Publications by Year: 2019

In the last five years, deep learning (DL) has become the state-of-the-art tool for solving various tasks in medical image analysis. Among the different methods that have been proposed to improve the performance of Convolutional Neural Networks (CNNs), one typical approach is the augmentation of the training data set through various transformations of the input image. Data augmentation is typically used in cases where a small amount of data is available, such as the majority of medical imaging problems, to present a more substantial amount of data to the network and improve the overall accuracy. However, the ability of the network to improve the accuracy of the results when a slightly modified version of the same input is presented is often overestimated. This overestimation is the result of the strong correlation between data samples when they are considered independently in the training phase. In this paper, we emphasize the importance of optimizing for accuracy as well as precision among multiple replicates of the same training data in the context of data augmentation. To this end, we propose a new approach that leverages the augmented data to help the network focus on the precision through a specifically-designed loss function, with the ultimate goal to improve both the overall performance and the network’s precision at the same time. We present two different applications of DL (regression and segmentation) to demonstrate the strength of the proposed strategy. We think that this work will pave the way to a explicit use of data augmentation within the loss function that helps the network to be invariant to small variations of the same input samples, a characteristic that is always required to every application in the medical imaging field.

Mastering detailed anatomy of the human deep brain in clinical neurosciences is challenging. Although numerous pioneering works have gathered a large dataset of structural and topographic information, it is still difficult to transfer this knowledge into practice, even with advanced magnetic resonance imaging techniques. Thus, classical histological atlases continue to be used to identify structures for stereotactic targeting in functional neurosurgery. Physicians mainly use these atlases as a template co-registered with the patient’s brain. However, it is possible to directly identify stereotactic targets on MRI scans, enabling personalized targeting. In order to help clinicians directly identify deep brain structures relevant to present and future medical applications, we built a volumetric MRI atlas of the deep brain (MDBA) on a large scale (infra millimetric). Twelve hypothalamic, 39 subthalamic, 36 telencephalic, and 32 thalamic structures were identified, contoured, and labeled. Nineteen coronal, 18 axial, and 15 sagittal MRI plates were created. Although primarily designed for direct labeling, the anatomic space was also subdivided in twelfths of AC-PC distance, leading to proportional scaling in the coronal, axial, and sagittal planes. This extensive work is now available to clinicians and neuroscientists, offering another representation of the human deep brain ([https://hal.archives-ouvertes.fr/] [hal-02116633]). The atlas may also be used by computer scientists who are interested in deciphering the topography of this complex region.

Cerebral microbleeds (CMBs), a common manifestation of mild traumatic brain injury (mTBI), have been sporadically implicated in the neurocognitive deficits of mTBI victims but their clinical significance has not been established adequately. Here we investigate the longitudinal effects of post-mTBI CMBs upon the fractional anisotropy (FA) of white matter (WM) in 21 older mTBI patients across the first  6 months post-injury. CMBs were segmented automatically from susceptibility-weighted imaging (SWI) by leveraging the intensity gradient properties of SWI to identify CMB-related hypointensities using gradient-based edge detection. A detailed diffusion magnetic resonance imaging (dMRI) atlas of WM was used to segment and cluster tractography streamlines whose prototypes were then identified. The correlation coefficient was calculated between (A) FA values at vertices along streamline prototypes and (B) topological (along-streamline) distances between these vertices and the nearest CMB. Across subjects, the CMB identification approach achieved a sensitivity of 97.1% ± 4.7% and a precision of 72.4% ± 11.0% across subjects. The correlation coefficient was found to be negative and, additionally, statistically significant for 12.3% ± 3.5% of WM clusters (p

Radiomics is an emerging field which extracts quantitative radiology data from medical images and explores their correlation with clinical outcomes in a non-invasive manner. This review aims to assess whether radiomics is a useful and reproducible method for clinical management of hepatocellular carcinoma (HCC) by reviewing the strengths and weaknesses of current radiomics literature pertaining specifically to HCC. From an initial set of 48 articles recovered through database searches, 23 articles were retained to be included in this review after full screening. Among these 23 studies, 7 used a radiomics approach in magnetic resonance imaging (MRI). Only two studies applied radiomics to positron emission tomography-computed tomography (PET-CT). In the remaining 14 articles, a radiomics analysis was performed on computed tomography (CT). Eight studies dealt with the relationship between biological signatures and imaging findings, and can be classified as radiogenomic studies. For each study included in our review, we computed a Radiomics Quality Score (RQS) as proposed by Lambin et al. We found that the RQS (mean ± standard deviation) was 8.35 ± 5.38 (out of a possible maximum value of 36). Although these scores are fairly low, and radiomics has not yet reached clinical utility in HCC, it is important to underscore the fact that these early studies pave the way for the radiomics field with a focus on HCC. Radiomics is still a very young field, and is far from being mature, but it remains a very promising technology for the future for developing adequate personalized treatment as a non-invasive approach, for complementing or replacing tumor biopsies, as well as for developing novel prognostic biomarkers in HCC patients.

The present survey describes the state-of-the-art techniques for dynamic cardiac magnetic resonance image reconstruction. Additionally, clinical relevance, main challenges, and future trends of this image modality are outlined. Thus, this paper aims to provide a general vision about cine MRI as the standard procedure in functional evaluation of the heart, focusing on technical methodologies.

Background: Factors of increased prevalence among individuals with Black racial identity (e.g., cardiovascular disease, CVD) may influence the association between exposure to repetitive head impacts (RHI) from American football and later-life neurological outcomes. Here, we tested the interaction between racial identity and RHI on neurobehavioral outcomes, brain volumetric measures, and cerebrospinal fluid (CSF) total tau (t-tau), phosphorylated tau (p-tau), and Aβ in symptomatic former National Football League (NFL) players.

Biomarker inference from biomedical images is one of the main tasks of medical image analysis. Standard techniques follow a segmentation-and-measure strategy, where the structure is first segmented and then the measurement is performed. Recent work has shown that such strategy could be replaced by a direct regression of the biomarker value in using regression networks. While achieving high correlation coefficients, such techniques operate as a ’black-box’, not offering quality-control images. We present a methodology to regress the biomarker from the image while simultaneously computing the quality control image. Our proposed methodology does not require segmentation masks for training, but infers the segmentations directly from the pixels that used to compute the biomarker value. The network proposed consists of two steps: a segmentation method to an unknown reference and a summation method for the biomarker estimation. The network is optimized using a dual loss function, L2 for the biomarkers and an L1 to enforce sparsity. We showcase our methodology in the problem of pectoralis muscle area (PMA) and subcutaneous fat area (SFA) inference in a single slice from chest-CT images. We use a database of 7000 cases to which only the value of the biomarker is known for training and a test set of 3000 cases with both, biomarkers and segmentations. We achieve a correlation coefficient of 0.97 for PMA and 0.98 for SFA with respect to the reference standard. The average DICE coefficient is of 0.88 (PMA) and 0.89 (SFA). Comparing with standard segment-and-measure techniques, we achieve the same correlation for the biomarkers but smaller DICE coefficients in segmentation. Such is of little surprise, since segmentation networks are the upper limit of performance achievable, and we are not using segmentation masks for training. We can conclude that it is possible to infer segmentation masks from biomarker regression networks.

We present a deep learning tractography segmentation method that allows fast and consistent white matter fiber tract identification across healthy and disease populations and across multiple diffusion MRI (dMRI) acquisitions. We create a large-scale training tractography dataset of 1 million labeled fiber samples (54 anatomical tracts are included). To discriminate between fibers from different tracts, we propose a novel 2D multi-channel feature descriptor (FiberMap) that encodes spatial coordinates of points along each fiber. We learn a CNN tract classification model based on FiberMap and obtain a high tract classification accuracy of 90.99%. The method is evaluated on a test dataset of 374 dMRI scans from three independently acquired populations across health conditions (healthy control, neuropsychiatric disorders, and brain tumor patients). We perform comparisons with two state-of-the-art white matter tract segmentation methods. Experimental results show that our method obtains a highly consistent segmentation result, where over 99% of the fiber tracts are successfully detected across all subjects under study, most importantly, including patients with space occupying brain tumors. The proposed method leverages deep learning techniques and provides a much faster and more efficient tool for large data analysis than methods using traditional machine learning techniques.

Gadolinium-based contrast agents (GBCAs) increase lesion detection and improve disease characterization for many cerebral pathologies investigated with MRI. These agents, introduced in the late 1980s, are in wide use today. However, some non-ionic linear GBCAs have been associated with the development of nephrogenic systemic fibrosis in patients with kidney failure. Gadolinium deposition has also been found in deep brain structures, although it is of unclear clinical relevance. Hence, new guidelines from the International Society for Magnetic Resonance in Medicine advocate cautious use of GBCA in clinical and research practice. Some linear GBCAs were restricted from use by the European Medicines Agency (EMA) in 2017.This review focuses on non-contrast-enhanced MRI techniques that can serve as alternatives for the use of GBCAs. Clinical studies on the diagnostic performance of non-contrast-enhanced as well as contrast-enhanced MRI methods, both well established and newly proposed, were included. Advantages and disadvantages together with the diagnostic performance of each method are detailed. Non-contrast-enhanced MRIs discussed in this review are arterial spin labeling (ASL), time of flight (TOF), phase contrast (PC), diffusion-weighted imaging (DWI), magnetic resonance spectroscopy (MRS), susceptibility weighted imaging (SWI), and amide proton transfer (APT) imaging.Ten common diseases were identified for which studies reported comparisons of non-contrast-enhanced and contrast-enhanced MRI. These specific diseases include primary brain tumors, metastases, abscess, multiple sclerosis, and vascular conditions such as aneurysm, arteriovenous malformation, arteriovenous fistula, intracranial carotid artery occlusive disease, hemorrhagic, and ischemic stroke.In general, non-contrast-enhanced techniques showed comparable diagnostic performance to contrast-enhanced MRI for specific diagnostic questions. However, some diagnoses still require contrast-enhanced imaging for a complete examination.