A personalized cortical atlas for functional regions of interest.

Advances in functional MRI (fMRI) allow mapping an individual's brain function in vivo. Task fMRI can localize domain-specific regions of cognitive processing or functional regions of interest (fROIs) within an individual. Moreover, data from resting state (no task) fMRI can be used to define an individual's connectome, which can characterize that individual's functional organization via connectivity-based parcellations. However, can connectivity-based parcellations alone predict an individual's fROIs? Here, we describe an approach to compute individualized rs-fROIs (i.e., regions that correspond to given fROI constructed using only resting state data) for motor control, working memory, high-level vision, and language comprehension. The rs-fROIs were computed and validated using a large sample of young adults (n = 1,018) with resting state and task fMRI from the Human Connectome Project. First, resting state parcellations were defined across a sequence of resolutions from broadscale to fine-grained networks in a training group of 500 individuals. Second, 21 rs-fROIs were defined from the training group by identifying the rs network that most closely matched task-defined fROIs across all individuals. Third, the selectivity of rs-fROIs was investigated in a training set of the remaining 518 individuals. All computed rs-fROIs were indeed selective for their preferred category. Critically, the rs-fROIs had higher selectivity than probabilistic atlas parcels for nearly all fROIs. In conclusion, we present a potential approach to define selective fROIs on an individual-level circumventing the need for multiple task-based localizers.NEW & NOTEWORTHY We compute individualized resting state parcels that identify an individual's own functional regions of interest (fROIs) for high-level vision, language comprehension, motor control, and working memory, using only their functional connectome. This approach demonstrates a rapid and powerful alternative for finding a large set of fROIs in an individual, using only their unique connectivity pattern, which does not require the costly acquisition of multiple fMRI localizer tasks.

Impact of Region-of-Interest Size on Immune Profiling Using Multiplex Immunofluorescence Tyramide Signal Amplification for Paraffin-Embedded Tumor Tissues.

INTRODUCTION: Representative regions of interest (ROIs) analysis from the whole slide images (WSI) are currently being used to study immune markers by multiplex immunofluorescence (mIF) and single immunohistochemistry (IHC). However, the amount of area needed to be analyzed to be representative of the entire tumor in a WSI has not been defined. METHODS: We labeled tumor-associated immune cells by mIF and single IHC in separate cohorts of non-small cell lung cancer (NSCLC) samples and we analyzed them as whole tumor area as well as using different number of ROIs to know how much area will be need to represent the entire tumor area. RESULTS: For mIF using the InForm software and ROI of 0.33 mm2 each, we observed that the cell density data from five randomly selected ROIs is enough to achieve, in 90% of our samples, more than 0.9 of Spearman correlation coefficient and for single IHC using ScanScope tool box from Aperio and ROIs of 1 mm2 each, we found that the correlation value of more than 0.9 was achieved using 5 ROIs in a similar cohort. Additionally, we also observed that each cell phenotype in mIF influence differently the correlation between the areas analyzed by the ROIs and the WSI. Tumor tissue with high intratumor epithelial and immune cells phenotype, quality, and spatial distribution heterogeneity need more area analyzed to represent better the whole tumor area. CONCLUSION: We found that at minimum 1.65 mm2 area is enough to represent the entire tumor areas in most of our NSCLC samples using mIF.

Linking MS1 and MS2 signals in positive and negative modes of LC-HRMS in untargeted metabolomics using the ROIMCR approach.

Data-independent acquisition (DIA) mode in liquid chromatography (LC) high-resolution mass spectrometry (HRMS) has emerged as a powerful strategy in untargeted metabolomics for detecting a broad range of metabolites. However, the use of this approach also represents a challenge in the analysis of the large datasets generated. The regions of interest (ROI) multivariate curve resolution (MCR) approach can help in the identification and characterization of unknown metabolites in their mixtures by linking their MS1 and MS2 DIA spectral signals. In this study, it is proposed for the first time the analysis of MS1 and MS2 DIA signals in positive and negative electrospray ionization modes simultaneously to increase the coverage of possible metabolites present in biological systems. In this work, this approach has been tested for the detection and identification of the amino acids present in a standard mixture solution and in fish embryo samples. The ROIMCR analysis allowed for the identification of all amino acids present in the analyzed mixtures in both positive and negative modes. The methodology allowed for the direct linking and correspondence between the MS signals in their different acquisition modes. Overall, this approach confirmed the advantages and possibilities of performing the proposed ROIMCR simultaneous analysis of mass spectrometry signals in their differing acquisition modes in untargeted metabolomics studies.

Ultrasound-based radiomics in the diagnosis of carpal tunnel syndrome: The influence of regions of interest delineation method on mode.

BACKGROUND: In the recent years, artificial intelligence (AI) algorithms have been used to accurately diagnose musculoskeletal diseases. However, it is not known whether the particular regions of interest (ROI) delineation method would affect the performance of the AI algorithm. PURPOSE: The purpose of this study was to investigate the influence of ROI delineation methods on model performance and observer consistency. METHODS: In this retrospective analysis, ultrasound (US) measures of median nerves affected with carpal tunnel syndrome (CTS) were compared to median nerves in a control group without CTS. Two methods were used for delineation of the ROI: (1) the ROI along the hyperechoic medial edge of the median nerve but not including the epineurium (MN) (ROI1); and (2) the ROI including the hyperechoic epineurium (ROI2), respectively. The intra group correlation coefficient (ICC) was used to compare the observer consistency of ROI features (i.e. the corresponding radiomics parameters). Parameters α1 and α2 were obtained based on the ICC of ROI1 features and ROI2 features. The ROC analysis was used to determine the area under the curve (AUC) and evaluate the performance of the radiologists and network. In addition, four indices, namely sensitivity, specificity, positive prediction and negative prediction were analyzed too. RESULTS: A total of 136 wrists of 77 CTS group and 136 wrists of 74 control group were included in the study. Control group was matched to CTS group according to the age and sex. The observer consistency of ROI features delineated by the two schemes was different, and the consistency of ROI1 features was higher (α1 Ëƒ α2). The intra-observer consistency was higher than the inter-observer consistency regardless of the scheme, and the intra-observer consistency was higher when chose scheme one. The performances of models based on the two ROI features were different, although the AUC of each model was greater than 0.8.The model performed better when the MN epineurium was included in the ROI. Among five artificial intelligence algorithms, the Forest models (model1 achieved an AUC of 0.921 in training datasets and 0.830 in testing datasets; model2 achieved an AUC of 0.967 in training datasets and 0.872 in testing datasets.) obtained the highest performance, followed by the support vector machine (SVM) models and the Logistic models. The performances of the models were significantly better than the inexperienced radiologist (Dr. B. Z. achieved an AUC of 0.702). CONCLUSION: Different ROI delineation methods may affect the performance of the model and the consistency of observers. Model performance was better when the ROI contained the MN epineurium, and observer consistency was higher when the ROI was delineated along the hyperechoic medial border of the MN.

Dual-ATME: Dual-Branch Attention Network for Micro-Expression Recognition.

Micro-expression recognition (MER) is challenging due to the difficulty of capturing the instantaneous and subtle motion changes of micro-expressions (MEs). Early works based on hand-crafted features extracted from prior knowledge showed some promising results, but have recently been replaced by deep learning methods based on the attention mechanism. However, with limited ME sample sizes, features extracted by these methods lack discriminative ME representations, in yet-to-be improved MER performance. This paper proposes the Dual-branch Attention Network (Dual-ATME) for MER to address the problem of ineffective single-scale features representing MEs. Specifically, Dual-ATME consists of two components: Hand-crafted Attention Region Selection (HARS) and Automated Attention Region Selection (AARS). HARS uses prior knowledge to manually extract features from regions of interest (ROIs). Meanwhile, AARS is based on attention mechanisms and extracts hidden information from data automatically. Finally, through similarity comparison and feature fusion, the dual-scale features could be used to learn ME representations effectively. Experiments on spontaneous ME datasets (including CASME II, SAMM, SMIC) and their composite dataset, MEGC2019-CD, showed that Dual-ATME achieves better, or more competitive, performance than the state-of-the-art MER methods.

Added value of semiquantitative analysis of brain FDG-PET for the differentiation between MCI-Lewy bodies and MCI due to Alzheimer's disease.

PURPOSE: FDG-PET is an established supportive biomarker in dementia with Lewy bodies (DLB), but its diagnostic accuracy is unknown at the mild cognitive impairment (MCI-LB) stage when the typical metabolic pattern may be difficultly recognized at the individual level. Semiquantitative analysis of scans could enhance accuracy especially in less skilled readers, but its added role with respect to visual assessment in MCI-LB is still unknown. METHODS: We assessed the diagnostic accuracy of visual assessment of FDG-PET by six expert readers, blind to diagnosis, in discriminating two matched groups of patients (40 with prodromal AD (MCI-AD) and 39 with MCI-LB), both confirmed by in vivo biomarkers. Readers were provided in a stepwise fashion with (i) maps obtained by the univariate single-subject voxel-based analysis (VBA) with respect to a control group of 40 age- and sex-matched healthy subjects, and (ii) individual odds ratio (OR) plots obtained by the volumetric regions of interest (VROI) semiquantitative analysis of the two main hypometabolic clusters deriving from the comparison of MCI-AD and MCI-LB groups in the two directions, respectively. RESULTS: Mean diagnostic accuracy of visual assessment was 76.8 ± 5.0% and did not significantly benefit from adding the univariate VBA map reading (77.4 ± 8.3%) whereas VROI-derived OR plot reading significantly increased both accuracy (89.7 ± 2.3%) and inter-rater reliability (ICC 0.97 [0.96-0.98]), regardless of the readers' expertise. CONCLUSION: Conventional visual reading of FDG-PET is moderately accurate in distinguishing between MCI-LB and MCI-AD, and is not significantly improved by univariate single-subject VBA but by a VROI analysis built on macro-regions, allowing for high accuracy independent of reader skills.

Adaptive Multi-ROI Agricultural Robot Navigation Line Extraction Based on Image Semantic Segmentation.

Automated robots are an important part of realizing sustainable food production in smart agriculture. Agricultural robots require a powerful and precise navigation system to be able to perform tasks in the field. Aiming at the problems of complex image background, as well as weed and light interference factors of the visual navigation system in field and greenhouse environments, a Faster-U-net model that retains the advantages of the U-net model feature jump connection is proposed. Based on the U-net model, pruning and optimization were carried out to predict crop ridges. Firstly, a corn dataset was trained to obtain the weight of the corn dataset. Then, the training weight of the obtained corn dataset was used as the pretraining weight for the cucumber, wheat, and tomato datasets, respectively. The three datasets were trained separately. Finally, the navigation line between ridges and the yaw angle of the robot were generated by B-spline curve fitting. The experimental results showed that the parameters of the improved path segmentation model were reduced by 65.86%, and the mPA was 97.39%. The recognition accuracy MIoU of the Faster-U-net model for maize, tomatoes, cucumbers, and wheat was 93.86%, 94.01%, 93.14%, and 89.10%, respectively. The processing speed of the single-core CPU was 22.32 fps/s. The proposed method had strong robustness in predicting rows of different crops. The average angle difference of the navigation line under a ridge environment such as that for corn, tomatoes, cucumbers, or wheat was 0.624°, 0.556°, 0.526°, and 0.999°, respectively. This research can provide technical support and reference for the research and development of intelligent agricultural robot navigation equipment in the field.

Remote Heart Rate Prediction in Virtual Reality Head-Mounted Displays Using Machine Learning Techniques.

Head-mounted displays are virtual reality devices that may be equipped with sensors and cameras to measure a patient's heart rate through facial regions. Heart rate is an essential body signal that can be used to remotely monitor users in a variety of situations. There is currently no study that predicts heart rate using only highlighted facial regions; thus, an adaptation is required for beats per minute predictions. Likewise, there are no datasets containing only the eye and lower face regions, necessitating the development of a simulation mechanism. This work aims to remotely estimate heart rate from facial regions that can be captured by the cameras of a head-mounted display using state-of-the-art EVM-CNN and Meta-rPPG techniques. We developed a region of interest extractor to simulate a dataset from a head-mounted display device using stabilizer and video magnification techniques. Then, we combined support vector machine and FaceMash to determine the regions of interest and adapted photoplethysmography and beats per minute signal predictions to work with the other techniques. We observed an improvement of 188.88% for the EVM and 55.93% for the Meta-rPPG. In addition, both models were able to predict heart rate using only facial regions as input. Moreover, the adapted technique Meta-rPPG outperformed the original work, whereas the EVM adaptation produced comparable results for the photoplethysmography signal.

Zoom-In Neural Network Deep-Learning Model for Alzheimer's Disease Assessments.

Deep neural networks have been successfully applied to generate predictive patterns from medical and diagnostic data. This paper presents an approach for assessing persons with Alzheimer's disease (AD) mild cognitive impairment (MCI), compared with normal control (NC) persons, using the zoom-in neural network (ZNN) deep-learning algorithm. ZNN stacks a set of zoom-in learning units (ZLUs) in a feedforward hierarchy without backpropagation. The resting-state fMRI (rs-fMRI) dataset for AD assessments was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The Automated Anatomical Labeling (AAL-90) atlas, which provides 90 neuroanatomical functional regions, was used to assess and detect the implicated regions in the course of AD. The features of the ZNN are extracted from the 140-time series rs-fMRI voxel values in a region of the brain. ZNN yields the three classification accuracies of AD versus MCI and NC, NC versus AD and MCI, and MCI versus AD and NC of 97.7%, 84.8%, and 72.7%, respectively, with the seven discriminative regions of interest (ROIs) in the AAL-90.

Assessing the Bone Age of Children in an Automatic Manner Newborn to 18 Years Range.

Bone age assessment (BAA) is a radiological process to identify the growth disorders in children. Although this is a frequent task for radiologists, it is cumbersome. The objective of this study is to assess the bone age of children from newborn to 18 years old in an automatic manner through computer vision methods including histogram of oriented gradients (HOG), local binary pattern (LBP), and scale invariant feature transform (SIFT). Here, 442 left-hand radiographs are applied from the University of Southern California (USC) hand atlas. In this experiment, for the first time, HOG-LBP-dense SIFT features with background subtraction are applied to assess the bone age of the subject group. For this purpose, features are extracted from the carpal and epiphyseal regions of interest (ROIs). The SVM and 5-fold cross-validation are used for classification. The accuracy of female radiographs is 73.88% and of the male is 68.63%. The mean absolute error is 0.5 years for both genders' radiographs. The accuracy a within 1-year range is 95.32% for female and 96.51% for male radiographs. The accuracy within a 2-year range is 100% and 99.41% for female and male radiographs, respectively. The Cohen's kappa statistical test reveals that this proposed approach, Cohen's kappa coefficients are 0.71 for female and 0.66 for male radiographs, p value < 0.05, is in substantial agreement with the bone age assessed by experienced radiologists within the USC dataset. This approach is robust and easy to implement, thus, qualified for computer-aided diagnosis (CAD). The reduced processing time and number of ROIs facilitate BAA.

Permutation Entropy and Statistical Complexity in Mild Cognitive Impairment and Alzheimer's Disease: An Analysis Based on Frequency Bands.

We present one of the first applications of Permutation Entropy (PE) and Statistical Complexity (SC) (measured as the product of PE and Jensen-Shanon Divergence) on Magnetoencephalography (MEG) recordings of 46 subjects suffering from Mild Cognitive Impairment (MCI), 17 individuals diagnosed with Alzheimer's Disease (AD) and 48 healthy controls. We studied the differences in PE and SC in broadband signals and their decomposition into frequency bands ( δ , θ , α and ß ), considering two modalities: (i) raw time series obtained from the magnetometers and (ii) a reconstruction into cortical sources or regions of interest (ROIs). We conducted our analyses at three levels: (i) at the group level we compared SC in each frequency band and modality between groups; (ii) at the individual level we compared how the [PE, SC] plane differs in each modality; and (iii) at the local level we explored differences in scalp and cortical space. We recovered classical results that considered only broadband signals and found a nontrivial pattern of alterations in each frequency band, showing that SC does not necessarily decrease in AD or MCI.

A Regions of Interest Voxel-Based Morphometry Study of the Human Brain During High-Frequency Spinal Cord Stimulation in Patients With Failed Back Surgery Syndrome.

INTRODUCTION: The effectiveness of spinal cord stimulation (SCS) as pain-relieving treatment for failed back surgery syndrome (FBSS) has already been demonstrated. However, potential structural and functional brain alterations resulting from subsensory SCS are less clear. The aim of this study was to test structural volumetric changes in a priori chosen regions of interest related to chronic pain after 1 month and 3 months of high-frequency SCS in patients with FBSS. METHODS: Eleven patients with FBSS who were scheduled for SCS device implantation were included in this study. All patients underwent a magnetic resonance imaging protocol before SCS device implantation 1 and 3 months after high-frequency SCS. Pain intensity, pain catastrophizing, and sleep quality were also measured. Regions-of-interest voxel-based morphometry was used to explore grey matter volumetric changes over time. Additionally, volumetric changes were correlated with changes in pain intensity, catastrophizing, and sleep quality. RESULTS: Significant decreases were found in volume in the left and right hippocampus over time. More specifically, a significant difference was revealed between volumes before SCS implantation and after 3 months of SCS. Repeated-measures correlations revealed a significant positive correlation between volumetric changes in the left hippocampus and changes in back pain score over time and between volumetric changes in the right hippocampus and changes in back pain score over time. CONCLUSION: In patients with FBSS, high-frequency SCS influences structural brain regions over time. The volume of the hippocampus was decreased bilaterally after 3 months of high-frequency SCS with a positive correlation with back pain intensity.

ROIMCR: a powerful analysis strategy for LC-MS metabolomic datasets.

BACKGROUND: The analysis of LC-MS metabolomic datasets appears to be a challenging task in a wide range of disciplines since it demands the highly extensive processing of a vast amount of data. Different LC-MS data analysis packages have been developed in the last few years to facilitate this analysis. However, most of these strategies involve chromatographic alignment and peak shaping and often associate each "feature" (i.e., chromatographic peak) with a unique m/z measurement. Thus, the development of an alternative data analysis strategy that is applicable to most types of MS datasets and properly addresses these issues is still a challenge in the metabolomics field. RESULTS: Here, we present an alternative approach called ROIMCR to: i) filter and compress massive LC-MS datasets while transforming their original structure into a data matrix of features without losing relevant information through the search of regions of interest (ROIs) in the m/z domain and ii) resolve compressed data to identify their contributing pure components without previous alignment or peak shaping by applying a Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) analysis. In this study, the basics of the ROIMCR method are presented in detail and a detailed description of its implementation is also provided. Data were analyzed using the MATLAB (The MathWorks, Inc., www.mathworks.com ) programming and computing environment. The application of the ROIMCR methodology is described in detail, with an example of LC-MS data generated in a lipidomic study and with other examples of recent applications. CONCLUSIONS: The methodology presented here combines the benefits of data filtering and compression based on the searching of ROI features, without the loss of spectral accuracy. The method has the benefits of the application of the powerful MCR-ALS data resolution method without the necessity of performing chromatographic peak alignment or modelling. The presented method is a powerful alternative to other existing data analysis approaches that do not use the MCR-ALS method to resolve LC-MS data. The ROIMCR method also represents an improved strategy compared to the direct applications of the MCR-ALS method that use less-powerful data compression strategies such as binning and windowing. Overall, the strategy presented here confirms the usefulness of the ROIMCR chemometrics method for analyzing LC-MS untargeted metabolomics data.

Analyzing brain structural differences associated with categories of blood pressure in adults using empirical kernel mapping-based kernel ELM.

BACKGROUND: Hypertension increases the risk of angiocardiopathy and cognitive disorder. Blood pressure has four categories: normal, elevated, hypertension stage 1 and hypertension stage 2. The quantitative analysis of hypertension helps determine disease status, prognosis assessment, guidance and management, but is not well studied in the framework of machine learning. METHODS: We proposed empirical kernel mapping-based kernel extreme learning machine plus (EKM-KELM+) classifier to discriminate different blood pressure grades in adults from structural brain MR images. ELM+ is the extended version of ELM, which integrates the additional privileged information about training samples in ELM to help train a more effective classifier. In this work, we extracted gray matter volume (GMV), white matter volume, cerebrospinal fluid volume, cortical surface area, cortical thickness from structural brain MR images, and constructed brain network features based on thickness. After feature selection and EKM, the enhanced features are obtained. Then, we select one feature type as the main feature to feed into KELM+, and the rest of the feature types are PI to assist the main feature to train 5 KELM+ classifiers. Finally, the 5 KELM+ classifiers are ensemble to predict classification result in the test stage, while PI is not used during testing. RESULTS: We evaluated the performance of the proposed EKM-KELM+ method using four grades of hypertension data (73 samples for each grade). The experimental results show that the GMV performs observably better than any other feature types with a comparatively higher classification accuracy of 77.37% (Grade 1 vs. Grade 2), 93.19% (Grade 1 vs. Grade 3), and 95.15% (Grade 1 vs. Grade 4). The most discriminative brain regions found using our method are olfactory, orbitofrontal cortex (inferior), supplementary motor area, etc. CONCLUSIONS: Using region of interest features and brain network features, EKM-KELM+ is proposed to study the most discriminative regions that have obvious structural changes in different blood pressure grades. The discriminative features that are selected using our method are consistent with the existing neuroimaging studies. Moreover, our study provides a potential approach to take effective interventions in the early period, when the blood pressure makes minor impacts on the brain structure and function.

Improved Faster R-CNN Traffic Sign Detection Based on a Second Region of Interest and Highly Possible Regions Proposal Network.

Traffic sign detection systems provide important road control information for unmanned driving systems or auxiliary driving. In this paper, the Faster region with a convolutional neural network (R-CNN) for traffic sign detection in real traffic situations has been systematically improved. First, a first step region proposal algorithm based on simplified Gabor wavelets (SGWs) and maximally stable extremal regions (MSERs) is proposed. In this way, the region proposal a priori information is obtained and will be used for improving the Faster R-CNN. This part of our method is named as the highly possible regions proposal network (HP-RPN). Second, in order to solve the problem that the Faster R-CNN cannot effectively detect small targets, a method that combines the features of the third, fourth, and fifth layers of VGG16 to enrich the features of small targets is proposed. Third, the secondary region of interest method to enhance the feature of detection objects and improve the classification capability of the Faster R-CNN is proposed. Finally, a method of merging the German traffic sign detection benchmark (GTSDB) and Chinese traffic sign dataset (CTSD) databases into one larger database to increase the number of database samples is proposed. Experimental results show that our method improves the detection performance, especially for small targets.

Functional connectivity corresponding to the tonotopic differentiation of the human auditory cortex.

Recent research has demonstrated that resting-state functional connectivity (RS-FC) within the human auditory cortex (HAC) is frequency-selective, but whether RS-FC between the HAC and other brain areas is differentiated by frequency remains unclear. Three types of data were collected in this study, including resting-state functional magnetic resonance imaging (fMRI) data, task-based fMRI data using six pure tone stimuli (200, 400, 800, 1,600, 3,200, and 6,400 Hz), and structural imaging data. We first used task-based fMRI to identify frequency-selective cortical regions in the HAC. Six regions of interest (ROIs) were defined based on the responses of 50 participants to the six pure tone stimuli. Then, these ROIs were used as seeds to determine RS-FC between the HAC and other brain regions. The results showed that there was RS-FC between the HAC and brain regions that included the superior temporal gyrus, dorsolateral prefrontal cortex (DL-PFC), parietal cortex, occipital lobe, and subcortical structures. Importantly, significant differences in FC were observed among most of the brain regions that showed RS-FC with the HAC. Specifically, there was stronger RS-FC between (1) low-frequency (200 and 400 Hz) regions and brain regions including the premotor cortex, somatosensory/-association cortex, and DL-PFC; (2) intermediate-frequency (800 and 1,600 Hz) regions and brain regions including the anterior/posterior superior temporal sulcus, supramarginal gyrus, and inferior frontal cortex; (3) intermediate/low-frequency regions and vision-related regions; (4) high-frequency (3,200 and 6,400 Hz) regions and the anterior cingulate cortex or left DL-PFC. These findings demonstrate that RS-FC between the HAC and other brain areas is frequency selective.

Real-Time Traffic Sign Detection and Recognition Method Based on Simplified Gabor Wavelets and CNNs.

Traffic sign detection and recognition plays an important role in expert systems, such as traffic assistance driving systems and automatic driving systems. It instantly assists drivers or automatic driving systems in detecting and recognizing traffic signs effectively. In this paper, a novel approach for real-time traffic sign detection and recognition in a real traffic situation was proposed. First, the images of the road scene were converted to grayscale images, and then we filtered the grayscale images with simplified Gabor wavelets (SGW), where the parameters were optimized. The edges of the traffic signs were strengthened, which was helpful for the next stage of the process. Second, we extracted the region of interest using the maximally stable extremal regions algorithm and classified the superclass of traffic signs using the support vector machine (SVM). Finally, we used convolution neural networks with input by simplified Gabor feature maps, where the parameters were the same as the detection stage, to classify the traffic signs into their subclasses. The experimental results based on Chinese and German traffic sign databases showed that the proposed method obtained a comparable performance with the state-of-the-art method, and furthermore, the processing efficiency of the whole process of detection and classification was improved and met the real-time processing demands.

Functional Connectivity of Brain Regions According to Resting State fMRI: Differences between Healthy and Depressed Subjects and Variability of the Results.

In depressed patients, changes in spontaneous brain activity, in particular, the strength of functional connectivity between different regions are observed. The data on changes in the synchrony of different regions of interest in the brain can serve as markers of depressive symptoms and as the targets for the corresponding therapy. The study involved 21 patients with mild depression and 21 healthy volunteers; by the time of second fMRI scanning, 15 and 19 subjects, respectively). The subjects underwent two 4-min sessions of resting state fMRI with 2-4 months interval between the recordings; on the basis of these data, functional connectivity between regions of interest was assessed. During the first session, depressed patients demonstrated more pronounced connection between the right frontal eye field and cerebellar area III. When the sample was restricted to subjects who underwent both fMRI sessions, depressed patients demonstrated closer relations of the right parietal operculum and cerebellar vermis area VIII. During the second recording, healthy subjects showed stronger connectivity between more than 20 frontal, temporal, and subcortical regions of interest and cerebellum area II. In healthy participants, brainstem functional interactions increased from the first to the second fMRI-recording. In depressed subjects a number of cortical areas split from left intraparietal sulcus, but the left temporal cortex became more intra-connected. The results confirm the differences in functional connectivity between depressed and healthy subjects. At the same time, attention should be paid to the variability of the data obtained.

Pedestrian Detection with Semantic Regions of Interest.

For many pedestrian detectors, background vs. foreground errors heavily influence the detection quality. Our main contribution is to design semantic regions of interest that extract the foreground target roughly to reduce the background vs. foreground errors of detectors. First, we generate a pedestrian heat map from the input image with a full convolutional neural network trained on the Caltech Pedestrian Dataset. Next, semantic regions of interest are extracted from the heat map by morphological image processing. Finally, the semantic regions of interest divide the whole image into foreground and background to assist the decision-making of detectors. We test our approach on the Caltech Pedestrian Detection Benchmark. With the help of our semantic regions of interest, the effects of the detectors have varying degrees of improvement. The best one exceeds the state-of-the-art.

Combined molecular MRI and immuno-spin-trapping for in vivo detection of free radicals in orthotopic mouse GL261 gliomas.

Free radicals play a major role in gliomas. By combining immuno-spin-trapping (IST) and molecular magnetic resonance imaging (mMRI), in vivo levels of free radicals were detected within mice bearing orthotopic GL261 gliomas. The nitrone spin trap DMPO (5,5-dimethyl pyrroline N-oxide) was administered prior to injection of an anti-DMPO probe (anti-DMPO antibody covalently bound to a bovine serum albumin (BSA)-Gd (gadolinium)-DTPA (diethylene triamine penta acetic acid)-biotin MRI contrast agent) to trap tumor-associated free radicals. mMRI detected the presence of anti-DMPO adducts by either a significant sustained increase (p<0.001) in MR signal intensity or a significant decrease (p<0.001) in T1 relaxation, measured as %T1 change. In vitro assessment of the anti-DMPO probe indicated a significant decrease (p<0.0001) in T1 relaxation in GL261 cells that were oxidatively stressed with hydrogen peroxide, compared to controls. The biotin moiety of the anti-DMPO probe was targeted with fluorescently-labeled streptavidin to locate the anti-DMPO probe in excised brain tissues. As a negative control a non-specific IgG antibody covalently bound to the albumin-Gd-DTPA-biotin construct was used. DMPO adducts were also confirmed in tumor tissue from animals administered DMPO, compared to non-tumor brain tissue. GL261 gliomas were found to have significantly increased malondialdehyde (MDA) protein adducts (p<0.001) and 3-nitrotyrosine (3-NT) (p<0.05) compared to normal mouse brain tissue, indicating increased oxidized lipids and proteins, respectively. Co-localization of the anti-DMPO probe with either 3-NT or 4-hydroxynonenal was also observed. This is the first report regarding the detection of in vivo levels of free radicals from a glioma model.