arXiv subjects

Natalia S. Rost

Publications and source records attributed to Natalia S. Rost.

9 recordsLinked to original sources

arXiv2024

Brain volume is a better biomarker of outcomes in ischemic stroke compared to brain atrophy

Brain parenchymal fraction (BPF) has been used as a surrogate measure of global brain atrophy, and as a biomarker of brain reserve in studies evaluating clinical outcomes after brain injury. Total brain volume at the time of injury has recently been shown to influence functional outcomes, where larger brain volumes are associated with better outcomes. Here, we assess if brain volume at the time of ischemic stroke injury is a better biomarker of functional outcome than BPF. Acute ischemic stroke cases at a single center between 2003 and 2011, with MR neuroimaging obtained within 48 hours from presentation were eligible. Functional outcomes represented by the modified Rankin Score (mRS) at 90 days post admission (mRS<3 deemed a favorable outcome) were obtained via patient interview or per chart review. Deep learning enabled automated segmentation pipelines were used to calculate brain volume, intracranial volume (ICV), and BPF on the acute neuroimaging data. Patient outcomes were modeled through logistic regressions, and model comparison was conducted using the Bayes Information Criterion (BIC). 467 patients with arterial ischemic stroke were included in the analysis. Median age was 65.8 years, and 65.3% were male. In both models, age and a larger stroke lesion volume were associated with worse functional outcomes. Higher BPF and a larger brain volume were both associated with favorable functional outcomes, however, comparison of both models suggested that the brain volume model (BIC=501) explains the data better compared to the BPF model (BIC=511). The extent of global brain atrophy has been regarded as an important biomarker of post-stroke functional outcomes and resilience to acute injury. Here, we demonstrate that a higher global brain volume at the time of injury better explains favorable functional outcomes, which can be directly clinically assessed.

q-bio.NC↗

arXiv2021

Hypernet-Ensemble Learning of Segmentation Probability for Medical Image Segmentation with Ambiguous Labels

Despite the superior performance of Deep Learning (DL) on numerous segmentation tasks, the DL-based approaches are notoriously overconfident about their prediction with highly polarized label probability. This is often not desirable for many applications with the inherent label ambiguity even in human annotations. This challenge has been addressed by leveraging multiple annotations per image and the segmentation uncertainty. However, multiple per-image annotations are often not available in a real-world application and the uncertainty does not provide full control on segmentation results to users. In this paper, we propose novel methods to improve the segmentation probability estimation without sacrificing performance in a real-world scenario that we have only one ambiguous annotation per image. We marginalize the estimated segmentation probability maps of networks that are encouraged to under-/over-segment with the varying Tversky loss without penalizing balanced segmentation. Moreover, we propose a unified hypernetwork ensemble method to alleviate the computational burden of training multiple networks. Our approaches successfully estimated the segmentation probability maps that reflected the underlying structures and provided the intuitive control on segmentation for the challenging 3D medical image segmentation. Although the main focus of our proposed methods is not to improve the binary segmentation performance, our approaches marginally outperformed the state-of-the-arts. The codes are available at \url{https://github.com/sh4174/HypernetEnsemble}.

eess.IV↗

arXiv2021

3D-StyleGAN: A Style-Based Generative Adversarial Network for Generative Modeling of Three-Dimensional Medical Images

Image synthesis via Generative Adversarial Networks (GANs) of three-dimensional (3D) medical images has great potential that can be extended to many medical applications, such as, image enhancement and disease progression modeling. However, current GAN technologies for 3D medical image synthesis need to be significantly improved to be readily adapted to real-world medical problems. In this paper, we extend the state-of-the-art StyleGAN2 model, which natively works with two-dimensional images, to enable 3D image synthesis. In addition to the image synthesis, we investigate the controllability and interpretability of the 3D-StyleGAN via style vectors inherited form the original StyleGAN2 that are highly suitable for medical applications: (i) the latent space projection and reconstruction of unseen real images, and (ii) style mixing. We demonstrate the 3D-StyleGAN's performance and feasibility with ~12,000 three-dimensional full brain MR T1 images, although it can be applied to any 3D volumetric images. Furthermore, we explore different configurations of hyperparameters to investigate potential improvement of the image synthesis with larger networks. The codes and pre-trained networks are available online: https://github.com/sh4174/3DStyleGAN.

eess.IV↗

arXiv2020

Brain volume: An important determinant of functional outcome after acute ischemic stroke

Objective: To determine whether brain volume is associated with functional outcome after acute ischemic stroke (AIS). Methods: We analyzed cross-sectional data of the multi-site, international hospital-based MRI-GENetics Interface Exploration (MRI-GENIE) study (July 1, 2014- March 16, 2019) with clinical brain magnetic resonance imaging (MRI) obtained on admission for index stroke and functional outcome assessment. Post-stroke outcome was determined using the modified Rankin Scale (mRS) score (0-6; 0: asymptomatic; 6 death) recorded between 60-190 days after stroke. Demographics and other clinical variables including acute stroke severity (measured as National Institutes of Health Stroke Scale score), vascular risk factors, and etiologic stroke subtypes (Causative Classification of Stroke) were recorded during index admission. Results: Utilizing the data from 912 acute ischemic stroke (AIS) patients (65+/-15 years of age, 58% male, 57% history of smoking, and 65% hypertensive) in a generalized linear model, brain volume (per 155.1cm^3 ) was associated with age (beta -0.3 (per 14.4 years)), male sex (beta 1.0) and prior stroke (beta -0.2). In the multivariable outcome model, brain volume was an independent predictor of mRS (beta -0.233), with reduced odds of worse long-term functional outcomes (OR: 0.8, 95% CI 0.7-0.9) in those with larger brain volumes. Conclusions: Larger brain volume quantified on clinical MRI of AIS patients at time of stroke purports a protective mechanism. The role of brain volume as a prognostic, protective biomarker has the potential to forge new areas of research and advance current knowledge of mechanisms of post-stroke recovery.

q-bio.NC↗

arXiv2020

Fast Learning-based Registration of Sparse 3D Clinical Images

We introduce SparseVM, a method that registers clinical-quality 3D MR scans both faster and more accurately than previously possible. Deformable alignment, or registration, of clinical scans is a fundamental task for many clinical neuroscience studies. However, most registration algorithms are designed for high-resolution research-quality scans. In contrast to research-quality scans, clinical scans are often sparse, missing up to 86% of the slices available in research-quality scans. Existing methods for registering these sparse images are either inaccurate or extremely slow. We present a learning-based registration method, SparseVM, that is more accurate and orders of magnitude faster than the most accurate clinical registration methods. To our knowledge, it is the first method to use deep learning specifically tailored to registering clinical images. We demonstrate our method on a clinically-acquired MRI dataset of stroke patients and on a simulated sparse MRI dataset. Our code is available as part of the VoxelMorph package at http://voxelmorph.mit.edu/.

cs.CV↗

arXiv2019

Multi-atlas image registration of clinical data with automated quality assessment using ventricle segmentation

Registration is a core component of many imaging pipelines. In case of clinical scans, with lower resolution and sometimes substantial motion artifacts, registration can produce poor results. Visual assessment of registration quality in large clinical datasets is inefficient. In this work, we propose to automatically assess the quality of registration to an atlas in clinical FLAIR MRI scans of the brain. The method consists of automatically segmenting the ventricles of a given scan using a neural network, and comparing the segmentation to the atlas' ventricles propagated to image space. We used the proposed method to improve clinical image registration to a general atlas by computing multiple registrations and then selecting the registration that yielded the highest ventricle overlap. Methods were evaluated in a single-site dataset of more than 1000 scans, as well as a multi-center dataset comprising 142 clinical scans from 12 sites. The automated ventricle segmentation reached a Dice coefficient with manual annotations of 0.89 in the single-site dataset, and 0.83 in the multi-center dataset. Registration via age-specific atlases could improve ventricle overlap compared to a direct registration to the general atlas (Dice similarity coefficient increase up to 0.15). Experiments also showed that selecting scans with the registration quality assessment method could improve the quality of average maps of white matter hyperintensity burden, instead of using all scans for the computation of the white matter hyperintensity map. In this work, we demonstrated the utility of an automated tool for assessing image registration quality in clinical scans. This image quality assessment step could ultimately assist in the translation of automated neuroimaging pipelines to the clinic.

eess.IV↗

arXiv2019

Patient-specific Conditional Joint Models of Shape, Image Features and Clinical Indicators

We propose and demonstrate a joint model of anatomical shapes, image features and clinical indicators for statistical shape modeling and medical image analysis. The key idea is to employ a copula model to separate the joint dependency structure from the marginal distributions of variables of interest. This separation provides flexibility on the assumptions made during the modeling process. The proposed method can handle binary, discrete, ordinal and continuous variables. We demonstrate a simple and efficient way to include binary, discrete and ordinal variables into the modeling. We build Bayesian conditional models based on observed partial clinical indicators, features or shape based on Gaussian processes capturing the dependency structure. We apply the proposed method on a stroke dataset to jointly model the shape of the lateral ventricles, the spatial distribution of the white matter hyperintensity associated with periventricular white matter disease, and clinical indicators. The proposed method yields interpretable joint models for data exploration and patient-specific statistical shape models for medical image analysis.

eess.IV↗

arXiv2019

Network Structural Dependency in the Human Connectome Across the Life-Span

Principles of network topology have been widely studied in the human connectome. Of particular interest is the modularity of the human brain, where the connectome is divided into subnetworks and subsequently changes with development, aging or disease are investigated. We present a weighted network measure, the Network Dependency Index (NDI), to identify an individual region's importance to the global functioning of the network. Importantly, we utilize NDI to differentiate four subnetworks (Tiers) in the human connectome following Gaussian Mixture Model fitting. We analyze the topological aspects of each subnetwork with respect to age and compare it to rich-club based subnetworks (rich-club, feeder and seeder). Our results first demonstrate the efficacy of NDI to identify more consistent, central nodes of the connectome across age-groups, when compared to the rich-club framework. Stratifying the connectome by NDI led to consistent subnetworks across the life-span revealing distinct patterns associated with age where, e.g., the key relay nuclei and cortical regions are contained in a subnetwork with highest NDI. The divisions of the human connectome derived from our data-driven NDI framework have the potential to reveal topological alterations described by network measures through the life-span.

q-bio.NC↗

arXiv2018

Medical Image Imputation from Image Collections

We present an algorithm for creating high resolution anatomically plausible images consistent with acquired clinical brain MRI scans with large inter-slice spacing. Although large data sets of clinical images contain a wealth of information, time constraints during acquisition result in sparse scans that fail to capture much of the anatomy. These characteristics often render computational analysis impractical as many image analysis algorithms tend to fail when applied to such images. Highly specialized algorithms that explicitly handle sparse slice spacing do not generalize well across problem domains. In contrast, we aim to enable application of existing algorithms that were originally developed for high resolution research scans to significantly undersampled scans. We introduce a generative model that captures fine-scale anatomical structure across subjects in clinical image collections and derive an algorithm for filling in the missing data in scans with large inter-slice spacing. Our experimental results demonstrate that the resulting method outperforms state-of-the-art upsampling super-resolution techniques, and promises to facilitate subsequent analysis not previously possible with scans of this quality. Our implementation is freely available at https://github.com/adalca/papago .

cs.CV↗