Search

Gliomas are the most frequent primary brain tumors in adults. Glioma change detection aims at finding the relevant parts of the image that change over time. Although Deep Learning (DL) shows promising performances in similar change detection tasks, the creation of large annotated datasets represents a major bottleneck for supervised DL applications in radiology. To overcome this, we propose a combined use of weak labels (imprecise, but fast-to-create annotations) and Transfer Learning (TL). Specifically, we explore inductive TL, where source and target domains are identical, but tasks are different due to a label shift: our target labels are created manually by three radiologists, whereas our source weak labels are generated automatically from radiology reports via NLP. We frame knowledge transfer as hyperparameter optimization, thus avoiding heuristic choices that are frequent in related works. We investigate the relationship between model size and TL, comparing a low-capacity VGG with a higher-capacity ResNeXt model. We evaluate our models on 1693 T2-weighted magnetic resonance imaging difference maps created from 183 patients, by classifying them into stable or unstable according to tumor evolution. The weak labels extracted from radiology reports allowed us to increase dataset size more than 3-fold, and improve VGG classification results from 75% to 82% AUC. Mixed training from scratch led to higher performance than fine-tuning or feature extraction. To assess generalizability, we ran inference on an open dataset (BraTS-2015: 15 patients, 51 difference maps), reaching up to 76% AUC. Overall, results suggest that medical imaging problems may benefit from smaller models and different TL strategies with respect to computer vision datasets, and that report-generated weak labels are effective in improving model performances. Code, in-house dataset and BraTS labels are released., Comment: This work has been submitted as Original Paper to a Journal

Accurate characterization of in utero human brain maturation is critical as it involves complex and interconnected structural and functional processes that may influence health later in life. Magnetic resonance imaging is a powerful tool to investigate equivocal neurological patterns during fetal development. However, the number of acquisitions of satisfactory quality available in this cohort of sensitive subjects remains scarce, thus hindering the validation of advanced image processing techniques. Numerical phantoms can mitigate these limitations by providing a controlled environment with a known ground truth. In this work, we present FaBiAN, an open-source Fetal Brain magnetic resonance Acquisition Numerical phantom that simulates clinical T2-weighted fast spin echo sequences of the fetal brain. This unique tool is based on a general, flexible and realistic setup that includes stochastic fetal movements, thus providing images of the fetal brain throughout maturation comparable to clinical acquisitions. We demonstrate its value to evaluate the robustness and optimize the accuracy of an algorithm for super-resolution fetal brain magnetic resonance imaging from simulated motion-corrupted 2D low-resolution series as compared to a synthetic high-resolution reference volume. We also show that the images generated can complement clinical datasets to support data-intensive deep learning methods for fetal brain tissue segmentation., Comment: 23 pages, 9 figures (including Supplementary Material), 4 tables, 1 supplement. Submitted to Scientific Reports (2021)

Brain aneurysm detection in Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) has undergone drastic improvements with the advent of Deep Learning (DL). However, performances of supervised DL models heavily rely on the quantity of labeled samples, which are extremely costly to obtain. Here, we present a DL model for aneurysm detection that overcomes the issue with ''weak'' labels: oversized annotations which are considerably faster to create. Our weak labels resulted to be four times faster to generate than their voxel-wise counterparts. In addition, our model leverages prior anatomical knowledge by focusing only on plausible locations for aneurysm occurrence. We frst train and evaluate our model through cross-validation on an in-house TOF-MRA dataset comprising 284 subjects (170 females / 127 healthy controls / 157 patients with 198 aneurysms). On this dataset, our best model achieved a sensitivity of 83%, with False Positive (FP) rate of 0.8 per patient. To assess model generalizability, we then participated in a challenge for aneurysm detection with TOF-MRA data (93 patients, 20 controls, 125 aneurysms). On the public challenge, sensitivity was 68% (FP rate=2.5), ranking 4th/18 on the open leaderboard. We found no signifcant diference in sensitivity between aneurysm risk-of-rupture groups (p=0.75), locations (p=0.72), or sizes (p=0.15). Data, code and model weights are released under permissive licenses. We demonstrate that weak labels and anatomical knowledge can alleviate the necessity for prohibitively expensive voxel-wise annotations., Comment: Paper accepted as Original Article in the journal Neuroinformatics (https://link.springer.com/article/10.1007/s12021-022-09597-0)

A commonly adopted approach to carry out detection tasks in medical imaging is to rely on an initial segmentation. However, this approach strongly depends on voxel-wise annotations which are repetitive and time-consuming to draw for medical experts. An interesting alternative to voxel-wise masks are so-called "weak" labels: these can either be coarse or oversized annotations that are less precise, but noticeably faster to create. In this work, we address the task of brain aneurysm detection as a patch-wise binary classification with weak labels, in contrast to related studies that rather use supervised segmentation methods and voxel-wise delineations. Our approach comes with the non-trivial challenge of the data set creation: as for most focal diseases, anomalous patches (with aneurysm) are outnumbered by those showing no anomaly, and the two classes usually have different spatial distributions. To tackle this frequent scenario of inherently imbalanced, spatially skewed data sets, we propose a novel, anatomically-driven approach by using a multi-scale and multi-input 3D Convolutional Neural Network (CNN). We apply our model to 214 subjects (83 patients, 131 controls) who underwent Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) and presented a total of 111 unruptured cerebral aneurysms. We compare two strategies for negative patch sampling that have an increasing level of difficulty for the network and we show how this choice can strongly affect the results. To assess whether the added spatial information helps improving performances, we compare our anatomically-informed CNN with a baseline, spatially-agnostic CNN. When considering the more realistic and challenging scenario including vessel-like negative patches, the former model attains the highest classification results (accuracy$\simeq$95\%, AUROC$\simeq$0.95, AUPR$\simeq$0.71), thus outperforming the baseline., Comment: 11 pages, 2 figures, Paper accepted at the "Machine Learning in Clinical Neuroimaging" (MLCN) MICCAI workshop 2020 (https://mlcnws.com/)

Relaxometry studies in preterm and at-term newborns have provided insight into brain microstructure, thus opening new avenues for studying normal brain development and supporting diagnosis in equivocal neurological situations. However, such quantitative techniques require long acquisition times and therefore cannot be straightforwardly translated to in utero brain developmental studies. In clinical fetal brain magnetic resonance imaging routine, 2D low-resolution T2-weighted fast spin echo sequences are used to minimize the effects of unpredictable fetal motion during acquisition. As super-resolution techniques make it possible to reconstruct a 3D high-resolution volume of the fetal brain from clinical low-resolution images, their combination with quantitative acquisition schemes could provide fast and accurate T2 measurements. In this context, the present work demonstrates the feasibility of using super-resolution reconstruction from conventional T2-weighted fast spin echo sequences for 3D isotropic T2 mapping. A quantitative magnetic resonance phantom was imaged using a clinical T2-weighted fast spin echo sequence at variable echo time to allow for super-resolution reconstruction at every echo time and subsequent T2 mapping of samples whose relaxometric properties are close to those of fetal brain tissue. We demonstrate that this approach is highly repeatable, accurate and robust when using six echo times (total acquisition time under 9 minutes) as compared to gold-standard single-echo spin echo sequences (several hours for one single 2D slice)., Comment: 11 pages, 4 figures, 1 table, 1 supplement, to appear in Proceedings in Medical Image Computing and Computer Assisted Intervention (MICCAI), Peru, October 2020

Structural connectivity in the brain is typically studied by reducing its observation to a single spatial resolution. However, the brain possesses a rich architecture organized over multiple scales linked to one another. We explored the multiscale organization of human connectomes using datasets of healthy subjects reconstructed at five different resolutions. We found that the structure of the human brain remains self-similar when the resolution of observation is progressively decreased by hierarchical coarse-graining of the anatomical regions. Strikingly, a geometric network model, where distances are not Euclidean, predicts the multiscale properties of connectomes, including self-similarity. The model relies on the application of a geometric renormalization protocol which decreases the resolution by coarse-graining and averaging over short similarity distances. Our results suggest that simple organizing principles underlie the multiscale architecture of human structural brain networks, where the same connectivity law dictates short- and long-range connections between different brain regions over many resolutions. The implications are varied and can be substantial for fundamental debates, such as whether the brain is working near a critical point, as well as for applications including advanced tools to simplify the digital reconstruction and simulation of the brain.

Communication of signals among nodes in a complex network poses fundamental problems of efficiency and cost. Routing of messages along shortest paths requires global information about the topology, while spreading by diffusion, which operates according to local topological features, is informationally "cheap" but inefficient. We introduce a stochastic model for network communication that combines varying amounts of local and global information about the network topology. The model generates a continuous spectrum of dynamics that converge onto shortest-path and random-walk (diffusion) communication processes at the limiting extremes. We implement the model on two cohorts of human connectome networks and investigate the effects of varying amounts of local and global information on the network's communication cost. We identify routing strategies that approach a (highly efficient) shortest-path communication process with a relatively small amount of global information. Moreover, we show that the cost of routing messages from and to hub nodes varies as a function of the amount of global information driving the system's dynamics. Finally, we implement the model to identify individual subject differences from a communication dynamics point of view. The present framework departs from the classical shortest paths vs. diffusion dichotomy, suggesting instead that brain networks may exhibit different types of communication dynamics depending on varying functional demands and the availability of resources.

Stochastic resonance is a phenomenon in which noise enhances the response of a system to an input signal. The brain is an example of a system that has to detect and transmit signals in a noisy environment, suggesting that it is a good candidate to take advantage of SR. In this work, we aim to identify the optimal levels of noise that promote signal transmission through a simple network model of the human brain. Specifically, using a dynamic model implemented on an anatomical brain network (connectome), we investigate the similarity between an input signal and a signal that has traveled across the network while the system is subject to different noise levels. We find that non-zero levels of noise enhance the similarity between the input signal and the signal that has traveled through the system. The optimal noise level is not unique; rather, there is a set of parameter values at which the information is transmitted with greater precision, this set corresponds to the parameter values that place the system in a critical regime. The multiplicity of critical points in our model allows it to adapt to different noise situations and remain at criticality.

Various hypotheses exist about the paths used for communication between the nodes of complex networks. Most studies simply suppose that communication goes via shortest paths, while others have more explicit assumptions about how routing (alternatively navigation or search) works or should work in real networks. However, these assumptions are rarely checked against real data. Here we directly analyze the structure of operational paths using real measurements. For this purpose we use existing and newly created datasets having both the topology of the network and a sufficient number of empirically-determined paths over it. Such datasets are processed for air transportation networks, the human brain, the Internet and the fit-fat-cat word ladder game. Our results suggest that from the great number of possible paths, nature seems to pick according to some simple rules, which we will refer to as routing policies. First we confirm, that the preference of short paths is an inevitable policy element, however the observed stretch of the paths suggests that there are other policies at work simultaneously. We identify two additional policies common in our networks: the "conform hierarchy", meaning that the paths should obey the structural hierarchy of the network, and the "prefer downstream" policy which promotes avoiding the network core if possible. Building upon these simple policies, we propose a synthetic routing policy which can recover the basic statistical properties of the operational paths in networks. Our results can be helpful in estimating the reaction of complex systems for stress coming from the outside more accurately than the shortest path assumption permits., Comment: 8 pages 5 figures

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., Comment: 38 pages, 5 figures + 19 supplemental figures, 1 table

We consider the problem of testing positively dependent multiple hypotheses assuming that a prior information about the dependence structure is available. We propose two-step multiple comparisons procedures that exploit the prior information of the dependence structure, without relying on strong assumptions. In the first step, we group the tests into subsets where tests are supposed to be positively dependent and in each of which we compute the standardized mean of the test scores. Given the subset mean scores or equivalently the subsets p-values, we apply a first screening at a predefined threshold, which results in two types of subsets. Based on this typing, the original single test p-values are modified such that they can be used in conjunction with any multiple comparison procedure. We show by means of different simulation that power is gained with the proposed two-step methods, and compare it with traditional multiple comparison procedures. As an illustration, our method is applied on real data comparing topological differences between two groups of human brain networks.

The human connectome has been widely studied over the past decade. A principal finding is that it can be decomposed into communities of densely interconnected brain regions. This result, however, may be limited methodologically. Past studies have often used a flawed modularity measure in order to infer the connectome's community structure. Also, these studies relied on the intuition that community structure is best defined in terms of a network's static topology as opposed to a more dynamical definition. In this report we used the partition stability framework, which defines communities in terms of a Markov process (random walk), to infer the connectome's multi-scale community structure. Comparing the community structure to observed resting-state functional connectivity revealed communities across a broad range of dynamical scales that were closely related to functional connectivity. This result suggests a mapping between communities in structural networks, models of communication processes, and brain function. It further suggests that communication in the brain is not limited to a single characteristic scale, leading us to posit a heuristic for scale-selective communication in the cerebral cortex., Comment: Corrected small typographical mistakes, changed order of authors and funding information, and also chose a more efficient compression for figures

In neuroimaging, a large number of correlated tests are routinely performed to detect active voxels in single-subject experiments or to detect regions that differ between individuals belonging to different groups. In order to bound the probability of a false discovery of pair-wise differences, a Bonferroni or other correction for multiplicity is necessary. These corrections greatly reduce the power of the comparisons which means that small signals (differences) remain hidden and therefore have been more or less successful depending on the application. We introduce a method that improves the power of a family of correlated statistical tests by reducing their number in an orderly fashion using our a-priori understanding of the problem . The tests are grouped by blocks that respect the data structure and only one or a few tests per group are performed. For each block we construct an appropriate summary statistic that characterizes a meaningful feature of the block. The comparisons are based on these summary statistics by a block-wise approach. We contrast this method with the one based on the individual measures in terms of power. Finally, we apply the method to compare brain connectivity matrices. Although the method is used in this study on the particular case of imaging, the proposed strategy can be applied to a large variety of problems that involves multiple comparisons when the tests can be grouped according to attributes that depend on the specific problem. Keywords and phrases: Multiple comparisons ; Family-wise error rate; False discovery rate; Bonferroni procedure; Human brain connectivity; Brain connectivity matrices.