Vincent Koppelmans | arXiv Science

arXiv2026

A Pipeline for ADNI Resting-State Functional MRI Processing and Quality Control

The Alzheimer's Disease Neuroimaging Initiative (ADNI) provides a comprehensive multimodal neuroimaging resource for studying aging and Alzheimer's disease (AD). Since its second wave, ADNI has increasingly collected resting-state functional MRI (rs-fMRI), a valuable resource for discovering brain connectivity changes predictive of cognitive decline and AD. A major barrier to its use is the considerable variability in acquisition protocols and data quality, compounded by missing imaging sessions and inconsistencies in how functional scans temporally align with clinical assessments. As a result, many studies only utilize a small subset of the total rs-fMRI data, limiting statistical power, reproducibility, and the ability to study longitudinal functional brain changes at scale. Here, we describe a pipeline for ADNI rs-fMRI data that encompasses the download of necessary imaging and clinical data, temporally aligning the clinical and imaging data, preprocessing, and quality control. We integrate data curation and preprocessing across all ADNI sites and scanner types using a combination of open-source software (Clinica, fMRIPrep, and MRIQC) and bespoke tools. Quality metrics and reports are generated for each subject and session to facilitate rigorous data screening. All scripts and configuration files are available to enable reproducibility. The pipeline, which currently supports ADNI-GO, ADNI-2, and ADNI-3 data releases, outputs high-quality rs-fMRI time series data adhering to the BIDS-derivatives specification. This protocol provides a transparent and scalable framework for curating and utilizing ADNI fMRI data, empowering large-scale functional biomarker discovery and integrative multimodal analyses in Alzheimer's disease research.

cs.DB↗

arXiv2021

Longitudinal surface-based spatial Bayesian GLM reveals complex trajectories of motor neurodegeneration in ALS

Longitudinal fMRI datasets hold great promise for the study of neurodegenerative diseases, but realizing their potential depends on extracting accurate fMRI-based brain measures in individuals over time. This is especially true for rare, heterogeneous and/or rapidly progressing diseases, which often involve small samples whose functional features may vary dramatically across subjects and over time, making traditional group-difference analyses of limited utility. One such disease is ALS, which results in extreme motor function loss and eventual death. Here, we analyze a rich longitudinal dataset containing 190 motor task fMRI scans from 16 ALS patients and 22 age-matched HCs. We propose a novel longitudinal extension to our cortical surface-based spatial Bayesian GLM, which has high power and precision to detect activations in individuals. Using a series of longitudinal mixed-effects models to subsequently study the relationship between activation and disease progression, we observe an inverted U-shaped trajectory: at relatively mild disability we observe enlarging activations, while at higher disability we observe severely diminished activation, reflecting progression toward complete motor function loss. We observe distinct trajectories depending on clinical progression rate, with faster progressors exhibiting more extreme hyper-activation and subsequent hypo-activation. These differential trajectories suggest that initial hyper-activation is likely attributable to loss of inhibitory neurons. By contrast, earlier studies employing more limited sampling designs and using traditional group-difference analysis approaches were only able to observe the initial hyper-activation, which was assumed to be due to a compensatory process. This study provides a first example of how surface-based spatial Bayesian modeling furthers scientific understanding of neurodegenerative disease.

stat.AP↗