MRI texture-based radiomics analysis for the identification of altered functional networks in alcoholic patients and animal models.

Alcohol use disorder (AUD) is a complex condition representing a leading risk factor for death, disease and disability. Its high prevalence and severe health consequences make necessary a better understanding of the brain network alterations to improve diagnosis and treatment. The purpose of this study was to evaluate the potential of resting-state fMRI 3D texture features as a novel source of biomarkers to identify AUD brain network alterations following a radiomics approach. A longitudinal study was conducted in Marchigian Sardinian alcohol-preferring msP rats (N = 36) who underwent resting-state functional and structural MRI before and after 30 days of alcohol or water consumption. A cross-sectional human study was also conducted among 33 healthy controls and 35 AUD patients. The preprocessed functional data corresponding to control and alcohol conditions were used to perform a probabilistic independent component analysis, identifying seven independent components as resting-state networks. Forty-three radiomic features extracted from each network were compared using a Wilcoxon signed-rank test with Holm correction to identify the network most affected by alcohol consumption. Features extracted from this network were then used in the machine learning process, evaluating two feature selection methods and six predictive models within a nested cross-validation structure. The classification was evaluated by computing the area under the ROC curve. Images were quantized using different numbers of gray-levels to test their influence on the results. The influence of ageing, data preprocessing, and brain iron accumulation were also analyzed. The methodology was validated using structural scans. The striatal network in alcohol-exposed msP rats presented the most significant number of altered features. The radiomics approach supported this result achieving good classification performance in animals (AUC = 0.915 ± 0.100, with 12 features) and humans (AUC = 0.724 ± 0.117, with 9 features) using a random forest model. Using the structural scans, high accuracy was achieved with a multilayer perceptron in both species (animals: AUC > 0.95 with 2 features, humans: AUC > 0.82 with 18 features). The best results were obtained using a feature selection method based on the p-value. The proposed radiomics approach is able to identify AUD patients and alcohol-exposed rats with good accuracy, employing a subset of 3D features extracted from fMRI. Furthermore, it can help identify relevant networks in drug addiction.

Structural connectivity centrality changes mark the path toward Alzheimer's disease.

INTRODUCTION: The pathophysiological process of Alzheimer's disease is thought to begin years before clinical decline, with evidence suggesting prion-like spreading processes of neurofibrillary tangles and amyloid plaques. METHODS: Using diffusion magnetic resonance imaging data from the Alzheimer's Disease Neuroimaging Initiative database, we first identified relevant features for dementia diagnosis. We then created dynamic models with the Nathan Kline Institute-Rockland Sample database to estimate the earliest detectable stage associated with dementia in the simulated disease progression. RESULTS: A classifier based on centrality measures provides informative predictions. Strength and closeness centralities are the most discriminative features, which are associated with the medial temporal lobe and subcortical regions, together with posterior and occipital brain regions. Our model simulations suggest that changes associated with dementia begin to manifest structurally at early stages. DISCUSSION: Our analyses suggest that diffusion magnetic resonance imaging-based centrality measures can offer a tool for early disease detection before clinical dementia onset.

Evaluating Functional Connectivity Alterations in Autism Spectrum Disorder Using Network-Based Statistics.

The study of resting-state functional brain networks is a powerful tool to understand the neurological bases of a variety of disorders such as Autism Spectrum Disorder (ASD). In this work, we have studied the differences in functional brain connectivity between a group of 74 ASD subjects and a group of 82 typical-development (TD) subjects using functional magnetic resonance imaging (fMRI). We have used a network approach whereby the brain is divided into discrete regions or nodes that interact with each other through connections or edges. Functional brain networks were estimated using the Pearson's correlation coefficient and compared by means of the Network-Based Statistic (NBS) method. The obtained results reveal a combination of both overconnectivity and underconnectivity, with the presence of networks in which the connectivity levels differ significantly between ASD and TD groups. The alterations mainly affect the temporal and frontal lobe, as well as the limbic system, especially those regions related with social interaction and emotion management functions. These results are concordant with the clinical profile of the disorder and can contribute to the elucidation of its neurological basis, encouraging the development of new clinical approaches.

A fully automated method for segmentation and classification of local field potential recordings. Preliminary results.

Identification of brain states measured with electrophysiological methods such as electroencephalography and local field potential (LFP) recordings is of great importance in numerous neuroscientific applications. For instance, in Brain Computer Interface, in the diagnosis of neurological disorders as well as to investigate how brain rhythms stem from synchronized physiological mechanisms (e.g., memory and learning). In this work, we propose a fully automated method with the aim of partitioning LFP signals into stationary segments as well as classifying each detected segment into three different classes (delta, regular theta or irregular theta rhythms). Our approach is computationally efficient since the process of detection and partition of signals into stationary segments is only based on two features (the variance and the so-called spectral error measure) and allow the classification at the same time. We developed the algorithm upon analyzing six anesthetized rats, resulting in a true positive rate of 97.5%, 91.8% and 79.1% in detecting delta, irregular theta and regular theta rhythms, respectively. This preliminary quantitative evaluation offers encouraging results for further research.

Evaluating network brain connectivity in alcohol postdependent state using Network-Based Statistic.

The use of functional magnetic resonance imaging (fMRI) to measure spontaneous fluctuations in blood oxygen level dependent (BOLD) signals has become an indispensable tool to investigate how brain regions interact and form long-range networks. Statistical dependency measures between brain regions obtained from BOLD signals can inform about brain functional states in longitudinal studies of neurological and psychiatric disorders. Furthermore, its non-invasive nature allows comparable measurements in clinical and animal studies, providing excellent translational capabilities. In the present study, we apply Network-Based Statistic (NBS) to investigate alterations in the functional connectivity (FC) of the rat brain in a post-dependent (PD) state, an established animal model of clinical relevant features in alcoholism. In contrast to mass-univariate tests, in which comparisons are performed at single link-level, NBS enhances the statistical power by assuming that the connections comprising the effect of interest are interconnected. Brain-wide resting-state fMRI signals were collected in 14 controls and 13 PD rats, and Pearson correlations computed between 47 brain regions of interest (ROIs). The NBS analysis revealed statistically significant differences in a connected network of structures including hippocampus, amygdala, lateral hypothalamus and the raphe nucleus, all regions with known relevance for addictive behaviors. In contrast, no individual connection could be found significant by univariate comparisons with false discovery rate (FDR) correction. Correlations between the structures in the identified subnetwork tend to decrease or become negative (anti-correlated) in the PD state compared to controls. We interpret this result as evidence for a disconnected subnetwork in the PD state.

Brain functional connectivity alterations in a rat model of excessive alcohol drinking: A resting-state network analysis.

Alcohol use disorders (AUD) are a major public health concern. Understanding the brain network alterations is of the utmost importance to diagnose and develop treatment strategies. Employing resting-state functional magnetic resonance imaging, we have performed a longitudinal study in a rat model of chronic excessive alcohol consumption, to identify functional alterations in brain networks triggered by alcohol drinking. Two time points were considered: 1) before alcohol consumption (control condition) and 2) after 30 days of alcohol drinking (alcohol condition). We first identified nine resting-state networks with group independent component analysis. Afterwards, dual regression was applied to obtain subject-specific time courses and spatial maps. L2-regularized partial correlation analysis between pairs of networks showed that functional connectivity (FC) between the retrosplenial-visual and striatal networks decreases due to alcohol consumption, whereas FC between the prefrontal-cingulate and striatal networks increases. Analysis of subject-specific spatial maps revealed FC decreases within networks after alcohol drinking, including the striatal, motor-parietal, prefrontal-cingulate, retrosplenial-visual and left motor-parietal networks. Overall, our results unveil a generalized decrease in brain FC induced by alcohol drinking in genetically predisposed animals, even after a relatively short period of exposure (1 month). The only exception to this hypo-connectivity state is the functional association between the striatal and prefrontal-cingulate networks, which increases after drinking, supporting evidence in human alcoholics.

Structural and functional, empirical and modeled connectivity in the cerebral cortex of the rat.

Connectomics data from animal models provide an invaluable opportunity to reveal the complex interplay between structure and function in the mammalian brain. In this work, we investigate the relationship between structural and functional connectivity in the rat brain cortex using a directed anatomical network generated from a carefully curated meta-analysis of published tracing data, along with resting-state functional MRI data obtained from a group of 14 anesthetized Wistar rats. We found a high correspondence between the strength of functional connections, measured as blood oxygen level dependent (BOLD) signal correlations between cortical regions, and the weight of the corresponding anatomical links in the connectome graph (maximum Spearman rank-order correlation ρ=0.48). At the network-level, regions belonging to the same functionally defined community tend to form more mutual weighted connections between each other compared to regions located in different communities. We further found that functional communities in resting-state networks are enriched in densely connected anatomical motifs. Importantly, these higher-order structural subgraphs cannot be explained by lower-order topological properties, suggesting that dense structural patterns support functional associations in the resting brain. Simulations of brain-wide resting-state activity based on neural mass models implemented on the empirical rat anatomical connectome demonstrated high correlation between the simulated and the measured functional connectivity (maximum Pearson correlation ρ=0.53), further suggesting that the topology of structural connections plays an important role in shaping functional cortical networks.

Automatic segmentation of the spine by means of a probabilistic atlas with a special focus on ribs suppression.

PURPOSE: The development of automatic and reliable algorithms for the detection and segmentation of the vertebrae are of great importance prior to any diagnostic task. However, an important problem found to accurately segment the vertebrae is the presence of the ribs in the thoracic region. To overcome this problem, a probabilistic atlas of the spine has been developed dealing with the proximity of other structures, with a special focus on ribs suppression. METHODS: The data sets used consist of Computed Tomography images corresponding to 21 patients suffering from spinal metastases. Two methods have been combined to obtain the final result: firstly, an initial segmentation is performed using a fully automatic level-set method; secondly, to refine the initial segmentation, a 3D volume indicating the probability of each voxel of belonging to the spine has been developed. In this way, a probability map is generated and deformed to be adapted to each testing case. RESULTS: To validate the improvement obtained after applying the atlas, the Dice coefficient (DSC), the Hausdorff distance (HD), and the mean surface-to-surface distance (MSD) were used. The results showed up an average of 10 mm of improvement accuracy in terms of HD, obtaining an overall final average of 15.51 ± 2.74 mm. Also, a global value of 91.01 ± 3.18% in terms of DSC and a MSD of 0.66 ± 0.25 mm were obtained. The major improvement using the atlas was achieved in the thoracic region, as ribs were almost perfectly suppressed. CONCLUSION: The study demonstrated that the atlas is able to detect and appropriately eliminate the ribs while improving the segmentation accuracy.

Automatic segmentation of the spine by means of a probabilistic atlas with a special focus on ribs suppression. Preliminary results.

Spine is a structure commonly involved in several prevalent diseases. In clinical diagnosis, therapy, and surgical intervention, the identification and segmentation of the vertebral bodies are crucial steps. However, automatic and detailed segmentation of vertebrae is a challenging task, especially due to the proximity of the vertebrae to the corresponding ribs and other structures such as blood vessels. In this study, to overcome these problems, a probabilistic atlas of the spine, including cervical, thoracic and lumbar vertebrae has been built to introduce anatomical knowledge in the segmentation process, aiming to deal with overlapping gray levels and the proximity to other structures. From a set of 3D images manually segmented by a physician (training data), a 3D volume indicating the probability of each voxel of belonging to the spine has been developed, being necessary the generation of a probability map and its deformation to adapt to each patient. To validate the improvement of the segmentation using the atlas developed in the testing data, we computed the Hausdorff distance between the manually-segmented ground truth and an automatic segmentation and also between the ground truth and the automatic segmentation refined with the atlas. The results are promising, obtaining a higher improvement especially in the thoracic region, where the ribs can be found and appropriately eliminated.

A fully automated level-set based segmentation method of thoracic and lumbar vertebral bodies in Computed Tomography images.

Spine is a structure commonly involved in several diseases. Identification and segmentation of the vertebral structures are of relevance to many medical applications related to the spine such as diagnosis, therapy or surgical intervention. However, the development of automatic and reliable methods are an unmet need. This work presents a fully automatic segmentation method of thoracic and lumbar vertebral bodies from Computed Tomography images. The procedure can be divided into four main stages: firstly, seed points were detected in the spinal canal in order to generate initial contours in the segmentation process, automating the whole process. Secondly, a processing step is performed to improve image quality. Third step was to carry out the segmentation using the Selective Binary Gaussian Filtering Regularized Level Set method and, finally, two morphological operations were applied in order to refine the segmentation result. The method was tested in clinical data coming from 10 trauma patients. To evaluate the result the average value of the DICE coefficient was calculated, obtaining a 90.86 ± 1.87% in the whole spine (thoracic and lumbar regions), a 86.08 ± 1.73% in the thoracic region and a 95,61 ±2,25% in the lumbar region. The results are highly competitive when compared to the results obtained in previous methods, especially for the lumbar region.

A fully automated method for spinal canal detection in computed tomography images.

This work presents a new automated method for spinal canal detection in Computed Tomography (CT) images. It uses both 2D and 3D information and the algorithm extracts the spinal canal automatically. The procedure can be divided into three main steps. Firstly, a thresholding and a set of morphological operations were applied. Secondly, 3D connectivity analysis was defined to extract the objects forming part of the spinal canal. Finally, the centroid of each slice constituting the spinal canal object was computed. Furthermore, interpolation and extrapolation of data were performed, if required. The method was applied on two different groups, each one coming from different acquisition systems. A total of 25 patients and 8704 images were used. An experienced radiologist evaluated the method qualitatively supporting the utility of it, as all extracted points fell into the spinal canal. Therefore, our method was able to reduce the workload and detect spinal canal objectively. We expect to carry out a quantitative evaluation in our future research. The qualitative outcome of this work suggests promising results.

Fully automatic spinal canal segmentation for radiation therapy using a gradient vector flow-based method on computed tomography images: A preliminary study.

Nowadays, radiotherapy is one of the key techniques for localized cancer treatment. Accurate identification of target volume (TV) and organs at risk (OAR) is a crucial step to therapy success. Spinal cord is one of the most radiosensitive OAR and its localization tends to be an observer-dependent and time-consuming task. Hence, numerous studies have aimed to carry out the contouring automatically. In CT images, there is a lack of contrast between soft tissues, making more challenge the delineation. That is the reason why the majority of researches have focused on spinal canal segmentation rather than spinal cord. In this work, we propose a fully automated method for spinal canal segmentation using a Gradient Vector Flow-based (GVF) algorithm. An experienced radiologist performed the manual segmentation, generating the ground truth. The method was evaluated on three different patients using the Dice coefficient, obtaining the following results: 79.50%, 83.77%, and 81.88%, respectively. Outcome reveals that more research has to be performed to improve the accuracy of the method.