Biology subjects

Tudorascu, D. L.

Publications and source records attributed to Tudorascu, D. L..

5 recordsLinked to original sources

bioRxiv2026

Evaluation of harmonization methods to mitigate assay and cohort effects in plasma p-tau217

INTRODUCTIONThe growing number of assay platforms measuring blood-based biomarkers (BBMs) for Alzheimers disease (AD) has introduced challenges in interpretability and comparability across assays. Differences across studies also limit comparability of data. To address these challenges, a systematic evaluation of harmonization methods is needed to support BBM data integration within or across studies. METHODSTwo multisite studies, Alzheimers Disease Neuroimaging Initiative (ADNI, n = 219) and Human Connectome Project (HCP, n = 111), were used to evaluate harmonization methods for mitigating assay and cohort effects in plasma p-tau217 measurements. Methods includes various normalization, regression, and standardization approaches, including the recently developed CentiMarker. Assay effects were evaluated using repeated-measures data across assay platforms within each cohort, whereas cohort effects were assessed using pooled ADNI and HCP data. Harmonization performance was evaluated using distributional statistics and downstream modeling of p-tau217. RESULTSQuantile normalization and quantile mapping methods were most effective for mitigating assay effects, whereas conditional quantile mapping performed best for pooled multi-cohort data. These methods also preserved biological variability. In contrast, simple means adjustment and reference-based z-score standardization were least effective for mitigating assay effects, while simple means adjustment, z-score standardization, and quantile normalization were least effective for mitigating cohort effects. CentiMarker had minimal impact on assay or cohort effects. DISCUSSIONBased on our evaluation, we recommend (conditional) quantile mapping for p-tau217 studies integrating data across multiple assays or cohorts. In contrast, we caution against using CentiMarker and z-score-based methods, as they limit comparability and do not effectively mitigate technical variability.

neuroscience↗

bioRxiv2026

Superpixel-ComBat multi-site harmonization of unpaired T1W MRI data in Huntingtons disease

BackgroundPooling multi-site MRI data is essential for well-powered neuroimaging analyses, particularly in Huntingtons disease (HD), where large cohorts are needed to study disease-stage heterogeneity and subtle progressive neuroanatomical change. However, scanner-related variability hinders direct data pooling, confounding image-level methods such as voxel-based morphometry (VBM). Superpixel-ComBat (SP-ComBat), a voxel-level image-harmonization framework, effectively removes scanner effects but depends on traveling-subject data that are rarely available retrospectively. We extend SP-ComBat to unpaired multi-site T1-weighted MRI by introducing a pseudo-pairing framework that leverages demographically matched controls across scanners as surrogate traveling subjects. MethodsTwo pipelines were developed to estimate scanner effects under retrospective constraints: pipeline 1 used a small set of well-matched pseudo-pairs (n = 4) with bootstrap resampling to address scanners with limited sample sizes, while pipeline 2 used the recommended number of pseudo-pairs (n = 16) without resampling. Pseudo-pair images were parcellated into 3D-superpixels, and ComBat was applied within clusters to estimate scanner-specific adjustments for native-space harmonization. Pipeline performance was assessed in a representative multi-study dataset comprising six scanners from three HD cohorts (HD-YAS, HD-CSF, TRACK-HD; N = 144) and replicated in the full multi-study dataset (FMD; N = 548). ResultsBoth pipelines improved image quality, aligned scanner-specific intensities, and preserved disease-related structural patterns. Pipeline 2 showed superior parameter re-estimation stability and was selected for the FMD. Harmonization eliminated systematic segmentation errors, enabled a single unified VBM pipeline across scanners, and increased sensitivity to HD-related voxel-wise atrophy. ConclusionsSP-ComBat was effectively adapted for harmonization of unpaired multi-site structural MRI, reducing scanner bias while preserving biological variability and supporting unified VBM analyses across scanners.

neuroscience↗

bioRxiv2025

Periventricular Diffusivity Reflects APOE4-modulated Amyloid Accumulation and Cognitive Impairment in Alzheimers Continuum

BackgroundAltered glymphatic-related fluid dynamics are increasingly recognized as a key feature of Alzheimers disease (AD). We generalized an established diffusion imaging technique to estimate periventricular diffusivity (PVeD), hypothesizing that fast diffusion signals in the periventricular region can reflect amyloid-beta (A{beta}) deposition across the Alzheimers continuum. MethodsParticipants from two multi-site cohorts (n = 440 and 414), comprising cognitively unimpaired individuals, those with mild cognitive impairment, and patients with AD, were included. We tested and validated the association of PVeD with A{beta} burden and core AD characteristics. ResultsLower PVeD was extensively associated with greater A{beta} burden, neurodegeneration, cognitive impairment, and clinical severity. Importantly, the relationship between PVeD and A{beta} burden was significantly modulated by APOE4 status, with APOE4 carriers showing a stronger negative association. Baseline PVeD also predicted longitudinal cognitive decline. DiscussionThese findings suggest that periventricular fast diffusion signals can reflect APOE4-modulated A{beta} burden and cognitive decline in AD. Research-in-ContextO_ST_ABSSystematic reviewC_ST_ABSA comprehensive PubMed literature search indicates that fluid movement related to glymphatic activity assessed by diffusion tensor image analysis along the perivascular space (DTI-ALPS) is associated with amyloid-beta deposition in Alzheimers disease (AD). However, recent evidence underscores certain limitations of DTI-ALPS, suggesting that it may not fully capture the diffusion processes involved in amyloid clearance. Moreover, no previous studies have investigated the role of APOE4 in modulating the relationship between glymphatic-related fast diffusion signals and amyloid-beta deposition. InterpretationThe transverse diffusion process along the perivenous space in the periventricular region appears to reflect glymphatic-related dysfunction manifested by amyloid-beta deposition. Reduced periventricular diffusivity is associated with greater amyloid burden across the AD continuum. This association is notably enhanced in APOE4 carriers, who exhibit higher amyloid accumulation for a given reduction in the periventricular diffusivity. Besides, periventricular diffusivity is related to other pathological markers of AD, including clinical symptom severity and neurodegeneration, and may also predict subsequent cognitive decline. Future directionsAlthough diffusion-based neuroimaging metrics hold promise as surrogate imaging biomarkers for glymphatic-related activity, they do not comprehensively capture the complex fluid dynamics such as convective bulk flow within the glymphatic system. By leveraging multimodal neuroimaging techniques and advanced analytic approaches, future research can refine these metrics into more sensitive, non-invasive tools capable of evaluating fluid dynamics related to glymphatic dysfunction in AD.

neuroscience↗

bioRxiv2022

MISPEL: A deep learning approach for harmonizingmulti-scanner matched neuroimaging data

Large-scale data obtained from aggregation of already collected multi-site neuroimaging datasets has brought benefits such as higher statistical power, reliability, and robustness to the studies. Despite these promises from growth in sample size, substantial technical variability stemming from differences in scanner specifications exists in the aggregated data and could inadvertently bias any downstream analyses on it. Such a challenge calls for data normalization and/or harmonization frameworks, in addition to comprehensive criteria to estimate the scanner-related variability and evaluate the harmonization frameworks. In this study, we propose MISPEL (Multi-scanner Image harmonization via Structure Preserving Embedding Learning), a supervised multi-scanner harmonization method that is naturally extendable to more than two scanners. We also designed a set of criteria to investigate the scanner-related technical variability and evaluate the harmonization techniques. As an essential requirement of our criteria, we introduced a multi-scanner matched dataset of 3T T1 images across four scanners, which, to the best of our knowledge is one of the few datasets of this kind. We also investigated our evaluations using two popular segmentation frameworks: FSL and segmentation in statistical parametric mapping (SPM). Lastly, we compared MISPEL to popular methods of normalization and harmonization, namely White Stripe, RAVEL, and CALAMITI. MISPEL outperformed these methods and is promising for many other neuroimaging modalities.

neuroscience↗

bioRxiv2022

ComBat Harmonization: Empirical Bayes versus Fully Bayes Approaches

Studying small effects or subtle neuroanatomical variation requires large-scale sample size data. As a result, combining neuroimaging data from multiple datasets is necessary. Variation in acquisition protocols, magnetic field strength, scanner build, and many other non-biologically related factors can introduce undesirable bias into studies. Hence, harmonization is required to remove the bias-inducing factors from the data. ComBat is one of the most common methods applied to features from structural images. ComBat models the data using a hierarchical Bayesian model and uses the empirical Bayes approach to infer the distribution of the unknown factors. The empirical Bayes harmonization method is computationally efficient and provides valid point estimates. However, it tends to underestimate uncertainty. This paper investigates a new approach, fully Bayesian ComBat, where Monte Carlo sampling is used for statistical inference. When comparing fully Bayesian and empirical Bayesian ComBat, we found Empirical Bayesian ComBat more effectively removed scanner strength information and was much more computationally efficient. Conversely, fully Bayesian ComBat better preserved biological disease and age-related information while performing more accurate harmonization on traveling subjects. The fully Bayesian approach generates a rich posterior distribution, which is useful for generating simulated imaging features for improving classifier performance in a limited data setting. We show the generative capacity of our model for augmenting and improving the detection of patients with Alzheimers disease. Posterior distributions for harmonized imaging measures can also be used for brain-wide uncertainty comparison and more principled downstream statistical analysis. Code for our new fully Bayesian ComBat extension is available at https://github.com/batmanlab/BayesComBat.

neuroscience↗