Search

Brain extraction is a critical pre-processing step in brain magnetic resonance imaging (MRI) analytical pipelines. In rodents, this is often achieved by manually editing brain masks slice-by-slice, a time-consuming task where workloads increase with higher spatial resolution datasets. We recently demonstrated successful automatic brain extraction via a deep-learning-based framework, U-Net, using 2D convolutions. However, such an approach cannot make use of the rich 3D spatial-context information from volumetric MRI data. In this study, we advanced our previously proposed U-Net architecture by replacing all 2D operations with their 3D counterparts and created a 3D U-Net framework. We trained and validated our model using a recently released CAMRI rat brain database acquired at isotropic spatial resolution, including T2-weighted turbo-spin-echo structural MRI and T2*-weighted echo-planar-imaging functional MRI. The performance of our 3D U-Net model was compared with existing rodent brain extraction tools, including Rapid Automatic Tissue Segmentation, Pulse-Coupled Neural Network, SHape descriptor selected External Regions after Morphologically filtering, and our previously proposed 2D U-Net model. 3D U-Net demonstrated superior performance in Dice, Jaccard, center-of-mass distance, Hausdorff distance, and sensitivity. Additionally, we demonstrated the reliability of 3D U-Net under various noise levels, evaluated the optimal training sample sizes, and disseminated all source codes publicly, with a hope that this approach will benefit rodent MRI research community.Significant Methodological Contribution: We proposed a deep-learning-based framework to automatically identify the rodent brain boundaries in MRI. With a fully 3D convolutional network model, 3D U-Net, our proposed method demonstrated improved performance compared to current automatic brain extraction methods, as shown in several qualitative metrics (Dice, Jaccard, PPV, SEN, and Hausdorff). We trust that this tool will avoid human bias and streamline pre-processing steps during 3D high resolution rodent brain MRI data analysis. The software developed herein has been disseminated freely to the community.

The loss of central norepinephrine (NE) released by neurons of the locus coeruleus (LC) occurs with aging, and is thought to be an important factor in producing the many of the nonmotor symptoms and exacerbating the degenerative process in animal models of Parkinson’s disease (PD). We hypothesize that selectively depleting noradrenergic LC neurons prior to the induction of chronic neuroinflammation may not only accelerate the rate of progressive neurodegeneration throughout the brain, but may exacerbate nonmotor and motor behavioral phenotypes that recapitulate symptoms of PD. For this reason, we used a “two-hit” mouse model whereby brain NE were initially depleted by DSP-4 one week prior to exposing mice to LPS. We found that pretreatment with DSP-4 potentiated LPS-induced sequential neurodegeneration in SNpc, hippocampus, and motor cortex, but not in VTA and caudate/putamen. Mechanistic study revealed that DSP-4 enhanced LPS-induced microglial activation and subsequently elevated neuronal oxidative stress in affected brain regions in a time-dependent pattern. To further characterize the effects of DSP-4 on non-motor and motor symptoms in the LPS model, physiological and behavioral tests were performed at different time points following injection. Consistent with the enhanced neurodegeneration, DSP-4 accelerated the progressive deficits of non-motor symptoms including hyposmia, constipation, anxiety, sociability, exaggerated startle response and impaired learning. Furthermore, notable decreases of motor functions, including decreased rotarod activity, grip strength, and gait disturbance, were observed in treated mice. In summary, our studies provided not only an accelerated “two-hit” PD model that recapitulates the features of sequential neuron loss and the progression of motor/non-motor symptoms of PD, but also revealed the critical role of early LC noradrenergic neuron damage in the pathogenesis of PD-like symptoms.

Accurate removal of magnetic resonance imaging (MRI) signal outside the brain, a.k.a., skull stripping, is a key step in the brain image pre-processing pipelines. In rodents, this is mostly achieved by manually editing a brain mask, which is time-consuming and operator dependent. Automating this step is particularly challenging in rodents as compared to humans, because of differences in brain/scalp tissue geometry, image resolution with respect to brain-scalp distance, and tissue contrast around the skull. In this study, we proposed a deep-learning-based framework, U-Net, to automatically identify the rodent brain boundaries in MR images. The U-Net method is robust against inter-subject variability and eliminates operator dependence. To benchmark the efficiency of this method, we trained and validated our model using both in-house collected and publicly available datasets. In comparison to current state-of-the-art methods, our approach achieved superior averaged Dice similarity coefficient to ground truth T2-weighted rapid acquisition with relaxation enhancement and T2∗-weighted echo planar imaging data in both rats and mice (all p < 0.05), demonstrating robust performance of our approach across various MRI protocols.

IL-11 induced differentiation and expansion of Th17 cells in patients with early relapsing-remitting multiple sclerosis (RRMS). In mice with relapsing-remitting experimental autoimmune encephalomyelitis (RREAE), IL-11 exacerbated disease, induced demyelination in the central nervous system (CNS), increased the percentage of IL-17A(+)CD4(+) Th17 cells in the CNS in the early acute phase, and up-regulated serum IL-17A levels and the percentage of IL-17A(+)CD4(+) Th17 cells in lymph nodes, and IFN-γ(+)CD4(+)T cells in spinal cord in the RR phase. IL-11 antagonist suppressed RREAE disease activities, inhibited IL-17A(+)CD4(+) cell infiltration and demyelination in the CNS, and decreased the percentage of IL-17A(+)CD4(+) T cells in peripheral blood mononuclear cells and ICAM1(+)CD4(+) T cells in brain and SC. Diffusion Tensor Imaging indicated that IL-11 antagonist inhibited demyelination in several brain regions. We conclude that by suppressing Th17 cell-mediated neuroinflammation and demyelination, IL-11 antagonist can be further studied as a potential selective and early therapy for RRMS.

This study developed a novel system combining a 16-channel micro-electrocorticography (μECoG) electrode array and functional photoacoustic microscopy (fPAM) to examine changes in neurovascular functions following transient ischemic attack (TIA) in rats. To mimic the pathophysiology of TIA, a modified photothrombotic ischemic model was developed by using 3 min illumination of 5 mW continuous-wave (CW) green laser light focusing on a distal branch of the middle cerebral artery (MCA). Cerebral blood volume (CBV), hemoglobin oxygen saturation (SO2), somatosensory evoked potentials (SSEPs) and alpha-to-delta ratio (ADR) were measured pre- and post-ischemia over a focal cortical region (i.e., 1.5 × 1.5 mm2). Unexpectedly, the SO2, peak-to-peak amplitude (PPA) of SSEPs and ADR recovered and achieved levels greater than the baseline values at the 4th hour post-ischemia induction without any intervention, whereas the CBV value only partially recovered. In other words, transient ischemia led to increased neural activity when the relative CBV was reduced, which may further compromise neural integrity or lead to subsequent vascular disease. This novel μECoG-fPAM system complements currently available imaging techniques and represents a promising technology for studying neurovascular coupling in animal models.

Postischemic angiogenesis is an important recovery mechanism. Both arteries and veins are upregulated during angiogenesis, but eventually there are more angiogenic veins than arteries in terms of number and length. It is critical to understand how the veins are modulated after ischemia and then transitioned into angiogenic vessels during the proangiogenic stage to finally serve as a restorative strength to the injured area. Using a rat model of transient focal cerebral ischemia, the hypercapnic blood oxygen level-dependent (BOLD) response was used to evaluate vascular reactivity, while the hyperoxic BOLD and tissue oxygen level-dependent (TOLD) responses were used to evaluate the vascular functionality at 1, 3, and 7 days after ischemia. Vessel-like venous signals appeared on R2* maps on days 3 and 7, but not on day 1. The large hypercapnic BOLD responses on days 3 and 7 indicated that these areas have high vascular reactivity. The temporal correlation between vascular reactivity and the immunoreactivity to desmin and VEGF further indicates that the integrity of vascular reactivity is associated with the pericyte coverage as regulated by the VEGF level. Vascular functionality remained low on days 1, 3, and 7, as reflected by the small hyperoxic BOLD and large hyperoxic TOLD responses, indicating the low oxygen consumption of the ischemic tissues. These functional changes in proangiogenic veins may be critical for angiogenesis.

To investigate the potential therapeutic effects of peripheral sensory stimulation during the hyperacute phase of stroke, the present study utilized electrophysiology and photoacoustic imaging techniques to evaluate neural and vascular responses of the rat cortex following ischemic insult. We employed a rat model of photothrombotic ischemia (PTI), which targeted the forelimb region of the primary somatosensory cortex (S1FL), due to its high reproducibility in creating localized ischemic injury. We also established a hybrid, dual-modality system, including six-channel electrocorticography (ECoG) and functional photoacoustic microscopy (fPAM), termed ECoG–fPAM, to image brain functional responses to peripheral sensory stimulation during the hyperacute phase of PTI. Our results showed that the evoked cerebral blood volume (CBV) and hemoglobin oxygen saturation (SO2) recovered to 84 ± 7.4% and 79 ± 6.2% of the baseline, respectively, when stimulation was delivered within 2.5 h following PTI induction. Moreover, neural activity significantly recovered, with 77 ± 8.6%, 76 ± 5.3% and 89 ± 8.2% recovery for the resting-state inter-hemispheric coherence, alpha-to-delta ratio (ADR) and somatosensory evoked potential (SSEP), respectively. Additionally, we integrated the CBV or SO2 with ADR values as a recovery indicator (RI) to assess functional recovery after PTI. The RI indicated that 80 ± 4.2% of neurovascular function was preserved when stimulation was delivered within 2.5 h. Additionally, stimulation treatment within this optimal time window resulted in a minimal infarct volume in the ischemic hemisphere (4.6 ± 2.1%). In contrast, the infarct volume comprised 13.7 ± 1.7% of the ischemic hemisphere when no stimulation treatment was applied.

Purpose To develop a series of robust and readily adoptable protocols for the application of deep brain stimulation (DBS)-functional MRI (fMRI) in rodents. Methods DBS-fMRI procedures were conducted in rat and mouse under varying anesthetic conditions (isoflurane in rat and mouse, α-chloralose in rat). A homemade two-channel tungsten microwire electrode was used to minimize magnetic susceptibility artifacts, and was targeted to the ventral posteromedial (VPM) thalamus for DBS-fMRI scanning procedures. Results Compared with a commercially available MR-compatible electrode, the tungsten microwire generated greatly reduced magnetic-susceptibility artifacts. In the rat, VPM-DBS using the microwire electrode resulted in robust positive blood-oxygen-level-dependent signal changes in somatosensory cortex that were relatively independent of anesthetic type. In the mouse, VPM-DBS similarly generated large, positive neurovascular responses in somatosensory cortex that were detected using cerebral blood volume measurements. Conclusion Collectively, this work describes reasonable and easily adoptable procedures for conducting DBS-fMRI studies in rodent models. The protocols developed herein may be extended to study DBS effects under numerous experimental conditions and at varying stimulation targets. Magn Reson Med 73:1246–1251, 2015. © 2014 Wiley Periodicals, Inc.

Background: Functional magnetic resonance imaging (fMRI) of deep brain stimulation (DBS) has potentials to reveal neuroanatomical connectivity of a specific brain region in vivo. Objective: This study aimed to demonstrate frequency and amplitude tunings of the thalamocortical tract using DBS fMRI at the rat ventral posteromedial thalamus. Methods: Blood oxygenation level dependent (BOLD) fMRI data were acquired in a total of twelve rats at a high-field 11.7 T MRI scanner with modulation of nine stimulus frequencies (1–40 Hz) and seven stimulus amplitudes (0.2–3.6 mA). Results: BOLD response in the barrel cortex peaked at 25 Hz. The response increased with stimulus amplitude and reached a plateau at 1 mA. Cortical spreading depolarization (CSD) was observed occasionally after DBS that carries >10% BOLD waves spanning the entire ipsilateral cortex. Conclusion: fMRI is sensitive to the frequency effect of DBS and has potential to investigate the function of a particular neuroanatomical pathway.

Purpose To compare basal retinal and cerebral blood flow (BF) values using continuous arterial spin labeling (CASL) MRI and fluorescent microspheres. Materials and Methods A total of 41 animals were used. BF was measured using an established microsphere technique (a mixture of 2.5 million 8 μm green and 0.5 million 10 μm blue fluorescent microspheres) and CASL MRI blood flow measurement in the rat retina and brain at 7 Tesla (T) and 11.7T, respectively. Results Retinal BF by MRI was 1.18 ± 0.57 mL/g/min and choroidal BF was 8.14 ± 1.8 mL/g/min (n = 6). Microsphere retinal BF was 9.12 ± 2.8 μL/min per tissue and choroidal BF was 73.38 ± 44 μL/min per tissue (n = 18), corresponding to a retinal BF value of 1.22 ± 0.36 mL/g/min by means of a wet weight conversion. The wet-weight of the choroid could not be determined. To corroborate our findings, cerebral BF (CBF) by MRI was also analyzed. In the cerebral cortices, CBF was 0.91 ± 0.29 mL/g/min (n = 14) by CASL MRI and 1.09 ± 0.37 mL/g/min (n = 6) by microspheres. There were no significant differences found between MRI and microsphere blood flow in the retina and brain. Conclusion BF values in the rat retina and cerebral cortex by MRI are in agreement with those obtained by the microsphere technique. J. Magn. Reson. Imaging 2014;40:609–615. © 2013 Wiley Periodicals, Inc.

Blood flow (BF) in many tissues is stable during significant fluctuations in systemic arterial blood pressure or perfusion pressure under normal conditions. The regulatory mechanisms responsible for this non-passive BF behavior include both local and neural control mechanisms. This study evaluated cerebral BF (CBF), retinal BF (RBF) and choroidal BF (ChBF) responses to acute blood pressure increases in rats using magnetic resonance imaging (MRI). A transient increase in blood pressure inside the MRI scanner was achieved by mechanically inflating a balloon catheter to occlude the descending aorta near the diaphragm. We verified the rat model of mechanical occlusion and MRI approach by first measuring blood-flow regulatory responses to changing BP in the brain under normoxia and hypercapnia where the phenomenon is well documented. Retinal and choroidal blood-flow responses to transient increased arterial pressure were then investigated. In response to an acute increase in blood pressure, RBF exhibited autoregulatory behavior and ChBF exhibited baroregulation similar to that seen in the cerebral circulation. This approach may prove useful to investigate retinal and choroidal vascular dysregulation in rat models of retinal diseases with suspected vascular etiology.

High-resolution functional-magnetic-resonance-imaging (fMRI) has been used to study brain functions at increasingly finer scale, but whether fMRI can accurately reflect layer-specific neuronal activities is less well understood. The present study investigated layer-specific cerebral-blood-volume (CBV) fMRI and electrophysiological responses in the rat cortex. CBV fMRI at 40×40 µm in-plane resolution was performed on an 11.7-T scanner. Electrophysiology used a 32-channel electrode array that spanned the entire cortical depth. Graded electrical stimulation was used to study activations in different cortical layers, exploiting the notion that most of the sensory-specific neurons are in layers II–V and most of the nociceptive-specific neurons are in layers V–VI. CBV response was strongest in layer IV of all stimulus amplitudes. Current source density analysis showed strong sink currents at cortical layers IV and VI. Multi-unit activities mainly appeared at layers IV–VI and peaked at layer V. Although our measures showed scaled activation profiles during modulation of stimulus amplitude and failed to detect specific recruitment at layers V and VI during noxious electrical stimuli, there appears to be discordance between CBV fMRI and electrophysiological peak responses, suggesting neurovascular uncoupling at laminar resolution. The technique implemented in the present study offers a means to investigate intracortical neurovascular function in the normal and diseased animal models at laminar resolution.

The Royal College of Surgeons (RCS) rat is an established animal model of retinitis pigmentosa, a family of inherited retinal diseases which starts with loss of peripheral vision and progresses to eventual blindness. Blood flow (BF), an important physiological parameter, is intricately coupled to metabolic function under normal physiological conditions and is perturbed in many neurological and retinal diseases. This study reports non-invasive high-resolution MRI (44 × 44 × 600 μm) to image quantitative retinal and choroidal BF and layer-specific retinal thicknesses in RCS rat retinas at different stages of retinal degeneration compared with age-matched controls. The unique ability to separate retinal and choroidal BF was made possible by the depth-resolved MRI technique. RBF decreased with progressive retinal degeneration, but ChBF did not change in RCS rats up to post-natal day 90. We concluded that choroidal and retinal circulations have different susceptibility to progressive retinal degeneration in RCS rats. Layer-specific retinal thickness became progressively thinner and was corroborated by histological analysis in the same animals. MRI can detect progressive anatomical and BF changes during retinal degeneration with laminar resolution.

The purpose of this study was to improve the accuracy of tissue segmentation on brain magnetic resonance (MR) images preprocessed by multiscale retinex (MSR), segmented with a combined boosted decision tree (BDT) and MSR algorithm (hereinafter referred to as the MSRBDT algorithm). Simulated brain MR (SBMR) T1-weighted images of different noise levels and RF inhomogeneities were adopted to evaluate the outcome of the proposed method; the MSRBDT algorithm was used to identify the gray matter (GM), white matter (WM), and cerebral-spinal fluid (CSF) in the brain tissues. The accuracy rates of GM, WM, and CSF segmentation, with spatial features (G, x, y, r, θ), were respectively greater than 0.9805, 0.9817, and 0.9871. In addition, images segmented with the MSRBDT algorithm were better than those obtained with the expectation maximization (EM) algorithm; brain tissue segmentation in MR images was significantly more precise. The proposed MSRBDT algorithm could be beneficial in clinical image segmentation.

MRI is being used increasingly for the noninvasive longitudinal monitoring of cellular processes in various pathophysiological conditions. Macrophages are the main stromal cells in neoplasms and have been suggested to be the major cell type ingesting superparamagnetic iron oxide (SPIO) nanoparticles. However, no MRI study has described longitudinally the presence of tumor-associated macrophages (TAMs) during tumorigenesis with histological confirmation. To address this, we injected SPIO nanoparticles into the circulation of tumor-bearing mice and used MRI and post-mortem histology to monitor TAMs at different time points. The MRI results demonstrated that TAMs, as hypointense signals, appeared continually with the expansion of the tumor. The histological findings also revealed that SPIO-labeled TAMs tended to deposit closer to the vessel lumen with time prior to rapid tumor growth. The present study demonstrates the potential of using MRI to assess longitudinally TAM accumulation during tumorigenesis, and provides the first in vivo insight into the topographical arrangement of TAMs in relation to the progression of tumors. In vivo monitoring of the presence of TAMs could be useful for the development of tumor treatments that target TAM functions.

Although optically based imaging techniques provide valuable functional and physiological information of the retina, they are mostly limited to the probing of the retinal surface and require an unobstructed light path. MRI, in contrast, could offer physiological and functional data without depth limitation. Blood oxygenation level-dependent functional MRI (BOLD fMRI) of the thin rat retina is, however, challenging because of the need for high spatial resolution, and the potential presence of eye movement and susceptibility artifacts. This study reports a novel application of high-resolution (111 × 111 × 1000 µm(3)) BOLD fMRI of visual stimulation in the anesthetized rat retina at 11.7 T. A high-field MRI scanner was utilized to improve the signal-to-noise ratio, spatial resolution and BOLD sensitivity. Visual stimuli (8 Hz diffuse achromatic light) robustly increased BOLD responses in the retina [5.0 ± 0.8% from activated pixels and 3.1 ± 1.1% from the whole-retina region of interest (mean ± SD), n = 12 trials on six rats, p 0.05). BOLD fMRI of visual stimulation has the potential to provide clinically relevant data to probe hemodynamic neurovascular coupling and dysfunction of the retina with depth resolution.

The present study reported the development of a novel functional photoacoustic microscopy (fPAM) system for investigating hemodynamic changes in rat cortical vessels associated with electrical forepaw stimulation. Imaging of blood optical absorption by fPAM at multiple appropriately-selected and distinct wavelengths can be used to probe changes in total hemoglobin concentration (HbT, i.e., cerebral blood volume [CBV]) and hemoglobin oxygen saturation (SO(2)). Changes in CBV were measured by images acquired at a wavelength of 570nm (lambda(570)), an isosbestic point of the molar extinction spectra of oxy- and deoxy-hemoglobin, whereas SO(2) changes were sensed by pixel-wise normalization of images acquired at lambda(560) or lambda(600) to those at lambda(570). We demonstrated the capacity of the fPAM system to image and quantify significant contralateral changes in both SO(2) and CBV driven by electrical forepaw stimulation. The fPAM system complements existing imaging techniques, with the potential to serve as a favorable tool for explicitly studying brain hemodynamics in animal models.

SummaryObjectiveThe purpose of this study was to investigate whether the effect of treatment with hyaluronic acid (HA) on cartilage in osteoarthritis (OA) can be determined by measuring the magnetic resonance (MR) T2 value of cartilage in an anterior cruciate ligament transection (ACLX) animal model.MethodEighteen male Sprague Dawley rats were separated randomly into three groups (n=6 for each group). Group 1 was given ACLX and intra-articular (IA) normal saline (NS) injection (ACLX+NS), group 2 was given ACLX and IA HA injection (ACLX+HA), and group 3 was the sham control. The ACLX+NS and ACLX+HA groups received ACLX on the right knee at 8 weeks of age and were then treated with IA NS or HA injection once a week, respectively, for 4 weeks starting at 13 weeks of age. In the sham-control group, the right knee joint was opened surgically but ACLX was not performed at 8 weeks of age. MR T2 measurements were obtained on all rats at 8, 12, and 21 weeks of age, and histological Mankin scoring was performed at 21 weeks of age.ResultsFive weeks after the 4-week treatment, the MR T2 value of the ACLX right knee cartilage was significantly lower in ACLX+HA (29.58±1.12ms) than in ACLX+NS (32.04±1.39ms) (P

Functional magnetic resonance imaging (fMRI) has revolutionized investigations of brain functions. Increases in fMRI signals are usually correlated with neuronal activation, but diverse explanations have been proposed for negative fMRI responses, including decreases in neuronal activity, the vascular-steal effect, and large increases in oxygen consumption. These possible scenarios, although encompassing a wide range of potential neurovascular responses, cannot yet be used to interpret certain types of negative fMRI signals. Recent studies have found that intravenous injection of dopamine D2receptor (D2DR) agonist reduced the hemodynamic responses in the caudate–putamen (CPu); however, whether endogenous dopaminergic neurotransmission contributes to fMRI signals remains obscure. Since it has been suggested that the D2DR is involved in pain modulation, and the CPu shows equivocal fMRI signals during noxious stimulation, the present study established an animal model based on graded electrical stimulation to elicit different levels of nociception, and aimed to determine whether nociception-induced endogenous dopaminergic neurotransmission is sufficient to generate negative fMRI responses. Our results from cerebral blood volume (CBV)-weighted fMRI, Fos immunohistochemistry, and electrophysiological recording demonstrated a salient bilateral CBV decreases associated with heightened neuronal activity in the CPu induced by unilateral noxious electrical stimulation. In addition, preinjection of D2DR antagonist reduced the observed CBV decreases. Our findings reveal the role of the D2DR in regulating striatal vascular responses and suggest that endogenous neurotransmission-induced CBV decreases underlie negative fMRI signals. Hence, the influence of endogenous neurotransmission should be considered when interpreting fMRI data, especially in an area involved in strong vasoactive neurotransmission.

Here we proposed an automatic segmentation method based on a decision tree to classify the brain tissues in magnetic resonance (MR) images. Two types of data - phantom MR images obtained from IBSR (http://www.cma.mgh.harvard.edu/ibsr) and simulated brain MR images obtained from BrainWeb (http://www.bic.mni.mcgill.ca/brainweb) - were segmented using an automatic decision tree algorithm to obtain images with improved visual rendition. Spatial information on the general gray level (G), spatial gray level (S), and two-dimensional wavelet transform (W) was combined in-plane in two coordinate systems (Euclidean coordinates (x, y) or polar coordinates (r, theta)). The decision tree was constructed based on a binary tree with nodes created by splitting the distribution of input features of the tree. The spatial information obtained from MR images with different noise levels and inhomogeneities were segmented to compare whether the use of a decision tree improved the identification of human anatomical structures in a neuroimage. The average accuracy rates of segmentation for phantom images with a noise variation of 15 gray levels were 0.9999 and 0.9973 with spatial information (G, x, y, r, theta) and (S, x, y, r, theta), respectively, and 0.9999 and 0.9819 with spatial information (G, x, y, S, r, theta) and (W, x, y, G, r, theta). The average accuracy rates of segmentation for simulated MR images with a noise level of 5% were 0.9532 and 0.9439 with spatial information (G, x, y, r, theta) and (S, x, y, r, theta), respectively, and 0.9446 and 0.9287 with spatial information (G, x, y, S, r, theta) and (W, x, y, G, r, theta). The accuracy rates of segmentation were highest for both simulated phantom and brain MR images, having the lowest noise levels, from a reduction of overlapping gray levels in the images. The accuracies of segmentation were higher when the spatial information included the general gray level than when it included the spatial gray level, which in turn were higher than when it included the wavelet transform. Furthermore, the performance of segmentation was also evaluated with a boundary detection methodology that is based on the Hausdorff distance to compare with the mean computer to observer difference (COD) and mean interobserver difference (IOD) for gray matter (GM), white matter (WM), and all areas (ALL) from images segmented using the decision tree. The values of mean COD are similar and around 12mm for GM segmented using the decision tree. Our segmentation method based on a decision tree algorithm presented an easy way to perform automatic segmentation for both phantom and tissue regions in brain MR images.

The purpose of this study was to improve the accuracy rate of brain tissue classification in magnetic resonance (MR) imaging using a boosted decision tree segmentation algorithm. Herein, we examined simulated phantom MR (SPMR) images, simulated brain MR (SBMR) images, and a real data. The accuracy rate and k index when classifying brain tissues as gray matter (GM), white matter (WM), or cerebral-spinal fluid (CSF) were better when using the boosted decision tree algorithm combined with a fuzzy threshold than when using a statistical region-growing (SRG) algorithm [Wolf I, Vetter M, Wegner I, Böttger T, Nolden M, Schöbinger M, et al. The medical imaging interaction toolkit. Med Imag Anal 2005;9:594-604] and an adaptive segmentation (AS) algorithm [Wells WM, Grimson WEL, Kikinis R, Jolesz FA. Adaptive segmentation of MRI data. IEEE Trans Med Imag 1996;15:429-42]. The segmentation performance when using this algorithm on real data from brain MR images was also better than those of SRG and AS algorithm. Segmentation of a real data using the boosted decision tree produced particularly clear brain MR imaging and permitted more accurate brain tissue segmentation. In conclusion, a decision tree with appropriate boost trials successfully improved the accuracy rate of MR brain tissue segmentation.

A spherical mass of the right psoas muscle was found incidentally by magnetic resonance health examination in a 43-year-old woman. The patient had complained of numbness in her right leg over the previous six months and neuralgia at the time of computed tomography-guided core biopsy, which was done to establish the diagnosis. To our knowledge, femoral nerve neuropathy caused by an intramuscular schwannoma arising from the psoas muscle has not been previously published.