The Knight ADRC has supported many investigators at Washington University and at other institutions over the years. We wish to avoid the situation where two investigators study the same research question to avoid duplication of effort and potential conflict. To determine if your topic has already been studied with our resources, please search our database. If you find that your topic or a related topic has been submitted, you may wish to contact the investigator to inquire about their findings to determine how you might proceed. You may wish to collaborate or modify your request to avoid overlap. The results below reflect requests made since online requests have been accepted. As such, not all fields will have data as certain information, such as aims, were not collected until recently. If an entry has been assigned an ID number (e.g. T1004), the full request has been submitted and is either approved, disapproved or in process. If an entry has no ID number, then it represents a submission that has not yet been reviewed. Search terms are applied across an entire requests application including variables not displayed below. A more specific, detailed search may yield better results depending upon your needs.
Search Terms:
Investigator: Greg Zaharchuk
Project Title: Evaluating MRI-Based Tau Prediction Across the Alzheimer’s Disease Spectrum
Date: October 31, 2026 at 10:50 am
Request ID: D2672
Aim 1: Improve and validate deep learning models for MRI-based tau prediction. Specifically, improve the ability of our existing model to predict regional tau burden using self-supervised pre-training, additional imaging sequences, demographic and plasma biomarker information, and different architectures.
Aim 2: Evaluate similar models to predict both future tau and risk of cognitive decline.
Aim 3:
Aim 4:
Investigator: Yize Zhao
Project Title: MRI-derived Surrogates to Improve Genetic Discovery of Regional Tau-PET in Alzheimer’s Disease
Date: October 30, 2026 at 5:09 pm
Request ID:
Aim 1: Identify common genetic variants associated with regional flortaucipir tau-PET burden, prioritizing temporal meta-regions, entorhinal cortex and inferior temporal cortex.
Aim 2: Develop MRI-derived tau surrogates using measured PET and tabular structural MRI, with genotype-eligible PET participants reserved for genetic analysis.
Aim 3: Evaluate whether MRI surrogates in separately authorized UK Biobank participants improve precision and detection of measured tau-PET associations without replacing observed PET.
Aim 4:
Investigator: Hui, Sai Kam
Project Title: Surface informed modelling of tau propagation and cognitive change in aging
Date: October 29, 2026 at 4:01 am
Request ID:
Aim 1: Our current research is focused on elderly individuals with impaired cognitive functions, aiming to utilize a modeling framework to describe the spatial and temporal propagation characteristics of the pathological processes of Alzheimer’s disease.
Aim 2: Rather than considering amyloid-β and tau solely as baseline predictors, our goal is also to model how pathological burden evolves across brain regions over time.
Aim 3:
Aim 4:
Investigator: Dandan Qiu
Project Title: Spatiotemporal flow-driven joint state-space modeling for multimodal Alzheimer’s disease progression prediction
Date: October 26, 2026 at 10:28 pm
Request ID:
Aim 1: Develop a time-aware state-space model to capture longitudinal AD dynamics from irregularly sampled visits, using time-interval encoding to learn temporal dependencies across variable follow-up schedules.
Aim 2: Extract regional features from T1-weighted structural MRI (sMRI) and use cognitive assessments to identify and enhance clinically relevant brain regions, capturing spatial patterns of AD-related changes.
Aim 3: Integrate sMRI with other modalities, including functional MRI, PET imaging, CSF and blood biomarkers, genetic data, and clinical measures, to learn complementary cross-modal relationships in AD progression.
Aim 4: Develop and validate multimodal survival models, including DeepHit-based approaches, to predict CN-to-MCI and MCI-to-AD conversion risks and timing while accounting for right censoring, and assess external generalizability using BIOCARD.
Investigator: Wei Cheng
Project Title: Cerebrospinal fluid proteomics maps regional brain atrophy and clinical progression
Date: October 26, 2026 at 7:36 am
Request ID:
Aim 1: To identify and independently validate associations between CSF protein levels and regional brain structural measures, including gray matter volume, cortical thickness, and surface area, across the Alzheimer’s disease continuum.
Aim 2: To determine whether baseline CSF protein profiles are associated with subsequent regional brain atrophy and longitudinal changes in brain structural measures, identifying molecular signatures of neurodegenerative progression.
Aim 3: To evaluate whether CSF proteins associated with brain structural alterations predict longitudinal cognitive decline and clinical progression, including conversion from mild cognitive impairment to Alzheimer’s disease dementia.
Aim 4:
Investigator: peng
Project Title: 11
Date: October 25, 2026 at 7:35 am
Request ID:
Aim 1: 12
Aim 2:
Aim 3:
Aim 4:
Investigator: Dongdong Chen
Project Title: Multimodal Deep Learning for Alzheimer’s Disease Progression and Brain Network Alterations in the KARI Cohort
Date: October 19, 2026 at 2:45 am
Request ID:
Aim 1: To develop a multimodal deep learning framework integrating T1w, dMRI, fMRI, and PET data to predict cognitive decline (PACC) and clinical progression (CDR) in the KARI cohort.
Aim 2: To investigate the impact of cerebrovascular comorbidities (e.g., WMH volume, cerebral microbleeds) on AD pathology (Amyloid/Tau) and brain structural/functional connectivity.
Aim 3: To utilize graph neural networks (GNN) on fMRI and dMRI to extract brain network topological features and identify hub regions associated with early cognitive impairment.
Aim 4:
Investigator: Dongdong Chen
Project Title: Multimodal Deep Learning for Alzheimer’s Disease Progression and Brain Network Alterations in the KARI Cohort
Date: October 19, 2026 at 2:44 am
Request ID:
Aim 1: To develop a multimodal deep learning framework integrating T1w, dMRI, fMRI, and PET data to predict cognitive decline (PACC) and clinical progression (CDR) in the KARI cohort.
Aim 2: To investigate the impact of cerebrovascular comorbidities (e.g., WMH volume, cerebral microbleeds) on AD pathology (Amyloid/Tau) and brain structural/functional connectivity.
Aim 3: To utilize graph neural networks (GNN) on fMRI and dMRI to extract brain network topological features and identify hub regions associated with early cognitive impairment.
Aim 4:
Investigator: Azhar Hadi
Project Title: Multimodal Deep Learning for Disease Classification
Date: October 4, 2026 at 10:49 am
Request ID: D2670
Aim 1: Develop and evaluate a multimodal deep learning model that combines structural MRI with clinical and cognitive measures to classify cognitively normal, mild cognitive impairment, and Alzheimer’s disease stages, using patient-level data splits to prevent leakage.
Aim 2: Test whether the model keeps its performance when amyloid or tau PET data are missing, using learned missing-modality handling (null tokens and modality dropout), since PET is available only for a subset of participants.
Aim 3: Predict amyloid and tau PET positivity in cognitively normal and mildly impaired participants from MRI and clinical data, to assess whether low-cost measures can support pre-screening for preclinical Alzheimer’s disease.
Aim 4: Evaluate cross-cohort generalization by testing models trained on external cohorts (ADNI) on the Knight ADRC data, after reconciling participant identifiers to avoid overlap with other publicly released Knight ADRC datasets.
Investigator: Michael Belloy
Project Title: Integrating Sex- and APOE ε4-Stratified Genetics and Multi-Omics to Identify Therapeutic Targets for Alzheimer’s Disease
Date: September 30, 2026 at 3:23 pm
Request ID:
Aim 1: Expand sex- and APOE ε4-stratified GWAS to identify AD risk loci and genes.
Aim 2: Characterize sex- and APOE ε4-dependent molecular changes and integrate them with stratified genetic findings.
Aim 3: Integrate complementary genetic and molecular evidence to prioritize subgroup-informed therapeutic targets and repurposable drugs.
Aim 4: