Arrow Research search

Author name cluster

Jean-Philippe Thiran

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

43 papers
2 author rows

Possible papers

43

YNIMG Journal 2026 Journal Article

Spatially regularized super-resolved constrained spherical deconvolution (SR 2 -CSD) of diffusion MRI data

  • Ekin Taskin
  • Gabriel Girard
  • Juan Luis Villarreal Haro
  • Jonathan Rafael-Patiño
  • Eleftherios Garyfallidis
  • Jean-Philippe Thiran
  • Erick Jorge Canales-Rodríguez

Constrained Spherical Deconvolution (CSD) is widely used to estimate the white matter fiber orientation distribution (FOD) from diffusion MRI data. Its angular resolution depends on the maximum spherical harmonic order ( l m a x ): low l m a x yields smooth but poorly resolved FODs, while high l m a x, as in Super-CSD, enables resolving fiber crossings with small inter-fiber angles but increases sensitivity to noise. In this proof-of-concept study, we introduce Spatially Regularized Super-Resolved CSD (SR 2 -CSD), a novel method that regularizes Super-CSD using a spatial FOD prior estimated via a self-calibrated total variation denoiser. We evaluated SR 2 -CSD against CSD and Super-CSD across four datasets: (i) the HARDI-2013 challenge numerical phantom, assessing angular and peak number errors across multiple signal-to-noise ratio (SNR) levels and CSD variants (single-/multi-shell, single-/multi-tissue); (ii) the Sherbrooke in vivo dataset, evaluating spatial coherence of FODs; (iii) a six-subject test–retest dataset acquired with both full (96 gradient directions) and subsampled (45 directions) protocols, assessing reproducibility; and (iv) the DiSCo phantom, evaluating tractography accuracy under varying SNR levels and multiple noise repetitions. Across all evaluations, SR 2 -CSD consistently reduced angular and peak number errors, improved spatial coherence, enhanced test–retest reproducibility, and yielded connectivity matrices more strongly correlated with ground-truth. Most improvements were statistically significant under multiple-comparison correction. These results demonstrate that incorporating spatial priors into CSD is feasible, mitigates estimation instability, and improves FOD reconstruction accuracy.

Details DOI

ICLR Conference 2025 Conference Paper

A Simple Framework for Open-Vocabulary Zero-Shot Segmentation

  • Thomas Stegmüller
  • Tim Lebailly
  • Nikola Dukic
  • Behzad Bozorgtabar
  • Tinne Tuytelaars
  • Jean-Philippe Thiran

Zero-shot classification capabilities naturally arise in models trained within a vision-language contrastive framework. Despite their classification prowess, these models struggle in dense tasks like zero-shot open-vocabulary segmentation. This deficiency is often attributed to the absence of localization cues in captions and the intertwined nature of the learning process, which encompasses both image/text representation learning and cross-modality alignment. To tackle these issues, we propose SimZSS, a $\textbf{Sim}$ple framework for open-vocabulary $\textbf{Z}$ero-$\textbf{S}$hot $\textbf{S}$egmentation. The method is founded on two key principles: i) leveraging frozen vision-only models that exhibit spatial awareness while exclusively aligning the text encoder and ii) exploiting the discrete nature of text and linguistic knowledge to pinpoint local concepts within captions. By capitalizing on the quality of the visual representations, our method requires only image-caption pair datasets and adapts to both small curated and large-scale noisy datasets. When trained on COCO Captions across 8 GPUs, SimZSS achieves state-of-the-art results on 7 out of 8 benchmark datasets in less than 15 minutes. Our code and pretrained models are publicly available at https://github.com/tileb1/simzss.

Details

NeurIPS Conference 2025 Conference Paper

ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains

  • Guillaume Vray
  • Devavrat Tomar
  • Xufeng Gao
  • Jean-Philippe Thiran
  • Evan Shelhamer
  • Behzad Bozorgtabar

This paper introduces ReservoirTTA, a novel plug–in framework designed for prolonged test–time adaptation (TTA) in scenarios where the test domain continuously shifts over time, including cases where domains recur or evolve gradually. At its core, ReservoirTTA maintains a reservoir of domain-specialized models—an adaptive test-time model ensemble—that both detects new domains via online clustering over style features of incoming samples and routes each sample to the appropriate specialized model, and thereby enables domain-specific adaptation. This multi-model strategy overcomes key limitations of single model adaptation, such as catastrophic forgetting, inter-domain interference, and error accumulation, ensuring robust and stable performance on sustained non-stationary test distributions. Our theoretical analysis reveals key components that bound parameter variance and prevent model collapse, while our plug–in TTA module mitigates catastrophic forgetting of previously encountered domains. Extensive experiments on scene-level corruption benchmarks (ImageNet-C, CIFAR-10/100-C), object-level style shifts (DomainNet-126, PACS), and semantic segmentation (Cityscapes→ACDC) — covering recurring and continuously evolving domain shifts — show that ReservoirTTA substantially improves adaptation accuracy and maintains stable performance across prolonged, recurring shifts, outperforming state-of-the-art methods. Our code is publicly available at https: //github. com/LTS5/ReservoirTTA.

PDF Details

ICLR Conference 2025 Conference Paper

Uncertainty modeling for fine-tuned implicit functions

  • Anna Susmelj
  • Mael Macuglia
  • Natasa Tagasovska
  • Reto Sutter
  • Sebastiano Caprara
  • Jean-Philippe Thiran
  • Ender Konukoglu

Implicit functions such as Neural Radiance Fields (NeRFs), occupancy networks, and signed distance functions (SDFs) have become pivotal in computer vision for reconstructing detailed object shapes from sparse views. Achieving optimal performance with these models can be challenging due to the extreme sparsity of inputs and distribution shifts induced by data corruptions. To this end, large, noise-free synthetic datasets can serve as shape priors to help models fill in gaps, but the resulting reconstructions must be approached with caution. Uncertainty estimation is crucial for assessing the quality of these reconstructions, particularly in identifying areas where the model is uncertain about the parts it has inferred from the prior. In this paper, we introduce Dropsembles, a novel method for uncertainty estimation in tuned implicit functions. We demonstrate the efficacy of our approach through a series of experiments, starting with toy examples and progressing to a real-world scenario. Specifically, we train a Convolutional Occupancy Network on synthetic anatomical data and test it on low-resolution MRI segmentations of the lumbar spine. Our results show that Dropsembles achieve the accuracy and calibration levels of deep ensembles but with significantly less computational cost.

Details

ICLR Conference 2025 Conference Paper

What to align in multimodal contrastive learning?

  • Benoit Dufumier
  • Javiera Castillo Navarro
  • Devis Tuia
  • Jean-Philippe Thiran

Humans perceive the world through multisensory integration, blending the information of different modalities to adapt their behavior. Contrastive learning offers an appealing solution for multimodal self-supervised learning. Indeed, by considering each modality as a different view of the same entity, it learns to align features of different modalities in a shared representation space. However, this approach is intrinsically limited as it only learns shared or redundant information between modalities, while multimodal interactions can arise in other ways. In this work, we introduce CoMM, a Contrastive Multimodal learning strategy that enables the communication between modalities in a single multimodal space. Instead of imposing cross- or intra- modality constraints, we propose to align multimodal representations by maximizing the mutual information between augmented versions of these multimodal features. Our theoretical analysis shows that shared, synergistic and unique terms of information naturally emerge from this formulation, allowing us to estimate multimodal interactions beyond redundancy. We test CoMM both in a controlled and in a series of real-world settings: in the former, we demonstrate that CoMM effectively captures redundant, unique and synergistic information between modalities. In the latter, CoMM learns complex multimodal interactions and achieves state-of-the-art results on seven multimodal tasks.

Details

AAAI Conference 2024 Conference Paper

Combining Graph Transformers Based Multi-Label Active Learning and Informative Data Augmentation for Chest Xray Classification

  • Dwarikanath Mahapatra
  • Behzad Bozorgtabar
  • Zongyuan Ge
  • Mauricio Reyes
  • Jean-Philippe Thiran

Informative sample selection in active learning (AL) helps a machine learning system attain optimum performance with minimum labeled samples, thus improving human-in-the-loop computer-aided diagnosis systems with limited labeled data. Data augmentation is highly effective for enlarging datasets with less labeled data. Combining informative sample selection and data augmentation should leverage their respective advantages and improve performance of AL systems. We propose a novel approach to combine informative sample selection and data augmentation for multi-label active learning. Conventional informative sample selection approaches have mostly focused on the single-label case which do not perform optimally in the multi-label setting. We improve upon state-of-the-art multi-label active learning techniques by representing disease labels as graph nodes, use graph attention transformers (GAT) to learn more effective inter-label relationships and identify most informative samples. We generate transformations of these informative samples which are also informative. Experiments on public chest xray datasets show improved results over state-of-the-art multi-label AL techniques in terms of classification performance, learning rates, and robustness. We also perform qualitative analysis to determine the realism of generated images.

PDF Details DOI

ICLR Conference 2024 Conference Paper

CrIBo: Self-Supervised Learning via Cross-Image Object-Level Bootstrapping

  • Tim Lebailly
  • Thomas Stegmüller
  • Behzad Bozorgtabar
  • Jean-Philippe Thiran
  • Tinne Tuytelaars

Leveraging nearest neighbor retrieval for self-supervised representation learning has proven beneficial with object-centric images. However, this approach faces limitations when applied to scene-centric datasets, where multiple objects within an image are only implicitly captured in the global representation. Such global bootstrapping can lead to undesirable entanglement of object representations. Furthermore, even object-centric datasets stand to benefit from a finer-grained bootstrapping approach. In response to these challenges, we introduce a novel $\textbf{Cr}$oss-$\textbf{I}$mage Object-Level $\textbf{Bo}$otstrapping method tailored to enhance dense visual representation learning. By employing object-level nearest neighbor bootstrapping throughout the training, CrIBo emerges as a notably strong and adequate candidate for in-context learning, leveraging nearest neighbor retrieval at test time. CrIBo shows state-of-the-art performance on the latter task while being highly competitive in more standard downstream segmentation tasks. Our code and pretrained models are publicly available at https://github.com/tileb1/CrIBo.

Details

ICLR Conference 2024 Conference Paper

Un-Mixing Test-Time Normalization Statistics: Combatting Label Temporal Correlation

  • Devavrat Tomar
  • Guillaume Vray
  • Jean-Philippe Thiran
  • Behzad Bozorgtabar

Recent test-time adaptation methods heavily rely on nuanced adjustments of batch normalization (BN) parameters. However, one critical assumption often goes overlooked: that of independently and identically distributed (i.i.d.) test batches with respect to unknown labels. This oversight leads to skewed BN statistics and undermines the reliability of the model under non-i.i.d. scenarios. To tackle this challenge, this paper presents a novel method termed '$\textbf{Un-Mix}$ing $\textbf{T}$est-Time $\textbf{N}$ormalization $\textbf{S}$tatistics' (UnMix-TNS). Our method re-calibrates the statistics for each instance within a test batch by $\textit{mixing}$ it with multiple distinct statistics components, thus inherently simulating the i.i.d. scenario. The core of this method hinges on a distinctive online $\textit{unmixing}$ procedure that continuously updates these statistics components by incorporating the most similar instances from new test batches. Remarkably generic in its design, UnMix-TNS seamlessly integrates with a wide range of leading test-time adaptation methods and pre-trained architectures equipped with BN layers. Empirical evaluations corroborate the robustness of UnMix-TNS under varied scenarios—ranging from single to continual and mixed domain shifts, particularly excelling with temporally correlated test data and corrupted non-i.i.d. real-world streams. This adaptability is maintained even with very small batch sizes or single instances. Our results highlight UnMix-TNS's capacity to markedly enhance stability and performance across various benchmarks. Our code is publicly available at https://github.com/devavratTomar/unmixtns.

Details

YNIMG Journal 2023 Journal Article

Tractography passes the test: Results from the diffusion-simulated connectivity (disco) challenge

  • Gabriel Girard
  • Jonathan Rafael-Patiño
  • Raphaël Truffet
  • Dogu Baran Aydogan
  • Nagesh Adluru
  • Veena A. Nair
  • Vivek Prabhakaran
  • Barbara B. Bendlin

Estimating structural connectivity from diffusion-weighted magnetic resonance imaging is a challenging task, partly due to the presence of false-positive connections and the misestimation of connection weights. Building on previous efforts, the MICCAI-CDMRI Diffusion-Simulated Connectivity (DiSCo) challenge was carried out to evaluate state-of-the-art connectivity methods using novel large-scale numerical phantoms. The diffusion signal for the phantoms was obtained from Monte Carlo simulations. The results of the challenge suggest that methods selected by the 14 teams participating in the challenge can provide high correlations between estimated and ground-truth connectivity weights, in complex numerical environments. Additionally, the methods used by the participating teams were able to accurately identify the binary connectivity of the numerical dataset. However, specific false positive and false negative connections were consistently estimated across all methods. Although the challenge dataset doesn't capture the complexity of a real brain, it provided unique data with known macrostructure and microstructure ground-truth properties to facilitate the development of connectivity estimation methods.

Details DOI

YNICL Journal 2022 Journal Article

Cortical lesions, central vein sign, and paramagnetic rim lesions in multiple sclerosis: Emerging machine learning techniques and future avenues

  • Francesco La Rosa
  • Maxence Wynen
  • Omar Al-Louzi
  • Erin S Beck
  • Till Huelnhagen
  • Pietro Maggi
  • Jean-Philippe Thiran
  • Tobias Kober

The current diagnostic criteria for multiple sclerosis (MS) lack specificity, and this may lead to misdiagnosis, which remains an issue in present-day clinical practice. In addition, conventional biomarkers only moderately correlate with MS disease progression. Recently, some MS lesional imaging biomarkers such as cortical lesions (CL), the central vein sign (CVS), and paramagnetic rim lesions (PRL), visible in specialized magnetic resonance imaging (MRI) sequences, have shown higher specificity in differential diagnosis. Moreover, studies have shown that CL and PRL are potential prognostic biomarkers, the former correlating with cognitive impairments and the latter with early disability progression. As machine learning-based methods have achieved extraordinary performance in the assessment of conventional imaging biomarkers, such as white matter lesion segmentation, several automated or semi-automated methods have been proposed as well for CL, PRL, and CVS. In the present review, we first introduce these MS biomarkers and their imaging methods. Subsequently, we describe the corresponding machine learning-based methods that were proposed to tackle these clinical questions, putting them into context with respect to the challenges they are facing, including non-standardized MRI protocols, limited datasets, and moderate inter-rater variability. We conclude by presenting the current limitations that prevent their broader deployment and suggesting future research directions.

Details DOI

YNIMG Journal 2022 Journal Article

Evaluating reproducibility and subject-specificity of microstructure-informed connectivity

  • Philipp J. Koch
  • Gabriel Girard
  • Julia Brügger
  • Andéol G. Cadic-Melchior
  • Elena Beanato
  • Chang-hyun Park
  • Takuya Morishita
  • Maximilian J. Wessel

Tractography enables identifying and evaluating the healthy and diseased brain's white matter pathways from diffusion-weighted magnetic resonance imaging data. As previous evaluation studies have reported significant false-positive estimation biases, recent microstructure-informed tractography algorithms have been introduced to improve the trade-off between specificity and sensitivity. However, a major limitation for characterizing the performance of these techniques is the lack of ground truth brain data. In this study, we compared the performance of two relevant microstructure-informed tractography methods, SIFT2 and COMMIT, by assessing the subject specificity and reproducibility of their derived white matter pathways. Specifically, twenty healthy young subjects were scanned at eight different time points at two different sites. Subject specificity and reproducibility were evaluated using the whole-brain connectomes and a subset of 29 white matter bundles. Our results indicate that although the raw tractograms are more vulnerable to the presence of false-positive connections, they are highly reproducible, suggesting that the estimation bias is subject-specific. This high reproducibility was preserved when microstructure-informed tractography algorithms were used to filter the raw tractograms. Moreover, the resulting track-density images depicted a more uniform coverage of streamlines throughout the white matter, suggesting that these techniques could increase the biological meaning of the estimated fascicles. Notably, we observed an increased subject specificity by employing connectivity pre-processing techniques to reduce the underlaying noise and the data dimensionality (using principal component analysis), highlighting the importance of these tools for future studies. Finally, no strong bias from the scanner site or time between measurements was found. The largest intraindividual variance originated from the sole repetition of data measurements (inter-run).

Details DOI

YNIMG Journal 2022 Journal Article

Insights from the IronTract challenge: Optimal methods for mapping brain pathways from multi-shell diffusion MRI

  • Chiara Maffei
  • Gabriel Girard
  • Kurt G. Schilling
  • Dogu Baran Aydogan
  • Nagesh Adluru
  • Andrey Zhylka
  • Ye Wu
  • Matteo Mancini

Limitations in the accuracy of brain pathways reconstructed by diffusion MRI (dMRI) tractography have received considerable attention. While the technical advances spearheaded by the Human Connectome Project (HCP) led to significant improvements in dMRI data quality, it remains unclear how these data should be analyzed to maximize tractography accuracy. Over a period of two years, we have engaged the dMRI community in the IronTract Challenge, which aims to answer this question by leveraging a unique dataset. Macaque brains that have received both tracer injections and ex vivo dMRI at high spatial and angular resolution allow a comprehensive, quantitative assessment of tractography accuracy on state-of-the-art dMRI acquisition schemes. We find that, when analysis methods are carefully optimized, the HCP scheme can achieve similar accuracy as a more time-consuming, Cartesian-grid scheme. Importantly, we show that simple pre- and post-processing strategies can improve the accuracy and robustness of many tractography methods. Finally, we find that fiber configurations that go beyond crossing (e.g., fanning, branching) are the most challenging for tractography. The IronTract Challenge remains open and we hope that it can serve as a valuable validation tool for both users and developers of dMRI analysis methods.

Details DOI

YNIMG Journal 2021 Journal Article

Resolving bundle-specific intra-axonal T2 values within a voxel using diffusion-relaxation tract-based estimation

  • Muhamed Barakovic
  • Chantal M.W. Tax
  • Umesh Rudrapatna
  • Maxime Chamberland
  • Jonathan Rafael-Patino
  • Cristina Granziera
  • Jean-Philippe Thiran
  • Alessandro Daducci

At the typical spatial resolution of MRI in the human brain, approximately 60–90% of voxels contain multiple fiber populations. Quantifying microstructural properties of distinct fiber populations within a voxel is therefore challenging but necessary. While progress has been made for diffusion and T 1-relaxation properties, how to resolve intra-voxel T 2 heterogeneity remains an open question. Here a novel framework, named COMMIT-T 2, is proposed that uses tractography-based spatial regularization with diffusion-relaxometry data to estimate multiple intra-axonal T2 values within a voxel. Unlike previously-proposed voxel-based T2 estimation methods, which (when applied in white matter) implicitly assume just one fiber bundle in the voxel or the same T2 for all bundles in the voxel, COMMIT-T 2 can recover specific T2 values for each unique fiber population passing through the voxel. In this approach, the number of recovered unique T 2 values is not determined by a number of model parameters set a priori, but rather by the number of tractography-reconstructed streamlines passing through the voxel. Proof-of-concept is provided in silico and in vivo, including a demonstration that distinct tract-specific T2 profiles can be recovered even in the three-way crossing of the corpus callosum, arcuate fasciculus, and corticospinal tract. We demonstrate the favourable performance of COMMIT-T2 compared to that of voxelwise approaches for mapping intra-axonal T2 exploiting diffusion, including a direction-averaged method and AMICO-T2, a new extension to the previously-proposed Accelerated Microstructure Imaging via Convex Optimization (AMICO) framework.

Details DOI

YNIMG Journal 2021 Journal Article

Revisiting the T2 spectrum imaging inverse problem: Bayesian regularized non-negative least squares

  • Erick Jorge Canales-Rodríguez
  • Marco Pizzolato
  • Thomas Yu
  • Gian Franco Piredda
  • Tom Hilbert
  • Joaquim Radua
  • Tobias Kober
  • Jean-Philippe Thiran

Multi-echo T2 magnetic resonance images contain information about the distribution of T2 relaxation times of compartmentalized water, from which we can estimate relevant brain tissue properties such as the myelin water fraction (MWF). Regularized non-negative least squares (NNLS) is the tool of choice for estimating non-parametric T2 spectra. However, the estimation is ill-conditioned, sensitive to noise, and highly affected by the employed regularization weight. The purpose of this study is threefold: first, we want to underline that the apparently innocuous use of two alternative parameterizations for solving the inverse problem, which we called the standard and alternative regularization forms, leads to different solutions; second, to assess the performance of both parameterizations; and third, to propose a new Bayesian regularized NNLS method (BayesReg). The performance of BayesReg was compared with that of two conventional approaches (L-curve and Chi-square (X2) fitting) using both regularization forms. We generated a large dataset of synthetic data, acquired in vivo human brain data in healthy participants for conducting a scan-rescan analysis, and correlated the myelin content derived from histology with the MWF estimated from ex vivo data. Results from synthetic data indicate that BayesReg provides accurate MWF estimates, comparable to those from L-curve and X2, and with better overall stability across a wider signal-to-noise range. Notably, we obtained superior results by using the alternative regularization form. The correlations reported in this study are higher than those reported in previous studies employing the same ex vivo and histological data. In human brain data, the estimated maps from L-curve and BayesReg were more reproducible. However, the T2 spectra produced by BayesReg were less affected by over-smoothing than those from L-curve. These findings suggest that BayesReg is a good alternative for estimating T2 distributions and MWF maps.

Details DOI

YNIMG Journal 2021 Journal Article

Tractography dissection variability: What happens when 42 groups dissect 14 white matter bundles on the same dataset?

  • Kurt G. Schilling
  • François Rheault
  • Laurent Petit
  • Colin B. Hansen
  • Vishwesh Nath
  • Fang-Cheng Yeh
  • Gabriel Girard
  • Muhamed Barakovic

White matter bundle segmentation using diffusion MRI fiber tractography has become the method of choice to identify white matter fiber pathways in vivo in human brains. However, like other analyses of complex data, there is considerable variability in segmentation protocols and techniques. This can result in different reconstructions of the same intended white matter pathways, which directly affects tractography results, quantification, and interpretation. In this study, we aim to evaluate and quantify the variability that arises from different protocols for bundle segmentation. Through an open call to users of fiber tractography, including anatomists, clinicians, and algorithm developers, 42 independent teams were given processed sets of human whole-brain streamlines and asked to segment 14 white matter fascicles on six subjects. In total, we received 57 different bundle segmentation protocols, which enabled detailed volume-based and streamline-based analyses of agreement and disagreement among protocols for each fiber pathway. Results show that even when given the exact same sets of underlying streamlines, the variability across protocols for bundle segmentation is greater than all other sources of variability in the virtual dissection process, including variability within protocols and variability across subjects. In order to foster the use of tractography bundle dissection in routine clinical settings, and as a fundamental analytical tool, future endeavors must aim to resolve and reduce this heterogeneity. Although external validation is needed to verify the anatomical accuracy of bundle dissections, reducing heterogeneity is a step towards reproducible research and may be achieved through the use of standard nomenclature and definitions of white matter bundles and well-chosen constraints and decisions in the dissection process.

Details DOI

YNIMG Journal 2020 Journal Article

Adaptive phase correction of diffusion-weighted images

  • Marco Pizzolato
  • Guillaume Gilbert
  • Jean-Philippe Thiran
  • Maxime Descoteaux
  • Rachid Deriche

Phase correction (PC) is a preprocessing technique that exploits the phase of images acquired in Magnetic Resonance Imaging (MRI) to obtain real-valued images containing tissue contrast with additive Gaussian noise, as opposed to magnitude images which follow a non-Gaussian distribution, e.g. Rician. PC finds its natural application to diffusion-weighted images (DWIs) due to their inherent low signal-to-noise ratio and consequent non-Gaussianity that induces a signal overestimation bias that propagates to the calculated diffusion indices. PC effectiveness depends upon the quality of the phase estimation, which is often performed via a regularization procedure. We show that a suboptimal regularization can produce alterations of the true image contrast in the real-valued phase-corrected images. We propose adaptive phase correction (APC), a method where the phase is estimated by using MRI noise information to perform a complex-valued image regularization that accounts for the local variance of the noise. We show, on synthetic and acquired data, that APC leads to phase-corrected real-valued DWIs that present a reduced number of alterations and a reduced bias. The substantial absence of parameters for which human input is required favors a straightforward integration of APC in MRI processing pipelines.

Details DOI

YNICL Journal 2020 Journal Article

Multiple sclerosis cortical and WM lesion segmentation at 3T MRI: a deep learning method based on FLAIR and MP2RAGE

  • Francesco La Rosa
  • Ahmed Abdulkadir
  • Mário João Fartaria
  • Reza Rahmanzadeh
  • Po-Jui Lu
  • Riccardo Galbusera
  • Muhamed Barakovic
  • Jean-Philippe Thiran

The presence of cortical lesions in multiple sclerosis patients has emerged as an important biomarker of the disease. They appear in the earliest stages of the illness and have been shown to correlate with the severity of clinical symptoms. However, cortical lesions are hardly visible in conventional magnetic resonance imaging (MRI) at 3T, and thus their automated detection has been so far little explored. In this study, we propose a fully-convolutional deep learning approach, based on the 3D U-Net, for the automated segmentation of cortical and white matter lesions at 3T. For this purpose, we consider a clinically plausible MRI setting consisting of two MRI contrasts only: one conventional T2-weighted sequence (FLAIR), and one specialized T1-weighted sequence (MP2RAGE). We include 90 patients from two different centers with a total of 728 and 3856 gray and white matter lesions, respectively. We show that two reference methods developed for white matter lesion segmentation are inadequate to detect small cortical lesions, whereas our proposed framework is able to achieve a detection rate of 76% for both cortical and white matter lesions with a false positive rate of 29% in comparison to manual segmentation. Further results suggest that our framework generalizes well for both types of lesion in subjects acquired in two hospitals with different scanners.

Details DOI

YNIMG Journal 2020 Journal Article

On the cortical connectivity in the macaque brain: A comparison of diffusion tractography and histological tracing data

  • Gabriel Girard
  • Roberto Caminiti
  • Alexandra Battaglia-Mayer
  • Etienne St-Onge
  • Karen S. Ambrosen
  • Simon F. Eskildsen
  • Kristine Krug
  • Tim B. Dyrby

Diffusion-weighted magnetic resonance imaging (DW-MRI) tractography is a non-invasive tool to probe neural connections and the structure of the white matter. It has been applied successfully in studies of neurological disorders and normal connectivity. Recent work has revealed that tractography produces a high incidence of false-positive connections, often from “bottleneck” white matter configurations. The rich literature in histological connectivity analysis studies in the macaque monkey enables quantitative evaluation of the performance of tractography algorithms. In this study, we use the intricate connections of frontal, cingulate, and parietal areas, well established by the anatomical literature, to derive a symmetrical histological connectivity matrix composed of 59 cortical areas. We evaluate the performance of fifteen diffusion tractography algorithms, including global, deterministic, and probabilistic state-of-the-art methods for the connectivity predictions of 1711 distinct pairs of areas, among which 680 are reported connected by the literature. The diffusion connectivity analysis was performed on a different ex-vivo macaque brain, acquired using multi-shell DW-MRI protocol, at high spatial and angular resolutions. Across all tested algorithms, the true-positive and true-negative connections were dominant over false-positive and false-negative connections, respectively. Moreover, three-quarters of streamlines had endpoints location in agreement with histological data, on average. Furthermore, probabilistic streamline tractography algorithms show the best performances in predicting which areas are connected. Altogether, we propose a method for quantitative evaluation of tractography algorithms, which aims at improving the sensitivity and the specificity of diffusion-based connectivity analysis. Overall, those results confirm the usefulness of tractography in predicting connectivity, although errors are produced. Many of the errors result from bottleneck white matter configurations near the cortical grey matter and should be the target of future implementation of methods.

Details DOI

YNIMG Journal 2019 Journal Article

Limits to anatomical accuracy of diffusion tractography using modern approaches

  • Kurt G. Schilling
  • Vishwesh Nath
  • Colin Hansen
  • Prasanna Parvathaneni
  • Justin Blaber
  • Yurui Gao
  • Peter Neher
  • Dogu Baran Aydogan

Diffusion MRI fiber tractography is widely used to probe the structural connectivity of the brain, with a range of applications in both clinical and basic neuroscience. Despite widespread use, tractography has well-known pitfalls that limits the anatomical accuracy of this technique. Numerous modern methods have been developed to address these shortcomings through advances in acquisition, modeling, and computation. To test whether these advances improve tractography accuracy, we organized the 3-D Validation of Tractography with Experimental MRI (3D-VoTEM) challenge at the ISBI 2018 conference. We made available three unique independent tractography validation datasets – a physical phantom and two ex vivo brain specimens - resulting in 176 distinct submissions from 9 research groups. By comparing results over a wide range of fiber complexities and algorithmic strategies, this challenge provides a more comprehensive assessment of tractography's inherent limitations than has been reported previously. The central results were consistent across all sub-challenges in that, despite advances in tractography methods, the anatomical accuracy of tractography has not dramatically improved in recent years. Taken together, our results independently confirm findings from decades of tractography validation studies, demonstrate inherent limitations in reconstructing white matter pathways using diffusion MRI data alone, and highlight the need for alternative or combinatorial strategies to accurately map the fiber pathways of the brain.

Details DOI

YNIMG Journal 2019 Journal Article

Sparse wars: A survey and comparative study of spherical deconvolution algorithms for diffusion MRI

  • Erick Jorge Canales-Rodríguez
  • Jon Haitz Legarreta
  • Marco Pizzolato
  • Gaëtan Rensonnet
  • Gabriel Girard
  • Jonathan Rafael- Patino
  • Muhamed Barakovic
  • David Romascano

Spherical deconvolution methods are widely used to estimate the brain's white-matter fiber orientations from diffusion MRI data. In this study, eight spherical deconvolution algorithms were implemented and evaluated. These included two model selection techniques based on the extended Bayesian information criterion (i. e. , best subset selection and the least absolute shrinkage and selection operator), iteratively reweighted l 2- and l 1-norm approaches to approximate the l 0-norm, sparse Bayesian learning, Cauchy deconvolution, and two accelerated Richardson-Lucy algorithms. Results from our exhaustive evaluation show that there is no single optimal method for all different fiber configurations, suggesting that further studies should be conducted to find the optimal way of combining solutions from different methods. We found l 0-norm regularization algorithms to resolve more accurately fiber crossings with small inter-fiber angles. However, in voxels with very dominant fibers, algorithms promoting more sparsity are less accurate in detecting smaller fibers. In most cases, the best algorithm to reconstruct fiber crossings with two fibers did not perform optimally in voxels with one or three fibers. Therefore, simplified validation systems as employed in a number of previous studies, where only two fibers with similar volume fractions were tested, should be avoided as they provide incomplete information. Future studies proposing new reconstruction methods based on high angular resolution diffusion imaging data should validate their results by considering, at least, voxels with one, two, and three fibers, as well as voxels with dominant fibers and different diffusion anisotropies.

Details DOI

YNIMG Journal 2019 Journal Article

Towards microstructure fingerprinting: Estimation of tissue properties from a dictionary of Monte Carlo diffusion MRI simulations

  • Gaëtan Rensonnet
  • Benoît Scherrer
  • Gabriel Girard
  • Aleksandar Jankovski
  • Simon K. Warfield
  • Benoît Macq
  • Jean-Philippe Thiran
  • Maxime Taquet

Many closed-form analytical models have been proposed to relate the diffusion-weighted magnetic resonance imaging (DW-MRI) signal to microstructural features of white matter tissues. These models generally make assumptions about the tissue and the diffusion processes which often depart from the biophysical reality, limiting their reliability and interpretability in practice. Monte Carlo simulations of the random walk of water molecules are widely recognized to provide near groundtruth for DW-MRI signals. However, they have mostly been limited to the validation of simpler models rather than used for the estimation of microstructural properties. This work proposes a general framework which leverages Monte Carlo simulations for the estimation of physically interpretable microstructural parameters, both in single and in crossing fascicles of axons. Monte Carlo simulations of DW-MRI signals, or fingerprints, are pre-computed for a large collection of microstructural configurations. At every voxel, the microstructural parameters are estimated by optimizing a sparse combination of these fingerprints. Extensive synthetic experiments showed that our approach achieves accurate and robust estimates in the presence of noise and uncertainties over fixed or input parameters. In an in vivo rat model of spinal cord injury, our approach provided microstructural parameters that showed better correspondence with histology than five closed-form models of the diffusion signal: MMWMD, NODDI, DIAMOND, WMTI and MAPL. On whole-brain in vivo data from the human connectome project (HCP), our method exhibited spatial distributions of apparent axonal radius and axonal density indices in keeping with ex vivo studies. This work paves the way for microstructure fingerprinting with Monte Carlo simulations used directly at the modeling stage and not only as a validation tool.

Details DOI

YNIMG Journal 2017 Journal Article

Transient networks of spatio-temporal connectivity map communication pathways in brain functional systems

  • Alessandra Griffa
  • Benjamin Ricaud
  • Kirell Benzi
  • Xavier Bresson
  • Alessandro Daducci
  • Pierre Vandergheynst
  • Jean-Philippe Thiran
  • Patric Hagmann

The study of brain dynamics enables us to characterize the time-varying functional connectivity among distinct neural groups. However, current methods suffer from the absence of structural connectivity information. We propose to integrate infra-slow neural oscillations and anatomical-connectivity maps, as derived from functional and diffusion MRI, in a multilayer-graph framework that captures transient networks of spatio-temporal connectivity. These networks group anatomically wired and temporary synchronized brain regions and encode the propagation of functional activity on the structural connectome. In a group of 71 healthy subjects, we find that these transient networks demonstrate power-law spatial and temporal size, globally organize into well-known functional systems and describe wave-like trajectories of activation across anatomically connected regions. Within the transient networks, activity propagates through polysynaptic paths that include selective ensembles of structural connections and differ from the structural shortest paths. In the light of the communication-through-coherence principle, the identified spatio-temporal networks could encode communication channels' selection and neural assemblies, which deserves further attention. This work contributes to the understanding of brain structure-function relationships by considering the time-varying nature of resting-state interactions on the axonal scaffold, and it offers a convenient framework to study large-scale communication mechanisms and functional dynamics.

Details DOI

YNICL Journal 2016 Journal Article

Brain network characterization of high-risk preterm-born school-age children

  • Elda Fischi-Gomez
  • Emma Muñoz-Moreno
  • Lana Vasung
  • Alessandra Griffa
  • Cristina Borradori-Tolsa
  • Maryline Monnier
  • François Lazeyras
  • Jean-Philippe Thiran

Higher risk for long-term cognitive and behavioral impairments is one of the hallmarks of extreme prematurity (EP) and pregnancy-associated fetal adverse conditions such as intrauterine growth restriction (IUGR). While neurodevelopmental delay and abnormal brain function occur in the absence of overt brain lesions, these conditions have been recently associated with changes in microstructural brain development. Recent imaging studies indicate changes in brain connectivity, in particular involving the white matter fibers belonging to the cortico-basal ganglia-thalamic loop. Furthermore, EP and IUGR have been related to altered brain network architecture in childhood, with reduced network global capacity, global efficiency and average nodal strength. In this study, we used a connectome analysis to characterize the structural brain networks of these children, with a special focus on their topological organization. On one hand, we confirm the reduced averaged network node degree and strength due to EP and IUGR. On the other, the decomposition of the brain networks in an optimal set of clusters remained substantially different among groups, talking in favor of a different network community structure. However, and despite the different community structure, the brain networks of these high-risk school-age children maintained the typical small-world, rich-club and modularity characteristics in all cases. Thus, our results suggest that brain reorganizes after EP and IUGR, prioritizing a tight modular structure, to maintain the small-world, rich-club and modularity characteristics. By themselves, both extreme prematurity and IUGR bear a similar risk for neurocognitive and behavioral impairment, and the here defined modular network alterations confirm similar structural changes both by IUGR and EP at school age compared to control. Interestingly, the combination of both conditions (IUGR + EP) does not result in a worse outcome. In such cases, the alteration in network topology appears mainly driven by the effect of extreme prematurity, suggesting that these brain network alterations present at school age have their origin in a common critical period, both for intrauterine and extrauterine adverse conditions.

Details DOI

YNIMG Journal 2016 Journal Article

Comparison of accelerated T1-weighted whole-brain structural-imaging protocols

  • Pavel Falkovskiy
  • Daniel Brenner
  • Thorsten Feiweier
  • Stephan Kannengiesser
  • Bénédicte Maréchal
  • Tobias Kober
  • Alexis Roche
  • Kaely Thostenson

Imaging in neuroscience, clinical research and pharmaceutical trials often employs the 3D magnetisation-prepared rapid gradient-echo (MPRAGE) sequence to obtain structural T1-weighted images with high spatial resolution of the human brain. Typical research and clinical routine MPRAGE protocols with ~1mm isotropic resolution require data acquisition time in the range of 5–10min and often use only moderate two-fold acceleration factor for parallel imaging. Recent advances in MRI hardware and acquisition methodology promise improved leverage of the MR signal and more benign artefact properties in particular when employing increased acceleration factors in clinical routine and research. In this study, we examined four variants of a four-fold-accelerated MPRAGE protocol (2D-GRAPPA, CAIPIRINHA, CAIPIRINHA elliptical, and segmented MPRAGE) and compared clinical readings, basic image quality metrics (SNR, CNR), and automated brain tissue segmentation for morphological assessments of brain structures. The results were benchmarked against a widely-used two-fold-accelerated 3T ADNI MPRAGE protocol that served as reference in this study. 22 healthy subjects (age=20–44yrs.) were imaged with all MPRAGE variants in a single session. An experienced reader rated all images of clinically useful image quality. CAIPIRINHA MPRAGE scans were perceived on average to be of identical value for reading as the reference ADNI-2 protocol. SNR and CNR measurements exhibited the theoretically expected performance at the four-fold acceleration. The results of this study demonstrate that the four-fold accelerated protocols introduce systematic biases in the segmentation results of some brain structures compared to the reference ADNI-2 protocol. Furthermore, results suggest that the increased noise levels in the accelerated protocols play an important role in introducing these biases, at least under the present study conditions.

Details DOI

YNIMG Journal 2016 Journal Article

Generative models of the human connectome

  • Richard F. Betzel
  • Andrea Avena-Koenigsberger
  • Joaquín Goñi
  • Ye He
  • Marcel A. de Reus
  • Alessandra Griffa
  • Petra E. Vértes
  • Bratislav Mišic

The human connectome represents a network map of the brain's wiring diagram and the pattern into which its connections are organized is thought to play an important role in cognitive function. The generative rules that shape the topology of the human connectome remain incompletely understood. Earlier work in model organisms has suggested that wiring rules based on geometric relationships (distance) can account for many but likely not all topological features. Here we systematically explore a family of generative models of the human connectome that yield synthetic networks designed according to different wiring rules combining geometric and a broad range of topological factors. We find that a combination of geometric constraints with a homophilic attachment mechanism can create synthetic networks that closely match many topological characteristics of individual human connectomes, including features that were not included in the optimization of the generative model itself. We use these models to investigate a lifespan dataset and show that, with age, the model parameters undergo progressive changes, suggesting a rebalancing of the generative factors underlying the connectome across the lifespan.

Details DOI

YNIMG Journal 2016 Journal Article

Surface-driven registration method for the structure-informed segmentation of diffusion MR images

  • Oscar Esteban
  • Dominique Zosso
  • Alessandro Daducci
  • Meritxell Bach-Cuadra
  • María J. Ledesma-Carbayo
  • Jean-Philippe Thiran
  • Andres Santos

Current methods for processing diffusion MRI (dMRI) to map the connectivity of the human brain require precise delineations of anatomical structures. This requirement has been approached by either segmenting the data in native dMRI space or mapping the structural information from T1-weighted (T1w) images. The characteristic features of diffusion data in terms of signal-to-noise ratio, resolution, as well as the geometrical distortions caused by the inhomogeneity of magnetic susceptibility across tissues hinder both solutions. Unifying the two approaches, we propose regseg, a surface-to-volume nonlinear registration method that segments homogeneous regions within multivariate images by mapping a set of nested reference-surfaces. Accurate surfaces are extracted from a T1w image of the subject, using as target image the bivariate volume comprehending the fractional anisotropy (FA) and the apparent diffusion coefficient (ADC) maps derived from the dMRI dataset. We first verify the accuracy of regseg on a general context using digital phantoms distorted with synthetic and random deformations. Then we establish an evaluation framework using undistorted dMRI data from the Human Connectome Project (HCP) and realistic deformations derived from the inhomogeneity fieldmap corresponding to each subject. We analyze the performance of regseg computing the misregistration error of the surfaces estimated after being mapped with regseg onto 16 datasets from the HCP. The distribution of errors shows a 95% CI of 0. 56–0. 66mm, that is below the dMRI resolution (1. 25mm, isotropic). Finally, we cross-compare the proposed tool against a nonlinear b0-to-T2w registration method, thereby obtaining a significantly lower misregistration error with regseg. The accurate mapping of structural information in dMRI space is fundamental to increase the reliability of network building in connectivity analyses, and to improve the performance of the emerging structure-informed techniques for dMRI data processing.

Details DOI

YNIMG Journal 2015 Journal Article

Accelerated Microstructure Imaging via Convex Optimization (AMICO) from diffusion MRI data

  • Alessandro Daducci
  • Erick J. Canales-Rodríguez
  • Hui Zhang
  • Tim B. Dyrby
  • Daniel C. Alexander
  • Jean-Philippe Thiran

Microstructure imaging from diffusion magnetic resonance (MR) data represents an invaluable tool to study non-invasively the morphology of tissues and to provide a biological insight into their microstructural organization. In recent years, a variety of biophysical models have been proposed to associate particular patterns observed in the measured signal with specific microstructural properties of the neuronal tissue, such as axon diameter and fiber density. Despite very appealing results showing that the estimated microstructure indices agree very well with histological examinations, existing techniques require computationally very expensive non-linear procedures to fit the models to the data which, in practice, demand the use of powerful computer clusters for large-scale applications. In this work, we present a general framework for Accelerated Microstructure Imaging via Convex Optimization (AMICO) and show how to re-formulate this class of techniques as convenient linear systems which, then, can be efficiently solved using very fast algorithms. We demonstrate this linearization of the fitting problem for two specific models, i. e. ActiveAx and NODDI, providing a very attractive alternative for parameter estimation in those techniques; however, the AMICO framework is general and flexible enough to work also for the wider space of microstructure imaging methods. Results demonstrate that AMICO represents an effective means to accelerate the fit of existing techniques drastically (up to four orders of magnitude faster) while preserving accuracy and precision in the estimated model parameters (correlation above 0. 9). We believe that the availability of such ultrafast algorithms will help to accelerate the spread of microstructure imaging to larger cohorts of patients and to study a wider spectrum of neurological disorders.

Details DOI

YNIMG Journal 2015 Journal Article

An efficient total variation algorithm for super-resolution in fetal brain MRI with adaptive regularization

  • Sébastien Tourbier
  • Xavier Bresson
  • Patric Hagmann
  • Jean-Philippe Thiran
  • Reto Meuli
  • Meritxell Bach Cuadra

Although fetal anatomy can be adequately viewed in new multi-slice MR images, many critical limitations remain for quantitative data analysis. To this end, several research groups have recently developed advanced image processing methods, often denoted by super-resolution (SR) techniques, to reconstruct from a set of clinical low-resolution (LR) images, a high-resolution (HR) motion-free volume. It is usually modeled as an inverse problem where the regularization term plays a central role in the reconstruction quality. Literature has been quite attracted by Total Variation energies because of their ability in edge preserving but only standard explicit steepest gradient techniques have been applied for optimization. In a preliminary work, it has been shown that novel fast convex optimization techniques could be successfully applied to design an efficient Total Variation optimization algorithm for the super-resolution problem. In this work, two major contributions are presented. Firstly, we will briefly review the Bayesian and Variational dual formulations of current state-of-the-art methods dedicated to fetal MRI reconstruction. Secondly, we present an extensive quantitative evaluation of our SR algorithm previously introduced on both simulated fetal and real clinical data (with both normal and pathological subjects). Specifically, we study the robustness of regularization terms in front of residual registration errors and we also present a novel strategy for automatically select the weight of the regularization as regards the data fidelity term. Our results show that our TV implementation is highly robust in front of motion artifacts and that it offers the best trade-off between speed and accuracy for fetal MRI recovery as in comparison with state-of-the art methods.

Details DOI

YNIMG Journal 2015 Journal Article

Improved statistical evaluation of group differences in connectomes by screening–filtering strategy with application to study maturation of brain connections between childhood and adolescence

  • Djalel-Eddine Meskaldji
  • Lana Vasung
  • David Romascano
  • Jean-Philippe Thiran
  • Patric Hagmann
  • Stephan Morgenthaler
  • Dimitri Van De Ville

Detecting local differences between groups of connectomes is a great challenge in neuroimaging, because the large number of tests that have to be performed and the impact on multiplicity correction. Any available information should be exploited to increase the power of detecting true between-group effects. We present an adaptive strategy that exploits the data structure and the prior information concerning positive dependence between nodes and connections, without relying on strong assumptions. As a first step, we decompose the brain network, i. e. , the connectome, into subnetworks and we apply a screening at the subnetwork level. The subnetworks are defined either according to prior knowledge or by applying a data driven algorithm. Given the results of the screening step, a filtering is performed to seek real differences at the node/connection level. The proposed strategy could be used to strongly control either the family-wise error rate or the false discovery rate. We show by means of different simulations the benefit of the proposed strategy, and we present a real application of comparing connectomes of preschool children and adolescents.

Details DOI

YNICL Journal 2014 Journal Article

Connectivity and tissue microstructural alterations in right and left temporal lobe epilepsy revealed by diffusion spectrum imaging

  • Alia Lemkaddem
  • Alessandro Daducci
  • Nicolas Kunz
  • François Lazeyras
  • Margitta Seeck
  • Jean-Philippe Thiran
  • Serge Vulliémoz

Focal epilepsy is increasingly recognized as the result of an altered brain network, both on the structural and functional levels and the characterization of these widespread brain alterations is crucial for our understanding of the clinical manifestation of seizure and cognitive deficits as well as for the management of candidates to epilepsy surgery. Tractography based on Diffusion Tensor Imaging allows non-invasive mapping of white matter tracts in vivo. Recently, diffusion spectrum imaging (DSI), based on an increased number of diffusion directions and intensities, has improved the sensitivity of tractography, notably with respect to the problem of fiber crossing and recent developments allow acquisition times compatible with clinical application. We used DSI and parcellation of the gray matter in regions of interest to build whole-brain connectivity matrices describing the mutual connections between cortical and subcortical regions in patients with focal epilepsy and healthy controls. In addition, the high angular and radial resolution of DSI allowed us to evaluate also some of the biophysical compartment models, to better understand the cause of the changes in diffusion anisotropy. Global connectivity, hub architecture and regional connectivity patterns were altered in TLE patients and showed different characteristics in RTLE vs LTLE with stronger abnormalities in RTLE. The microstructural analysis suggested that disturbed axonal density contributed more than fiber orientation to the connectivity changes affecting the temporal lobes whereas fiber orientation changes were more involved in extratemporal lobe changes. Our study provides further structural evidence that RTLE and LTLE are not symmetrical entities and DSI-based imaging could help investigate the microstructural correlate of these imaging abnormalities.

Details DOI

YNIMG Journal 2013 Journal Article

Comparing connectomes across subjects and populations at different scales

  • Djalel Eddine Meskaldji
  • Elda Fischi-Gomez
  • Alessandra Griffa
  • Patric Hagmann
  • Stephan Morgenthaler
  • Jean-Philippe Thiran

Brain connectivity can be represented by a network that enables the comparison of the different patterns of structural and functional connectivity among individuals. In the literature, two levels of statistical analysis have been considered in comparing brain connectivity across groups and subjects: 1) the global comparison where a single measure that summarizes the information of each brain is used in a statistical test; 2) the local analysis where a single test is performed either for each node/connection which implies a multiplicity correction, or for each group of nodes/connections where each subset is summarized by one single test in order to reduce the number of tests to avoid a penalizing multiplicity correction. We comment on the different levels of analysis and present some methods that have been proposed at each scale. We highlight as well the possible factors that could influence the statistical results and the questions that have to be addressed in such an analysis.

Details DOI