04-008.
Predicting brain atrophy in Alzheimer’s disease using 3D conditional rectified flow model
Impact: This study successfully generated high-quality 3D MR images for subject-specific brain atrophy to predict Alzheimer's disease progression. Incorporating meta information improved the performances, especially in AD-spectrum subjects. Quantifying the influence of the meta information further enhanced model interpretability.
|
||
|
465-04-009.
Staging of Alzheimer’s Disease Using Multi-Regional Volumetric Z-Scores and Cognitive Assessments
Impact: Combining brain volumetry with cognitive scores improves the distinction between healthy
aging, mild cognitive impairment, and Alzheimer’s disease. Volumetric measures add biological information that complement cognitive assessments, highlighting the
value of using both for more accurate and individualized diagnosis.
|
||
|
465-04-010.
PDDF-Net: A deep neural network for diagnosing Parkinson's Disease using QSM and T1w images
Impact: This work introduces a new DL paradigm for
diagnosing PD based on QSM and T1w images, which looks into the 3D volume data of
patients in a slice-by-slice manner, achieving both improved explainability and
diagnostic accuracy.
|
||
|
465-04-011.
Structural Connectivity Alterations of Cortical Rich-Club in Parkinson’s Disease with Freezing of Gait
Impact: Our study identifies maladaptive strengthening of structural hub
networks as a mechanism underlying freezing of gait in Parkinson's disease.
Rich-Club topology may serve as an objective biomarker for early FOG risk
identification, informing targeted interventions.
|
||
|
465-04-012.
Edge-Centric Functional Connectivity and ALFF Reveal Dual-Track Network Alterations Across Different Cognitive States in PD
Impact: Integration
of edge-centric functional connectivity and ALFF enables multiscale
characterization of Parkinson’s disease–related cognitive decline, offering
mechanistic insights into hierarchical network reconfiguration underlying
impaired cognitive processing.
|
||
|
465-04-013.
Anatomically-Guided Deep Learning for Early Parkinson's Disease Diagnosis Using Dual-Modal Diffusion MRI
Impact: Our anatomically-guided approach improves early PD diagnostic accuracy by effectively integrating multi-modal neuroimaging features. This fusion strategy utilizing anatomical knowledge provides a valuable reference paradigm for deep learning techniques in the multimodal neuroimaging field.
|