Search

BACKGROUND: Many subjects with major depression (MDD) exhibit subthreshold mania symptoms (MDD+). This study investigated, for the first time, using emotional inhibition tasks, whether the neural organization of MDD+ subjects is more similar to bipolar depression (BDD) or to MDD subjects without subthreshold bipolar symptoms (MDD−). METHOD: This study included 118 medication-free young adults (15 – 30 yrs.): 20 BDD, 28 MDD+, 41 MDD− and 29 HC subjects. Participants underwent fMRI during emotional and non-emotional Go/No-go tasks during which they responded for Go stimuli and inhibited response for happy, fear, and non-emotional (gender) faces No-go stimuli. Univariate linear mixed-effects (LME) analysis for group effects and multivariate Gaussian Process Classifier (GPC) analyses were conducted. RESULTS: MDD− group compared to both the BDD and MDD+ groups, exhibited significantly lower activation in parietal, temporal and frontal regions (cluster-wise corrected p

Differentiation between Bipolar Disorder Depression (BDD) and Unipolar Major Depressive Disorder (MDD) is critical to clinical practice. This study investigated machine learning classification of BDD and MDD using graph properties of Diffusion-weighted Imaging (DWI)-based structural connectome.This study included a large number of medication-free (N =229) subjects: 60 BDD, 95 MDD, and 74 Healthy Control (HC) subjects. DWI probabilistic tractography was performed to create Fractional Anisotropy (FA) and Total Streamline (TS)-based structural connectivity matrices. Global and nodal graph properties were computed from these matrices and tested for group differences. Next, using identified graph properties, machine learning classification (MLC) between BDD, MDD, MDD with risk factors for developing BD (MDD+), and MDD without risk factors for developing BD (MDD-) was conducted.Communicability Efficiency of the left superior frontal gyrus (SFG) was significantly higher in BDD vs. MDD. In particular, Communicability Efficiency using TS-based connectivity in the left SFG as well as FA-based connectivity in the right middle anterior cingulate area was higher in the BDD vs. MDD- group. There were no significant differences in graph properties between BDD and MDD+. Direct comparison between MDD+ and MDD- showed differences in Eigenvector Centrality (TS-based connectivity) of the left middle frontal sulcus. Acceptable Area Under Curve (AUC) for classification were seen between the BDD and MDD- groups, and between the MDD+ and MDD- groups, using the differing graph properties.Graph properties of DWI-based connectivity can discriminate between BDD and MDD subjects without risk factors for BD.

Major Depressive Disorder (MDD) is frequently associated with risk factors for the development of Bipolar Disorder (BD). Using graph theory, we investigated brain network properties associated with BD risk factors in young MDD subjects.Resting-state fMRI was acquired from a large cohort (N= 104) of medication-free currently depressed participants (25 BD depression (BDD), 79 MDD). Lifetime mania symptom count (LMSC), current Young Mania Rating Scale (YMRS) score, and family history of mood disorders (FHMD) were examined as BD risk factors. Functional connectivity matrices from 280 regions of interests (ROIs) were first entered into the Network Based Statistic (NBS) toolbox to identify connections that varied with each risk factor. Next, within the correlated network for each risk factor, global and nodal graph properties for the top five linked nodes were calculated. Last, using identified graph properties, machine learning classification (MLC) between BDD, MDD with BD risk factors (MDD+), and without BD risk factors (MDD-) was conducted.LMSC positively correlated with left lateral orbitofrontal cortex (LOFC) Communication Efficiency and with left middle temporal Eigenvector Centrality. Current YMRS score positively correlated with right amygdala Communication Efficiency and Closeness Centrality. FHMD positively correlated with right insula Eigenvector Centrality. Acceptable MLC accuracy was seen between BDD and MDD- using middle temporal Eigenvector Centrality, whereas moderate accuracy was seen between MDD+ and MDD- using OFC Communication Efficiency.Although participants were medication-free, they were not medication-naïve.Functional connectome graph properties may serve as BD vulnerability biomarkers in young individuals with MDD.

Individuals with or at risk for bipolar disorder (BD) often present initially for the treatment of depressive symptoms. Unfortunately, pharmacological treatments for major depressive disorder (MDD) can be iatrogenic, precipitating mania that may not have otherwise occurred. Current diagnostic procedures rely solely on self-reported/observable symptoms, and thus alternative data sources, such as brain network properties, are needed to supplement current self-report/observation-based indices of risk for mania.Brain connectivity during affect maintenance/regulation was examined in a large (N = 249), medication-free sample of currently depressed patients with BD (n = 50) and MDD (n = 116) and healthy control subjects (n = 83). BD risk was categorized in a subset of patients with MDD. We used graph theory to identify emergent network properties that differentiated between patients with BD and MDD and between patients with MDD at high and low risk for BD.BD and MDD differed in the extent to which the rostral anterior cingulate cortex was embedded in the local network, amount of influence the hippocampus exerted over global network communication, and clarity of orbitofrontal cortex communication. Patients with MDD at high risk for BD showed a pattern of local network clustering around the right amygdala that was similar to the pattern observed in healthy control subjects, whereas patients with MDD at low risk for BD deviated from this pattern.BD and MDD differed in emergent network mechanisms subserving affect regulation, and amygdala properties tracked BD risk in patients with MDD. If replicated, our findings may be combined with other markers to assess the presence of BD and/or BD risk in individuals presenting with depressive symptoms to prevent the use of iatrogenic treatments.

BackgroundA significant number of subjects with major depression (MDD) exhibit subthreshold mania symptoms (MDD+). This study investigated, for the first time, using emotional inhibition tasks, whether the neurobiology of MDD+ subjects is more akin to bipolar disorder depression (BDD) or to MDD subjects without any subthreshold bipolar symptoms (MDD−).MethodThis study included 118 medication-free young adult subjects (16 – 30 yrs.): 20 BDD, 28 MDD+, 41 MDD−, and 29 HC subjects. Participants underwent fMRI during emotional and non-emotional Go/No-go tasks during which they responded for Go stimuli and inhibited response for happy, fear, emotional (happy + fear) and non-emotional (gender) faces No-go stimuli. Linear mixed effects (LME) analysis for group effects and Gaussian Process Classifier (GPC) analyses was conducted.ResultsMDD− group compared to both the BDD and MDD+ groups, exhibited significantly lower activation in parietal, temporal and frontal regions (cluster-wise corrected p ConclusionUsing an fMRI emotional Go-Nogo task, MDD− subjects can be discriminated from BDD subjects and MDD+.

Lithium is a highly effective medication for bipolar disorder, but its mechanism of action remains unknown. In this study, brain MRI scans and blood samples for gene expression (total of 110 scans and 109 blood samples) were collected from 21 bipolar subjects before and after 2 and 8 weeks of lithium monotherapy and at the same time-points from untreated 16 healthy controls. We used linear mixed-effects models to identify brain structural features and genes with expression changed after lithium treatment, with correction for multiple testing, and correlated their concurrent changes to identify molecular pathways associated with lithium effects. There are significant increases in gray matter fraction, global cortical thickness, and the frontal and parietal cortices after 8 weeks of lithium treatment (corrected p p p

Radiation treatment of patients with lung cancer can benefit from knowledge of regional lung function. 4-D CT provides structural and anatomical information during tidal breathing which maybe used to measure mechanical properties related to deformation of the lung tissue including Jacobian and strain of deformation at high resolutions and as a function of time. Single photon emission computed tomography (SPECT) lung imaging also provides functional information using radioactive tracers with SPECT ventilation and SPECT perfusion imaging. Due to availability and ease of use, we performed B-spline deformable registration — this yielded 3D deformation fields between successive respiratory phases of the 4D CT data from which measures of regional lung function in 11 lung cancer patients were derived. Furthermore, we registered the max-inhale CT volume to both V SPECT and Q SPECT data sets using the Demon's non-rigid registration algorithm in all subjects. Subsequently, statistical correlation between CT ventilation images, Jacobian and strain values, with both V SPECT and Q SPECT was undertaken. Through statistical analysis with the Spearman's rank correlation coefficient (ρ), we found that Jacobian values have the highest correlation with both V SPECT and Q SPECT . Among strain values, λ 3 showed higher correlation with SPECT than the other measures.

Segmentation of the aorta from CT and MR data is important in order to quantitatively assess diseases of the aorta including aortic dissection and distention of aortic aneurysm, among others. In this paper, we propose a segmentation method to extract exact the 3D boundary of the aorta via graph-cuts segmentation. The graph-cuts technique is able to avoid local minima with global optimization and can be applied to 3D and higher dimension with fast computation. We performed 3D segmentation using this method for five CT data sets. The user selects seed points for aorta region as 'object' and surrounding tissues as 'background' on an axial slice of the 3D CT data and the algorithm calculates the cost of n-link (neighborhood-link) and t-link (terminal-link), and computes the minimum cut separating the aorta from the background by applying the max-flow/min-cut algorithm. Results were validated against manually traced aorta boundaries. The mean Dice Similarity Coefficient for the five 3D segmentations was 0.9381. The 3D segmentation took less than five minutes for data sets of size 512×512×244 to 512×512×284.

The conventional graph cuts technique has been widely used for image segmentation due to its ability to find the global minimum and its ease of implementation. However, it is an intensity-based technique and as a result is limited to segmentation applications where there is significant contrast between the object and the background. We modified the conventional graph cuts method by adding shape prior and motion information. Active shape models (ASM) with signed distance functions were used to capture the shape prior information, preventing unwanted surrounding tissue from becoming part of the segmented object. The optical flow method was used to estimate the local motion and to extend 3D segmentation to 4D by warping a prior shape model through time. The method has been applied to segmentation of whole lung boundary and whole liver boundary from respiratory gated CT data. 4D lung boundary segmentation in five patients, and 4D liver boundary segmentation in five patients were performed and in each case, results were compared with the results from expert-delineated ground truth. 4D segmentation for five phases of CT data took approximately ten minutes on a PC workstation with AMD Phenom II and 32GB of memory. An important by-product is quantitative whole organ volumes from respiratory gated CT from end-inspiration to end-expiration which can be determined with high accuracy.

Purpose Inaccuracy of conventional four-dimensional (4D) flow MR imaging in the presence of random unsteady and turbulent blood flow distal to a narrowing has been an important challenge. Previous investigations have revealed that shorter echo times (TE) decrease the errors, leading to more accurate flow assessments. Methods In this study, as part of a 4D flow acquisition, an Ultra-Short TE (UTE) method was adopted. UTE works based on a center-out radial k-space trajectory that inherently has a short TE. By employing free induction decay sampling starting from read-out gradient ramp-up, and by combining the refocusing lobe of the slice select gradient with the bipolar flow encoding gradient, TEs of ≈1 msec may be achieved. Results Both steady and pulsatile flow regimes, and in each case a range of Reynolds numbers, were studied in an in-vitro model. Flow assessment at low and medium flow rates demonstrated a good agreement between 4D UTE and conventional 4D flow techniques. However, 4D UTE flow significantly outperformed conventional 4D flow, at high flow rates for both steady and pulsatile flow regimes. Feasibility of the method in one patient with Aortic Stenosis was also demonstrated. Conclusion For both steady and pulsatile high flow rates, the measured flow distal to the stenotic narrowing using conventional 4D flow revealed more than 20% error compared to the ground-truth flow. This error was reduced to less than 5% using the 4D UTE flow technique. Magn Reson Med 73:939–950, 2015. © 2014 Wiley Periodicals, Inc.

Lung tumor segmentation is important for therapy in the radiation treatment of patients with thoracic malignancies. In this paper, we describe a 4D image segmentation method based on graph-cuts optimization, shape prior and optical flow. Due to small size, the location, and low contrast between the tumor and the surrounding tissue, tumor segmentation in 3D+t is challenging. We performed 4D lung tumor segmentation in 5 patients, and in each case compared the results with the expert-delineated lung nodules. In each case, 4D image segmentation took approximately ten minutes on a PC with AMD Phenom II and 32GB of memory for segmenting tumor in five phases of lung CT data.

Lung boundary image segmentation is important for many tasks including for example in development of radiation treatment plans for subjects with thoracic malignancies. In this paper, we describe a method and parameter settings for accurate 3D lung boundary segmentation based on graph-cuts from X-ray CT data1. Even though previously several researchers have used graph-cuts for image segmentation, to date, no systematic studies have been performed regarding the range of parameter that give accurate results. The energy function in the graph-cuts algorithm requires 3 suitable parameter settings: K, a large constant for assigning seed points, c, the similarity coefficient for n-links, and λ, the terminal coefficient for t-links. We analyzed the parameter sensitivity with four lung data sets from subjects with lung cancer using error metrics. Large values of K created artifacts on segmented images, and relatively much larger value of c than the value of λ influenced the balance between the boundary term and the data term in the energy function, leading to unacceptable segmentation results. For a range of parameter settings, we performed 3D image segmentation, and in each case compared the results with the expert-delineated lung boundaries. We used simple 6-neighborhood systems for n-link in 3D. The 3D image segmentation took 10 minutes for a 512x512x118 ~ 512x512x190 lung CT image volume. Our results indicate that the graph-cuts algorithm was more sensitive to the K and λ parameter settings than to the C parameter and furthermore that amongst the range of parameters tested, K=5 and λ=0.5 yielded good results.

Use of phase-contrast (PC) MRI in assessment of hemodynamics has significant clinical importance. In this paper we develop a novel approach to determination of hemodynamic pressures. 3D gradients of pressure obtained from Navier-Stokes equation are expanded into a series of orthogonal basis functions, and are subsequently projected onto an integrable subspace. Before the projection step however, a scheme is devised to eliminate the discontinuity at the vessel and image boundaries. In terms of the computation time, the proposed approach significantly improves on previous iterative methods for pressure calculations. The method has been validated using computational fluid dynamic simulations and in-vitro MRI studies of stenotic flows.

Assessment of blood flow is an important factor in diagnosis of cardiovascular disease. Vascular stenosis result in disturbed blood flow, flow recirculation, turbulence, and flow jet. These types of flows cause erroneous quantification of blood flow using conventional Phase contrast (PC) MRI techniques. Previous investigations have revealed that shorter Echo Times (TE) can decrease the quantification errors. In this paper, we performed phantom studies under steady flow to validate the UTE technique. Investigation of three different constant flow rates revealed a significant improvement in flow quantification and reduction of flow artifacts in comparison to Cartesian Phase-Contrast MRI.

Phase-contrast MRI (PC MRI) is a powerful technique for imaging the flow velocities in-vivo. In PC MRI, the velocity encoding parameter, V enc , plays a central role in fidelity of velocities acquired with this technique. Typically V enc is chosen to be the highest velocity in the flow. However, when acquiring velocity images at a number of slice locations, it may be advantageous to change the V enc as a function of position. The goal of this study was to determine the effect of V enc on the accuracy of the estimated velocity profiles in a range of locations in a phantom model of a 90% area stenosis. To this end, velocities from PC MRI were compared with velocities from computational fluid dynamics (CFD) simulations for identical geometry and flow conditions. Our results confirm that although using smaller V enc 's leads to wrapping artifact, it provides more accurate velocity profiles. However, there is a lower limit to setting the V enc , beyond which phase-unwrapping becomes uncertain.

Purpose: To investigate whether pre-treatment images can be used in predicting microsphere distribution in tumors. When intra-arterial radioembolization using Y90 microspheres was performed, the microspheres were often delivered non-uniformly within the tumor, which could lead to an inefficient therapy. Therefore, it is important to estimate the distribution of microspheres. Methods: Early arterial phase CT and FDG PET images were acquired for patients with primary liver cancer prior to radioembolization (RE) using Y90 microspheres. Tumor volume was delineated on CT images and fused with FDG PET images. From each voxel (3.9×3.9×3.3 mm3) in the tumor, the Hounsfield unit (HU) from the CT and SUV values from the FDG PET were harvested. We binned both HU and SUV into 11 bins and then calculated a normalized joint-histogram in an 11×11 array.Patients also underwent a post-treatment Y90 PET imaging. Radiation dose for the tumor was estimated using convolution of the Y90 distribution with a dose-point kernel. We also calculated a fraction of the tumor volume that received a radiation dose great than 100Gy. Results: Averaged over 40 patients, 55% of tumor volume received a dose greater than 100Gy (range : 1.1 – 100%). The width of the joint histogram was narrower for patients with a high dose. For patients with a low dose, the width was wider and a larger fraction of tumor volume had low HU. Conclusion: We have shown the pattern of joint histogram of the HU and SUV depends on delivered dose. The patterns can predict the efficacy of uniform intra-arterial delivery of Y90 microspheres.