SignificanceComorbidities such as mood and cognitive disorders are often found in individuals with epilepsy after seizures. Cortex processes sensory, motor, and cognitive information. Brain circuit changes can be studied by observing functional network changes in epileptic mice’s cortex.AimThe cortex is easily accessible for non-invasive brain imaging and electroencephalogram recording (EEG). However, the impact of seizures on cortical activity and functional connectivity has been rarely studied in vivo.ApproachIntrinsic optical signal and EEG were used to monitor cortical activity in awake mice within 4 h after pilocarpine induction. It was divided into three periods according to the behavior and EEG of the mice: baseline, onset of seizures (onset, including seizures and resting in between seizure events), and after seizures (post, without seizures). Changes in cortical activity were compared between the baseline and after seizures.ResultsHemoglobin levels increased significantly, particularly in the parietal association cortex (PT), retrosplenial cortex (RS), primary visual cortex (V1), and secondary visual cortex (V2). The network-wide functional connectivity changed post seizures, e.g., hypoconnectivity between PT and visual-associated cortex (e.g., V1 and V2). In contrast, connectivity between the motor-associated cortex and most other regions increased. In addition, the default mode network (DMN) also changed after seizures, with decreased connectivity between primary somatosensory region (SSp) and visual region (VIS), but increased connectivity involving anterior cingulate cortex (AC) and RS.ConclusionsOur results provide references for understanding the mechanisms behind changes in brain circuits, which may explain the profound effects of seizures on comorbid health conditions.
SignificanceRobust segmentations of neurons greatly improve neuronal population reconstruction, which could support further study of neuron morphology for brain research.AimPrecise segmentation of 3D neuron structures from optical microscopy (OM) images is crucial to probe neural circuits and brain functions. However, the high noise and low contrast of images make neuron segmentation challenging. Convolutional neural networks (CNNs) can provide feasible solutions for the task but they require large manual labels for training. Labor-intensive labeling is highly expensive and heavily limits the algorithm generalization.ApproachWe devise a weakly supervised learning framework Docker-based deep network plus (DDeep3M+) for neuron segmentation without any manual labeling. A Hessian analysis based adaptive enhancement filter is employed to generate pseudo-labels for segmenting neuron images. The automated segmentation labels are input for training a DDeep3M to extract neuronal features. We mine more undetected weak neurites from the probability map based on neuronal structures, thereby modifying the pseudo-labels. We iteratively refine the pseudo-labels and retrain the DDeep3M model with the pseudo-labels to obtain a final segmentation result.ResultsThe proposed method achieves promising results with the F1 score of 0.973, which is close to that of the CNN model with manual labels and superior to several segmentation algorithms.ConclusionsWe propose an accurate weakly supervised neuron segmentation method. The high precision results achieved on 3D OM datasets demonstrate the superior generalization of our DDeep3M+.
The number of neuronal cells is fundamentally important for brain functions. However, it can be difficult to obtain the accurate number of neuronal cells in large-scale brain imaging, which is nearly inevitable with traditional image segmentation techniques due to the low contrast and noisy background. Here, we introduce a Docker-based deep convolutional neural network (DDeep3M) for better counting neurons in the stimulated Raman scattering (SRS) microscopy images. To reconcile the memory limit of computational resource, a high-resolution 2D SRS image of whole coronal slice of mouse brain is divided into multiple patch images. Each patch image is then fed into the DDeep3M and predicted as a probability map. A higher contrast image targeting neurons (i.e. the predicted image) can be acquired by stitching the patches of probability map together. With this routine segmentation method applied in both raw SRS image and the predicted image, the DDeep3M achieves the accuracy of over 0.96 for cell counting which is much better than the result of traditional segmentation methods. Compared with the U-Net, which is one of the most popular deep learning networks for medical image segmentation, DDeep3M demonstrates a better result when handling such large-scale image. Thus, DDeep3M can be really helpful for large-scale cell counting in brain research.
Neuronal cells play very important role on metabolism regulation and mechanism control, so cell number is a fundamental determinant of brain function. Combined suitable cell-labeling approaches with recently proposed three-dimensional optical imaging techniques, whole mouse brain coronal sections can be acquired with 1-μm voxel resolution. We have developed a completely automatic pipeline to perform cell centroids detection, and provided three-dimensional quantitative information of cells in the primary motor cortex of C57BL/6 mouse. It involves four principal steps: i) preprocessing; ii) image binarization; iii) cell centroids extraction and contour segmentation; iv) laminar density estimation. Investigations on the presented method reveal promising detection accuracy in terms of recall and precision, with average recall rate 92.1% and average precision rate 86.2%. We also analyze laminar density distribution of cells from pial surface to corpus callosum from the output vectorizations of detected cell centroids in mouse primary motor cortex, and find significant cellular density distribution variations in different layers. This automatic cell centroids detection approach will be beneficial for fast cell-counting and accurate density estimation, as time-consuming and error-prone manual identification is avoided.
Cortical spreading depression (CSD) is an important neurophysiological phenomenon correlating with some neural
disorders, such as migraine, cerebral ischemia and epilepsy. By now, we are still not clear about the mechanisms of
CSD's initiation and propagation, also the relevance between CSD and those neural diseases. Nevertheless,
characterization of CSD, especially the spatiotemporal evolution, will promote the understanding of the CSD's nature
and mechanisms. Besides the previous experimental work on charactering the spatiotemporal evolution of CSD in rats by
optical intrinsic signal imaging, a computational study based on a generalized cellular automaton (CA) model was
proposed here. In the model, we exploited a generalized neighborhood connection rule: a central CA cell is related with a
group of surrounding CA cells with different weight coefficients. By selecting special parameters, the generalized CA
model could be transformed to the traditional CA models with von Neumann, Moore and hexagon neighborhood
connection means. Hence, the new model covered several properties of CSD simulated in traditional CA models: 1)
expanding from the origin site like a circular wave; 2) annihilation of two waves traveling in opposite directions after
colliding; 3) wavefront of CSD breaking and recovering when and after encountering an obstacle. By setting different
refractory period in the different CA lattice field, different connection coefficient in different direction within the defined
neighborhood, inhomogeneous propagation of CSD was simulated with high fidelity. The computational results were
analogous to the reported time-varying CSD waves by optical imaging. So, the generalized CA model would be useful to
study CSD because of its intuitive appeal and computational efficiency.
This study aimed to investigate the variation of propagation patterns of successive cortical spreading depression (CSD)
waves induced by K+ or pinprick in rat cortex. In the K+ induction group, 18 Sprague-Dawley rats under
α-chloralose/urethane anesthesia were used to elicit CSD by 1 M KCl solution in the frontal cortex. Optical intrinsic
signal imaging (OISI) at an isosbestic point of hemoglobin (550 nm) was applied to examine regional cerebral blood
volume (CBV) changes in the parieto-occipital cortex. In 6 of the 18 rats, OISI was performed in conjunction with DC
potential recording of the cortex. The results of this group were reported previously. In the pinprick group, 6 rats were
used to induce CSD by pinprick with 8 min interval, and the other 6 rats were pricked with 4 min. CBV changes during
CSD appeared as repetitive propagation of wave-like hyperemia at a speed of 3.7±0.4 mm/min, which was characterized
by a significant negative peak (-14.3±3.2%) in the reflectance signal. Except for the first CSD wave, the following waves
don't spread fully in the observed cortex all the time and they might abort in the medial area. Independent on the
stimulation of pinprick or K+, a short interval of the current CSD to the last CSD no more than 4 min would induce the
current CSD be partially propagated. For the first time, the data reveals the time-varying propagation patterns of CSD
waves might be affected by the interval between CSD waves. The results suggest that the propagation patterns of a series
of CSD waves are time-varying in different regions of rat cortex, and the variation is related to the interval between CSD
waves.
Spreading depression (SD) has been found involved in focal cerebral ischemia which may result in severe or lethal
neurological deficits. Electrical recording of SD has been used for acute and long term monitoring of focal cerebral
ischemia but with an inherently low resolution. Here, we presented optical intrinsic signal imaging (OISI) to characterize
the spontaneous SD waves following permanent middle cerebral artery occlusion (MCAO) in rats with high spatial
resolution. During each SD episode, the measured optical reflectance varied regionally: decreased (-12.5±2.8%) in the
area near the midline, remained flat (3.1±2.5%) in the lateral region, and increased (12.1±3.6%) in the intermediate
cortex. The three types of changes yielded identifications for three biological relevant zones: nonischemic cortex,
penumbra and infarct core. Accompanying recurrent SD waves, the suggested penumbral area reduced by about
6.4±2.5% of the whole imaged area per SD event, indicating a growth of the infracted area. Staining with 2%
2,3,5-triphenyltetrazolium chloride (TTC) 4 h post-occlusion proved the infarct cortex to be consistent with the lateral
region where the final SD wave did not invade (r=0.86±0.10). The results suggest that OISI based on SD can effectively
used to distinguish nonischemic cortex, penumbra and infarct core in the ischemic hemisphere and monitor the
development of ischemia with high spatial resolution.
Spreading depression (SD) waves occur in focal cerebral ischemia of the brain. Optical reflectance imaging at 550±10-nm wavelength using a charge-coupled device (CCD) camera, called optical intrinsic signal imaging (OISI) in the neuroscience community, provides high resolution imaging of SD waves based on changes in blood perfusion. We present optical images of SD waves in normal rat brain induced by a pinprick, and the spontaneous SD waves that follow middle cerebral artery occlusion (MCAO). The images of change in reflectance are calculated as A=(I–Io)/Io, where I is pixel intensity as some timepoint and Io is the initial intensity just prior to an SD wave. Difference images B=[I(i)–I(i–1)]/Io, where I(i) is the image at time i and I(i–1) is the previous image at time i–1 (a 6.4-s interval), significantly sharpen the boundaries between leading and trailing edges of the SD wave. Maximum rate-of-change images C=max(B) display the maximum pixel value of B within the duration of a single SD wave, and provide an image that visualizes the entire penumbra. The penumbra appear bright due to a rapid drop in perfusion, while the normal brain and infarct area appear dark.
A series of images are acquired by optical intrinsic signal imaging (OISI) at 550 nm during cortical spreading depression (CSD) in rats. Temporal clustering analysis (TCA), which is an exploratory data-driven technique that has been proposed for the analysis of fMRI data and laser speckle contrast images, is applied to tract the extreme response during CSD. The minimum of optical intrinsic signals (OIS) during CSD in each pixel, corresponding to the maximum change of regional cerebral blood volume (CBV), is determined by TCA. Interestingly, the spatial pattern of the maximum activation shows the ongoing expanding circle. In order to describe the circular pattern quantitatively, we present the least square estimation (LSE) to detect the three parameters of the expanding circles (radius R, center coordinates (a, b)) at each time point (i.e. in each frame). The evaluated mean centers of the circles (1.50±0.06 mm, 2.62±0.03 mm) were tightly correlated with the pinprick site (1.4±0.2 mm, 2.5±0.2 mm). According to the varying values of the radiuses at different time point, we calculate the propagated speed of the CSD at 3.96 mm/min. The results substantiate the CSD spreads like a wave from the induced site to the periphery at 2~5 mm/min over the cortex. So, the combination of TCA and LSE enables the image analysis of OISI more automatic and effective.
Intrinsic optical signals imaging (IOSI) is a novel technique for functional neuroimaging in vivo, especially in the study of cortical spreading depression (CSD). At 550 nm wavelength, the optical images during CSD showed significant vasodilatation of some small arteries in the surface of cortex of rats. In order to quantify the arteries’ diameter change, two kinds of threshold segmentation methods are applied, one is Isodata algorithm thresholding, the other is Otsu’s thresholding. Firstly, we set up a simple model to prove that segmentation of the vessel in a rectangle region could be equally transferred to describe the diameter change. The two methods could automatically select right thresholds for segmentation, so they were suitable for acquiring the dynamic vasodilatation in a series of optical images by computer. Comparing with the traditional method, the new methods were more robust and with high performance. By the methods, we found the vasodilatation could be distinguished as two processes during one CSD episode, a small vasodilatation before the great one that had been commonly reported before. The hemodynamic character during CSD deserves further study. And the methods can be easily applied to the other optical imaging experiments when the vascular dynamic is concerned.
The optical intrinsic signal imaging is an indirect mapping of neuronal activity. The change in light intensity due to neuronal activity are often very small, no more than 0.1-6% of the total intensity of the reflected light in optimal cases, and the noise, which arise from either the biological noise associated with the respiration, circulation and irrelevant physiological activity or the instrumentation noise such as digitization noise, illumination noise, movement artifacts, etc. are usually large. In previous studies, a couple of analysis methods such as Standard Difference, Principal Component Analysis (PCA), Independent Component Analysis (ICA), Truncated Differences were used to suppress these large background noises and extract the small signal of interest from the noisy raw data. The performance of these methods for improving the determination of spatial pattern and time course of the response signal were examined and compared in this paper. The evaluations were employed to both simulated data and experimental optical intrinsic signal imaging data from rat somatosensory cortex during the electrical stimulation at contralateral sciatic nerve.
Intrinsic optical signals imaging (IOSI) and laser speckle imaging (LSI) are both novel techniques for functional neuroimaging in vivo. Combining them to study cortical spreading depression (CSD) which is an important disease model for migraine and other neurological disorders. CSD were induced by pinprick in Sprague-Dawley rats. Intrinsic optical signals (IOS) at 540 nm showed CSD evolution happened in one hemisphere cortex at speeds of 3.7±0.4 mm/min, and the vasodilation closely correlated a four-phasic response. By LSI, we observed a transient and significant increase cerebral blood flow (CBF). In this paper, optical imaging would be showed as a powerful tool for describing the hemodynamic character during CSD in rat.
Cortical spreading depression (CSD) is an important disease model for migraine and cerebral ischemia. We investigated the spatio-temporal characteristics of the intrinsic optical signals (IOS) at 570 nm and the cerebral blood vessel responses during CSD simultaneously by optical reflectance imaging in vivo. The CSD were induced by pinprick in 10 α-chloralose/urethane anesthetized Sprague-Dawley rats. A four-phasic IOS response was observed at pial arteries and parenchymal sites in all experimental animals and an initial slight pial arteries dilation (21.5%±13.6%) and constriction (-4.2%±3.5%) precedes the dramatic dilation (69.2%±26.1%) of pial arterioles was recorded. Our experimental results show a high correlation (r = 0.89±0.025) between the IOS response and the diameter changes of the cerebral blood vessels during CSD in rats.
Optical imaging method was applied into observing the temporal-spatial characteristic of rat primary somatosensory cortex during graded electrical stimulation of the sciatic nerve (5hz,duration of 2s,0.5ms puls,1x,10x and 20x muscle twitch threshold). We found that the temporal and spatial properties of hindlimb somatosensory cortex were modulated by graded intensity electrical stimulation of the sciatic nerve. The magnitude and time course were larger and longer with the intensity raising. And the spatial extent was wider at 20x stimulus than the other two kinds of stimulus. Therefore, our optical imaging was based on 570nm, which only reflect the changes of blood volume. Then our future study will reveal more information of pain modulation in primary somatosensory cortex.
The spatio-temporal characteristics of changes in cerebral blood volume associated with neuronal activity were investigated in the hindlimb somatosensory cortex of α-chloralose/urethan anesthetized rats (n=10) with optical imaging at 570nm through a thinned skull. Activation of cortex was carried out by electrical stimulation of the contralateral sciatic nerve with 5Hz, 0.3V pulses (0.5ms) for duration of 2s. The stimulation evoked a monophasic optical reflectance decrease at cortical parenchyma and arteries sites rapidly after the onset of stimulation, whereas no similar response was observed at vein compartments. The optical signal changes reached 10% of the peak response 0.70±0.32s after stimulation onset and no significant time lag in this 10% start latency time was observed between the response at cortical parenchyma and arteries compartments. The evoked optical reflectance decrease reached the peak (0.25%±0.047%)2.66±0.61s after the stimulus onset at parenchyma site, 0.40±0.20s earlier (P<0.05) than that at arteries site (0.50%±0.068% 3.06±0.70s). Variable location within the cortical parenchyma and arteries compartment themselves didn’t affect the temporal characteristics of the evoked signal significantly. These results suggest that the sciatic nerve stimulation evokes a local blood volume increase at both capillaries (cortical parenchyma) and arterioles rapidly after the stimulus onset but the evoked blood volume increase in capillaries could not be entirely accounted for by the dilation of arterioles.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.