Predicting Parkinson's disease progression using MRI-based white matter radiomic biomarker and machine learning: a reproducibility and replicability study

Arafe, M.; Bhagwat, N. P.; Chatelain, Y.; Dugre, M.; Sokolowski, A.; Wang, M.; Xiao, Y.; Sharp, M.; Poline, J.-B.; Glatard, T.

doi:10.1101/2023.05.05.539590

Home/Archive/bioRxiv

bioRxiv · 10.1101/2023.05.05.539590

Predicting Parkinson's disease progression using MRI-based white matter radiomic biomarker and machine learning: a reproducibility and replicability study

Arafe, M.·Bhagwat, N. P.·Chatelain, Y.·Dugre, M.·Sokolowski, A.·Wang, M.·Xiao, Y.·Sharp, M.·Poline, J.-B.·Glatard, T.

View original source ↗Export metadata ↓

Abstract

BackgroundThe availability of reliable biomarkers of Parkinsons disease (PD) progression is critical to the understanding of the disease and development of treatment options. Magnetic Resonance Imaging (MRI) provides a promising source of PD biomarkers, however, neuroimaging results have been shown to be markedly sensitive to analytical conditions and population sampling, which motivates investigations of their robustness. This study is part of a project to investigate the replicability of 11 structural MRI measures of PD identified in a recent review. ObjectiveThis paper attempts to reproduce (similar data, similar analysis) and replicate (variations in data and analysis) the design of the machine learning (ML) model described in [1] to predict PD progression from T1-weighted MRIs. MethodsWe used the Parkinsons Progression Markers Initiative dataset (PPMI, ppmi-info.org) used in [1] and we followed as closely as possible the original methods. We also investigated slight methodological variations in cohort selection, feature extraction, ML model design, and evaluation techniques. ResultsThe Area under the ROC Curve (AUC) achieved by our model closely reproducing the original study remained lower than 0.5. Across all tested models, we obtained a peak AUC of 0.685, which is better than chance performance but remained lower than the AUC value of 0.795 reported in [1]. ConclusionWe managed to train a model that predicts disease progression with a performance better than chance on a cohort extracted from the PPMI dataset, using methods adapted from [1]. However, the performance of this model remains substantially lower than the one reported in [1]. Our difficulties to reproduce or replicate the original work are likely explained by the relatively low sample size in the original study. We provide recommendations on how to improve the reproducibility of MRI-based ML models of PD in the future.

Explore related subjects

neuroscience Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Arafe, M., Bhagwat, N. P., Chatelain, Y., Dugre, M., Sokolowski, A., Wang, M., Xiao, Y., Sharp, M., Poline, J.-B., Glatard, T.. 2023-05-05. Predicting Parkinson's disease progression using MRI-based white matter radiomic biomarker and machine learning: a reproducibility and replicability study. https://doi.org/10.1101/2023.05.05.539590

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

The Unreasonable Effectiveness of Cell Types in Describing Neuronal Physiological Features

Single-cell RNA sequencing (scRNA-seq) captures detailed gene expression profiles at scale, while patch-clamp recordings measure intrinsic neuronal electrophysiological properties. Modeling the relations between these two modalities remains a challenge. Here, we compare how well electrophysiological features can be predicted by traditional transcriptomic cell type classification, representations derived from a foundational model (scGPT) pretrained on large-scale scRNA-seq datasets, ion channel-coding genes, and highly variable genes. Using paired transcriptomic and electrophysiological patch-sequencing data from 495 human neurons from neurosurgical tissue, we find that cluster-level cell type representations consistently outperform highly variable gene selection, ion channel gene selection, and context-enriched scGPT embeddings. Notably, performance varies across model architectures and initializations, and the best results are obtained by combining the outputs of separate cell type and scGPT-based models. Together, these findings suggest that traditional discrete cellular classification is highly effective in predicting physiological features. For maximum performance it can be complemented by pretrained transformer models.

neuroscience↗

bioRxiv2026

A nonlinear inhibition pathway underlying cortical responses to tuned holographic optogenetic perturbations

Optogenetics enables causal manipulation of cortical activity. Perturbation responses can be counterintuitive due to network interactions, making theory essential for predicting them. Existing approaches often rely on linear approximations, which fail for many biologically relevant perturbations. Here we develop a nonlinear theory of responses to holographic perturbations in cell-type-specific recurrent networks with structured connectivity. We fit a nonlinear model to mouse V1 data, which shows cotuned-ensemble suppression: perturbing spatially clustered neurons with similar preferred orientations yields markedly stronger short-range suppression than perturbing untuned ensembles. We show that cotuned-ensemble suppression arises from a feature-tuned, nonlinear inhibition pathway implicating somatostatin-positive (SST) interneurons. The theory predicts that cotuned ensembles suppress parvalbumin-positive (PV) neurons but facilitate SST neurons, and links the degree of cotuned-ensemble suppression or facilitation to the variance of the SST response. This framework identifies mechanisms by which nonlinear inhibition sculpts cortical dynamics and establishes a predictive basis for targeted optogenetic interventions.

neuroscience↗

bioRxiv2026

Proteomic signatures of APOE ε4 across human tissues and cell types in Alzheimers disease

The apolipoprotein E {varepsilon}4 (APOE {varepsilon}4) allele is the strongest genetic risk factor for late-onset Alzheimers disease (AD). However, the underlying molecular mechanisms remain unclear. This study included 1691 participants from the Religious Orders Study and Rush Memory and Aging Project (ROSMAP), 1226 participants from the Accelerating Medicines Partnership - Alzheimers Disease (AMP-AD) Diverse Cohorts Study, and 735 participants from the Alzheimers Disease Neuroimaging Initiative (ADNI). To characterise APOE {varepsilon}4 molecular effects, we analysed proteomic data from plasma, cerebrospinal fluid (CSF), and induced pluripotent stem cell (iPSC)-derived astrocytes and neurons, as well as transcriptomic and proteomic data from multiple brain regions. The association of APOE {varepsilon}4 with AD neuropathology was also examined. APOE {varepsilon}4 carriers shared a plasma proteomic signature enriched for immune processes, irrespective of AD diagnosis. A machine learning classifier trained on this signature discriminated APOE {varepsilon}4 carriers from non-carriers in an independent cohort using CSF proteomics. APOE {varepsilon}4 carriage was associated with higher Braak stages and Consortium to Establish a Registry for Alzheimers Disease (CERAD) score. However, only limited APOE {varepsilon}4-associated transcriptomic and proteomic changes were observed in bulk brain tissue, with poor cross-layer concordance. Proteomic analyses of iPSC-derived astrocytes and neurons further revealed cell-type-specific APOE {varepsilon}4-associated changes. APOE {varepsilon}4 is associated with a consistent proteomic signature across plasma and CSF. Its molecular effects in the brain differ across cell types, brain regions and molecular layers. These findings support the need for cell-type-resolved multi-omic studies to elucidate how APOE {varepsilon}4 confers AD risk.

neuroscience↗