TL-MSE2-Net: Transfer learning based nested model for cerebrovascular segmentation with aneurysms.

Cerebrovascular (i.e., cerebral vessel) segmentation is essential for diagnosing and treating brain diseases. Convolutional neural network models, such as U-Net, are commonly used for this purpose. Unfortunately, such models may not be entirely satisfactory in dealing with cerebrovascular segmentation with tumors due to the following issues: (1) Relatively small number of clinical datasets from patients obtained through different modalities such as computed tomography (CT) and magnetic resonance imaging (MRI), leading to inadequate training and lack of transferability in the modeling; (2) Insufficient feature extraction caused by less attention to both convolution sizes and cerebral vessel edges. Inspired by the existence of similar features on cerebral vessels between normal subjects and patients, we propose a transfer learning strategy based on a pre-trained nested model called TL-MSE2-Net. This model uses one of the publicly available datasets for cerebrovascular segmentation with aneurysms. To address issue (1), our transfer learning strategy leverages a pre-trained model that uses a large number of datasets from normal subjects, providing a potential solution to the lack of sufficient clinical datasets. To tackle issue (2), we structure the pre-trained model based on 3D U-Net, comprising three blocks: ResMul, DeRes, and REAM. The ResMul and DeRes blocks enhance feature extraction by utilizing multiple convolution sizes to capture multiscale features, and the REAM block increases the weight of the voxels on the edges of the given 3D volume. We evaluated the proposed model on one small private clinical dataset and two publicly available datasets. The experimental results demonstrated that our MSE2-Net framework achieved an average Dice score of 70.81 % and 89.08 % on the two publicly available datasets, outperforming other state-of-the-art methods. Ablation studies were also conducted to validate the effectiveness of each block. The proposed TL-MSE2-Net yielded better results than MSE2-Net on a small private clinical dataset, with increases of 5.52 %, 3.37 %, 6.71 %, and 0.85 % for the Dice score, sensitivity, Jaccard index, and precision, respectively.

Evaluation of the Effects of Myopic Astigmatism Correction and Anterior Corneal Curvature on Functional Optical Zone After SMILE.

PURPOSE: To evaluate the influence of different degrees of myopic astigmatism correction and preoperative anterior corneal curvature on the functional optical zone (FOZ) following small incision lenticule extraction (SMILE). METHODS: In this retrospective study, 68 patients (106 eyes) treated with SMILE were grouped according to myopic astigmatism correction: control (0.00 diopters [D]), moderate astigmatism (-0.50 to -2.00 D), and high astigmatism (> -2.00 D). The FOZ was measured and compared between the three groups for 3 months. Correlations between attempted correction, anterior corneal curvature, corneal aberrations, and the FOZ were analyzed. RESULTS: The preoperative mean treatment spherical equivalent was comparable among the three groups. The average FOZ was 5.06 ± 0.24 mm in the control group, 5.19 ± 0.25 mm in the moderate astigmatism group, and 5.35 ± 0.20 mm in the high astigmatism group The FOZ showed statistically significant differences among the three groups (P < .001), particularly between the high astigmatism group and the other two groups (P < .001 and .018). Correlation analysis showed that the total higher order aberrations, coma, and spherical aberration change were correlated with the FOZ (P < .001). Preoperative steep keratometry, average keratometry, and corneal astigmatism were significantly correlated with the FOZ (P < .05). The correlation remained after excluding the influence of attempted correction on the FOZ (P < .05). After adjusting for other risk factors using multiple linear regression analysis, there was still a significant positive association between preoperative steep keratometry and the FOZ (P < .001). CONCLUSIONS: Patients with higher myopic astigmatism achieved a larger FOZ and less induced horizontal coma than the control and moderate astigmatism groups. A larger FOZ after SMILE can be achieved in eyes with steeper keratometry. [J Refract Surg. 2023;39(2):135-141.].

MuLHiTA: A Novel Multiclass Classification Framework With Multibranch LSTM and Hierarchical Temporal Attention for Early Detection of Mental Stress.

Mental stress is an increasingly common psychological issue leading to diseases such as depression, addiction, and heart attack. In this study, an early detection framework based on electroencephalogram (EEG) data is developed for reducing the risk of these diseases. In existing frameworks, signals are often segmented into smaller sections prior to being input to a deep neural network. However, this approach ignores the fundamental nature of EEG signals as a carrier of valuable information (e.g., the integrity of frequency and phase, and temporal fluctuations of EEG components). As such, this type of segmenting may lead to information loss and a failure to effectively identify mental stress levels. Thus, we propose a novel multiclass classification framework termed multibranch LSTM and hierarchical temporal attention (MuLHiTA) for the early identification of mental stress levels. It specifically focuses on not only intraslice (within each slice) but also interslice (between different slices) samples in parallel. This was achieved by including two complementary branches, each of which integrated a specifically designed attention module into a bidirectional long short-term memory (BLSTM) network, enabling extraction of the most discriminative features from interslice and intraslice EEG signals simultaneously. The outputs of attention modules were then summed to obtain a feature representation that contributes to reduce overfitting and more effective multiclass classification. In addition, electrode positions were optimized using neural activity areas under high-stress conditions, thereby reducing computational costs by minimizing the number of critical electrodes. MuLHiTA was evaluated across one private [Montreal imaging stress task (MIST)] and two publicly available EEG datasets [EEG during mental arithmetic tasks (DMAT) and Simultaneous task EEG workload (STEW)]. These were divided into training and test sets using an 8:2 ratio, and the training data were further divided into training and validation sets using a fivefold cross-validation (CV) method, in which the model with the highest accuracy among the five was selected. The model was trained once more with the full training set, and the test data were then used to evaluate its performance. This approach achieved average classification accuracies of 93.58%, 91.80%, and 99.71% for the MIST, STEW, and DMAT datasets, respectively. Experimental results showed MuLHiTA was superior to state-of-the-art algorithms, including EEGNet, BLSTM, EEGLearn, convolutional neural network (CNN)-long short-term memory (LSTM), and convolutional recurrent attention model (CRAM), for multiclass classification. This demonstrates the viability of MuLHiTA for the early detection of mental stress.

Mechanism of piR-1245/PIWI-like protein-2 regulating Janus kinase-2/signal transducer and activator of transcription-3/vascular endothelial growth factor signaling pathway in retinal neovascularization.

Inhibiting retinal neovascularization is the optimal strategy for the treatment of retina-related diseases, but there is currently no effective treatment for retinal neovascularization. P-element-induced wimpy testis (PIWI)-interacting RNA (piRNA) is a type of small non-coding RNA implicated in a variety of diseases. In this study, we found that the expression of piR-1245 and the interacting protein PIWIL2 were remarkably increased in human retinal endothelial cells cultured in a hypoxic environment, and cell apoptosis, migration, tube formation and proliferation were remarkably enhanced in these cells. Knocking down piR-1245 inhibited the above phenomena. After intervention by a p-JAK2 activator, piR-1245 decreased the expression of hypoxia inducible factor-1α and vascular endothelial growth factor through the JAK2/STAT3 pathway. For in vivo analysis, 7-day-old newborn mice were raised in 75 ± 2% hyperoxia for 5 days and then piR-1245 in the retina was knocked down. In these mice, the number of newly formed vessels in the retina was decreased, the expressions of inflammation-related proteins were reduced, the number of apoptotic cells in the retina was decreased, the JAK2/STAT3 pathway was inhibited, and the expressions of hypoxia inducible factor-1α and vascular endothelial growth factor were decreased. Injection of the JAK2 inhibitor JAK2/TYK2-IN-1 into the vitreous cavity inhibited retinal neovascularization in mice and reduced expression of hypoxia inducible factor-1α and vascular endothelial growth factor. These findings suggest that piR-1245 activates the JAK2/STAT3 pathway, regulates the expression of hypoxia inducible factor-1α and vascular endothelial growth factor, and promotes retinal neovascularization. Therefore, piR-1245 may be a new therapeutic target for retinal neovascularization.

Effect of macular vascular density on visual quality in young myopic adults.

Objective: To evaluate macular vascular density using optical coherence tomography angiography (OCTA) and to investigate its impact on best-corrected visual acuity (BCVA), contrast sensitivity function (CSF), and higher-order aberrations (HOAs) in young myopic adults. Methods: This cross-sectional study included 109 eyes with axial length (AL) between 22 and 26 mm in the medium AL group and 90 eyes with AL > 26 mm in the long AL group. OCTA was used to obtain 3 × 3 mm en face images, and the vessel length density (VLD), perfusion density (PD), and fovea avascular zone (FAZ) of the superficial layer were evaluated. Visual quality was assessed using the CSF and HOAs. Results: Significant differences were found in the inferior VLD, parafoveal PD, and FAZ areas between the groups. AL and macular vascular density showed negative correlations in the inferior and nasal areas. The spherical correction (SE) also showed a positive correlation with vascular density in these two areas. FAZ area and perimeter had a significant negative association with AL, and FAZ circularity was correlated with SE. CSF with bright around 6 and 12 spatial frequencies showed positive correlations with nasal PD. The parafoveal PD showed a significant correlation with BCVA after adjusting for other factors. Conclusion: The superficial macular vascular density of young myopic adults decreased with lower SE and longer AL in the parafovea area. An eye with a long AL has a smaller FAZ, and myopia decreases the FAZ circularity index. The decrease in vessel density could contribute to worse BCVA and may be correlated with lower CSF, but not with HOAs.

3D vessel-like structure segmentation in medical images by an edge-reinforced network.

The vessel-like structure in biomedical images, such as within cerebrovascular and nervous pathologies, is an essential biomarker in understanding diseases' mechanisms and in diagnosing and treating diseases. However, existing vessel-like structure segmentation methods often produce unsatisfactory results due to challenging segmentations for crisp edges. The edge and nonedge voxels of the vessel-like structure in three-dimensional (3D) medical images usually have a highly imbalanced distribution as most voxels are non-edge, making it challenging to find crisp edges. In this work, we propose a generic neural network for the segmentation of the vessel-like structures in different 3D medical imaging modalities. The new edge-reinforced neural network (ER-Net) is based on an encoder-decoder architecture. Moreover, a reverse edge attention module and an edge-reinforced optimization loss are proposed to increase the weight of the voxels on the edge of the given 3D volume to discover and better preserve the spatial edge information. A feature selection module is further introduced to select discriminative features adaptively from an encoder and decoder simultaneously, which aims to increase the weight of edge voxels, thus significantly improving the segmentation performance. The proposed method is thoroughly validated using four publicly accessible datasets, and the experimental results demonstrate that the proposed method generally outperforms other state-of-the-art algorithms for various metrics.

E3 ligase RNF5 inhibits type I interferon response in herpes simplex virus keratitis through the STING/IRF3 signaling pathway.

Herpes simplex keratitis (HSK), caused by the herpes simplex virus 1 (HSV-1), is a major blinding disease in developed countries. HSV-1 can remain latent in the host for life and cannot be eradicated. The infection causes the secretion of various cytokines and aggregation of inflammatory cells. In the early stage of inflammation, mainly neutrophils infiltrate the cornea, and CD4+ T cells mediate the immunopathological changes in herpetic stromal keratitis in the subsequent progression. The STING/IRF3-mediated type I interferon (IFN) response can effectively inhibit viral replication and control infection, but the activity of STING is affected by various ubiquitination modifications. In this study, we found that the expression of RNF5 was elevated in corneal tissues and corneal epithelial cells after infection with HSV-1. Immunofluorescence staining confirmed that RNF5 was mainly expressed in the corneal epithelial layer. We silenced and overexpressed RNF5 expression in corneal epithelial cells and then inoculated them with HSV-1. We found that the expressions of STING, p-IRF3, p-TBK1, and IFN-ß mRNA increased after RNF5 silencing. The opposite results were obtained after RNF5 overexpression. We also used siRNA to silence RNF5 in the mouse cornea and then established the HSK model. Compared with the siRNA-control group, the siRNA-RNF5 group showed significantly improved corneal inflammation, reduced clinical scores and tear virus titers, and significantly increased corneal IFN-ß expression. In addition, the expressions of the proinflammatory cytokines IL-6 and TNF-α in the corneal tissue were significantly decreased, indicating that RNF5 silencing could effectively promote IFN-I expression, inhibit virus replication, alleviate inflammation, and reduce corneal inflammatory damage. In summary, our results suggest that RNF5 limits the type I IFN antiviral response in HSV corneal epithelitis by inhibiting STING/IRF3 signaling.

A nested parallel multiscale convolution for cerebrovascular segmentation.

PURPOSE: Cerebrovascular segmentation in magnetic resonance imaging (MRI) plays an important role in the diagnosis and treatment of cerebrovascular diseases. Many segmentation frameworks based on convolutional neural networks (CNNs) or U-Net-like structures have been proposed for cerebrovascular segmentation. Unfortunately, the segmentation results are still unsatisfactory, particularly in the small/thin cerebrovascular due to the following reasons: (1) the lack of attention to multiscale features in encoder caused by the convolutions with single kernel size; (2) insufficient extraction of shallow and deep-seated features caused by the depth limitation of transmission path between encoder and decoder; (3) insufficient utilization of the extracted features in decoder caused by less attention to multiscale features. METHODS: Inspired by U-Net++, we propose a novel 3D U-Net-like framework termed Usception for small cerebrovascular. It includes three blocks: Reduction block, Gap block, and Deep block, aiming to: (1) improve feature extraction ability by grouping different convolution sizes; (2) increase the number of multiscale features in different layers by grouping paths of different depths between encoder and decoder; (3) maximize the ability of decoder in recovering multiscale features from Reduction and Gap block by using convolutions with different kernel sizes. RESULTS: The proposed framework is evaluated on three public and in-house clinical magnetic resonance angiography (MRA) data sets. The experimental results show that our framework reaches an average dice score of 69.29%, 87.40%, 77.77% on three data sets, which outperform existing state-of-the-art methods. We also validate the effectiveness of each block through ablation experiments. CONCLUSIONS: By means of the combination of Inception-ResNet and dimension-expanded U-Net++, the proposed framework has demonstrated its capability to maximize multiscale feature extraction, thus achieving competitive segmentation results for small cerebrovascular.

Alterations of Regional Homogeneity in Preschool Boys With Autism Spectrum Disorders.

OBJECTIVES: The study was aimed at investigating the alterations of local spontaneous brain activity in preschool boys with autism spectrum disorders (ASD). METHODS: Based on regional homogeneity (ReHo), the acquired resting state functional magnetic resonance imaging (fMRI) data sets, which included 86 boys with ASD and 54 typically developing (TD) boys, were used to detect regional brain activity. Pearson correlation analysis was used to study the relationship between abnormal ReHo value and the Childhood Autism Rating Scale (CARS), Autism Behavior Checklist (ABC), developmental quotient, and age. RESULTS: In the ASD group, we found increased ReHo in the right calcarine as well as decreased ReHo in the opercular part of the left inferior frontal gyrus, the left middle temporal gyrus, the left angular gyrus, and the right medial orbital frontal cortex (p < 0.05, false discovery rate correction). We did not find a correlation between the results of brain regions and the CARS, ABC, and age. CONCLUSIONS: Our study found spontaneous activity changes in multiple brain regions, especially the visual and language-related areas of ASD, that may help to further understand the clinical characteristics of boys with ASD.

TRIM21 Aggravates Herpes Simplex Virus Epithelial Keratitis by Attenuating STING-IRF3-Mediated Type I Interferon Signaling.

Herpes simplex virus-1 (HSV-1) is the leading cause of infectious blindness in the developed world. HSV-1 infection can occur anywhere in the eye, and the most common presentation is epithelial keratitis. In the HSV epithelial keratitis mice model, we detected the expression of TRIM21 and then investigated the clinical relationship between TRIM21 and HSV epithelial keratitis by silencing TRIM21. Through the clinical scores and histopathology examination, we found that TRIM21 can effectively reduce the severity of HSV epithelial keratitis. Furthermore, silencing TRIM21 significantly controlled the virus particle release at 1, 3, and 5 days post-HSV-1 infection. Notably, the production of IFN-ß was enhanced, and the secretion of pro-inflammatory cytokines (IL-6 and TNF-a) was inhibited. Next, human corneal epithelial cells were pretreated with lentivirus or siRNA, respectively, so that TRIM21 expression was overexpressed or silenced. We focused on the regulation of STING-IRF3 and type I interferon signaling after infected with HSV-1. In conclusion, our results have identified that TRIM21 is abnormally high expressed in HSV epithelial keratitis. TRIM21 enhances the replication of HSV-1 in corneal epithelial cells via suppressing the production of type I IFN by inhibiting STING/IRF3 signaling. It also promotes the secretion of IL-6 and TNF-a, thereby aggravating the severity of HSV epithelial keratitis.

Blockade of IL-27 signaling ameliorates herpes stromal keratitis with upregulated CD4+ Foxp3+ regulatory T cells influx in mice.

Purpose: The purpose of this study was to investigate the production of IL-27 p28 and EBI3 in the ocular inflammatory sites, and the role of IL-27 signaling in a model of HSV-1 induced herpetic stromal keratitis (HSK). Methods: The BALB/c mice were injected intraperitoneally (24 h before infection) with anti-IL-27 antibody or IgG antibody as control, infected with HSV-1 via corneal scarification, and then injected intraperitoneally with anti-IL-27 antibody or IgG antibody at 1, 3, and 5 days postinfection. Slit lamp and histopathology were used to assess disease outcome. The levels of IL-27 p28 and EBI3 in corneas were determined by western blotting and immunofluorescence. Furthermore, viral titers were determined, and immune cell infiltrates were collected and analyzed by flow cytometry. Results: We found that the levels of IL-27 p28 and EBI3 in corneas were elevated significantly at the peak of HSK, and both of them were expressed simultaneously in the epithelium, stroma, and endothelium of corneas. In the group of anti-IL-27 treatment, the severity of the corneal lesion and CD4+ T cells infiltration were significantly decreased, and the percentage of CD4+ Foxp3+ Tregs was upregulated markedly in the spleen, DLNs and cornea of HSK mice compared to IgG treatment. Conclusion: These results provided evidence that IL-27 as a pathogenic pro-inflammatory cytokine controlled CD4+ Foxp3+ Tregs production in HSK, which ultimately resulted in promoting the progression of HSK and poor prognosis.

A failed top-down control from the prefrontal cortex to the amygdala in generalized anxiety disorder: Evidence from resting-state fMRI with Granger causality analysis.

In generalized anxiety disorder (GAD), abnormal top-down control from the prefrontal cortex (PFC) to the amygdala is a widely accepted hypothesis through which an "emotional dysregulation model" may be explained. However, whether and how the PFC directly exerts abnormal top-down control on the amygdala remains largely unknown. We aimed to investigate the amygdala-based effective connectivity by using Granger causality analysis (GCA). Thirty-five drug-naive patients with GAD and thirty-six healthy controls (HC) underwent resting-state functional MR imaging. We used seed-based Granger causality analysis to examine the effective connectivity between the bilateral amygdala and the whole brain. The amygdala-based effective connectivity was compared between the HC and GAD groups. The results showed that, in the HC group, the left middle frontal gyrus exerted an inhibitory influence on the right amygdala, while in the GAD group, this influence was disrupted (single voxel P < 0.001, Gaussian random field corrected with P < 0.01). Our findings support and advance the "insufficient top-down control" hypothesis by identifying a failed top-down control from the prefrontal cortex to the amygdala in GAD.

Dynamic functional abnormalities in generalized anxiety disorders and their increased network segregation of a hyperarousal brain state modulated by insomnia.

BACKGROUND: Insomnia is frequently accompanied by the generalized anxiety disorder (GAD) but mostly fMRI studies investigated their aberrant functional connectivity (FC) without this issue. Recently, dynamic FC approach is prevailing to capture the time-varying fluctuations of spontaneous brain activities. Nevertheless, it is unclear how the dynamic FC characteristics are altered by insomnia in GAD. METHODS: We acquired resting state fMRI and neuropsychological tests for the 17 comorbid GAD with insomnia (GAD/IS), 15 GAD and 24 healthy controls (HC). Then, based on the sliding window correlations, we estimated distinct brain states and statistically compared their dynamic properties. Further combining with graph theory, their network properties of each state among groups were accessed. Lastly, we examined associations between abnormal parameters and neuropsychological tests. RESULTS: We identified four brain states but did not observe significance on the state transitions. The mean dwell time and fraction of one globally hypoactive state accounted for high proportion of brain activities were significantly different (GAD > HC > GAD/IS). Meanwhile, we found gradual decreases in a brain state representing slight sleep/drowsiness (HC > GAD/IS > GAD). Additionally, we observed the GAD/IS patients had significantly increased network segregation and posterior cingulate cortex in a hyperarousal state, as well as significant associations with anxiety and insomnia severity. LIMITATIONS: The influences of depression on dynamic FC properties in GAD are unclear yet and more subjects should be recruited. CONCLUSIONS: These results provide new insights about the temporal features in GAD and offer potential biomarkers to evaluate the impacts of insomnia.

Three-year results of small incision lenticule extraction and wavefront-guided femtosecond laser-assisted laser in situ keratomileusis for correction of high myopia and myopic astigmatism.

AIM: To compare and calculate the 3-year refractive results, higher-order aberrations (HOAs), contrast sensitivity (CS) and dry eye parameters after small incision lenticule extraction (SMILE) and wavefront-guided femtosecond laser-assisted laser in situ keratomileusis (FS-LASIK) for correction of high myopia and myopic astigmatism. METHODS: In this prospective, non-randomized comparative study, 78 eyes with spherical equivalent (SE) of -8.11±1.09 diopters (D) received a SMILE surgery, and 65 eyes with SE of -8.05±1.12 D received a wavefront-guided FS-LASIK surgery with the VisuMax femtosecond laser (Carl Zeiss Meditec, Jena, Germany) for flap cutting. Visual acuity, manifest refraction, CS, HOAs, ocular surface disease index (OSDI) and tear break-up time (TBUT) were evaluated during a 3-year follow-up. RESULTS: The difference of uncorrected distance visual acuity (UDVA) postoperatively was achieved at 1mo and at 3mo, whereas the difference of the mean UDVA between two groups at 3y were not statistically significant (t=-1.59, P=0.13). The postoperative change of SE was 0.89 D in the FS-LASIK group (t=5.76, P=0.00), and 0.14 D in the SMILE group (t=0.54, P=0.59) from 1mo to 3y after surgery. At 3-year postoperatively, both HOAs and spherical aberrations in the SMILE group were obviously less than those in the FS-LASIK group (P=0.00), but the coma root mean square (RMS) was higher in the SMILE group (0.59±0.26) than in the FS-LASIK group (0.29±0.14, P=0.00). The mesopic CS values between two groups were not statistically significant at 3y postoperatively. Compared with the FS-LASIK group, lower OSDI scores and longer TBUT values were found in the SMILE group at 1mo and 3mo postoperatively. With regard to safety, no eye lost any line of CDVA in both groups at 3y after surgery. CONCLUSION: Both SMILE and wavefront-guided FS-LASIK procedures provide good visual outcomes. Both procedures are effective and safe, but SMILE surgery achieve more stable long-term refractive outcome and better control of early postoperative dry eye as compared to FS-LASIK.

Spontaneous alterations of regional brain activity in patients with adult generalized anxiety disorder.

OBJECTIVE: We aimed to examine how spontaneous brain activity might be related to the pathophysiology of generalized anxiety disorder (GAD). PATIENTS AND METHODS: Using resting-state functional MRI, we examined spontaneous regional brain activity in 31 GAD patients (mean age, 36.87±9.16 years) and 36 healthy control participants (mean age, 39.53±8.83 years) matched for age, education, and sex from December 2014 to October 2015. We performed a two-sample t-test on the voxel-based analysis of the regional homogeneity (ReHo) maps. We used Pearson correlation analysis to compare scores from the Hamilton Anxiety Rating Scale, Hamilton Depression Rating Scale, State-Trait Anxiety Scale-Trait Scale, and mean ReHo values. RESULTS: We found abnormal spontaneous activity in multiple regions of brain in GAD patients, especially in the sensorimotor cortex and emotional regions. GAD patients showed decreased ReHo values in the right orbital middle frontal gyrus, left anterior cingulate cortex, right middle frontal gyrus, and bilateral supplementary motor areas, with increased ReHo values in the left middle temporal gyrus, left superior temporal gyrus, and right superior occipital gyrus. The ReHo value of the left middle temporal gyrus correlated positively with the Hamilton Anxiety Rating Scale scores. CONCLUSION: These results suggest that altered local synchronization of spontaneous brain activity may be related to the pathophysiology of GAD.

An EEG-based machine learning method to screen alcohol use disorder.

Screening alcohol use disorder (AUD) patients has been challenging due to the subjectivity involved in the process. Hence, robust and objective methods are needed to automate the screening of AUD patients. In this paper, a machine learning method is proposed that utilized resting-state electroencephalography (EEG)-derived features as input data to classify the AUD patients and healthy controls and to perform automatic screening of AUD patients. In this context, the EEG data were recorded during 5 min of eyes closed and 5 min of eyes open conditions. For this purpose, 30 AUD patients and 15 aged-matched healthy controls were recruited. After preprocessing the EEG data, EEG features such as inter-hemispheric coherences and spectral power for EEG delta, theta, alpha, beta and gamma bands were computed involving 19 scalp locations. The selection of most discriminant features was performed with a rank-based feature selection method assigning a weight value to each feature according to a criterion, i.e., receiver operating characteristics curve. For example, a feature with large weight was considered more relevant to the target labels than a feature with less weight. Therefore, a reduced set of most discriminant features was identified and further be utilized during classification of AUD patients and healthy controls. As results, the inter-hemispheric coherences between the brain regions were found significantly different between the study groups and provided high classification efficiency (Accuracy = 80.8, sensitivity = 82.5, and specificity = 80, F-Measure = 0.78). In addition, the power computed in different EEG bands were found significant and provided an overall classification efficiency as (Accuracy = 86.6, sensitivity = 95, specificity = 82.5, and F-Measure = 0.88). Further, the integration of these EEG feature resulted into even higher results (Accuracy = 89.3 %, sensitivity = 88.5 %, specificity = 91 %, and F-Measure = 0.90). Based on the results, it is concluded that the EEG data (integration of the theta, beta, and gamma power and inter-hemispheric coherence) could be utilized as objective markers to screen the AUD patients and healthy controls.

Increased insular connectivity with emotional regions in primary insomnia patients: a resting-state fMRI study.

OBJECTIVE: To explore the abnormal connectivity patterns between the insular and the voxels of the brain in primary insomnia (PI) with insular-based functional connectivity (FC). METHODS: With the resting-state fMRI data acquired from 57 PI patients and 46 healthy controls, a two-sample t test was performed on individual FC correlation maps from two groups. The person correlation analysis was used to evaluate the relationship between the abnormal FC and clinical features. RESULTS: PI patients show enhanced connectivity between the left insula with the right anterior cingulate cortex (p < 0.05 and p < 0.001, AlphaSim-corrected), right frontal sup orb, bilateral thalamus and left precuneus,as well as decreased connectivity with the left middle temporal gyrus and right fusiform (p < 0.05, AlphaSim-corrected). Correlation analysis indicated the enhanced connectivities in the PI patients have significant negative correlations with Self-Rating Depression Scale(SDS)and Self-Rating Anxiety Scale(SAS)scores. In addition, the decreased functional connectivities showed positive correlations with SDS and SAS scores. CONCLUSION: Our study showed the increased connectivity regions with insula were mainly in the emotional circle and decreased connectivity was in cognitive-related regions. These provide additional evidence from functional integration view to understand the possible underlying neural- mechanisms of PI. KEY POINTS: • The aberrant insular-based connectivity pattern of PI patients was detected. • Regions showing increased connectivity with left insular were mainly in emotional circle. • Significant correlations between changed FC and SDS and SAS score were found.

A wavelet-based technique to predict treatment outcome for Major Depressive Disorder.

Treatment management for Major Depressive Disorder (MDD) has been challenging. However, electroencephalogram (EEG)-based predictions of antidepressant's treatment outcome may help during antidepressant's selection and ultimately improve the quality of life for MDD patients. In this study, a machine learning (ML) method involving pretreatment EEG data was proposed to perform such predictions for Selective Serotonin Reuptake Inhibitor (SSRIs). For this purpose, the acquisition of experimental data involved 34 MDD patients and 30 healthy controls. Consequently, a feature matrix was constructed involving time-frequency decomposition of EEG data based on wavelet transform (WT) analysis, termed as EEG data matrix. However, the resultant EEG data matrix had high dimensionality. Therefore, dimension reduction was performed based on a rank-based feature selection method according to a criterion, i.e., receiver operating characteristic (ROC). As a result, the most significant features were identified and further be utilized during the training and testing of a classification model, i.e., the logistic regression (LR) classifier. Finally, the LR model was validated with 100 iterations of 10-fold cross-validation (10-CV). The classification results were compared with short-time Fourier transform (STFT) analysis, and empirical mode decompositions (EMD). The wavelet features extracted from frontal and temporal EEG data were found statistically significant. In comparison with other time-frequency approaches such as the STFT and EMD, the WT analysis has shown highest classification accuracy, i.e., accuracy = 87.5%, sensitivity = 95%, and specificity = 80%. In conclusion, significant wavelet coefficients extracted from frontal and temporal pre-treatment EEG data involving delta and theta frequency bands may predict antidepressant's treatment outcome for the MDD patients.

Feature Learning Based Random Walk for Liver Segmentation.

Liver segmentation is a significant processing technique for computer-assisted diagnosis. This method has attracted considerable attention and achieved effective result. However, liver segmentation using computed tomography (CT) images remains a challenging task because of the low contrast between the liver and adjacent organs. This paper proposes a feature-learning-based random walk method for liver segmentation using CT images. Four texture features were extracted and then classified to determine the classification probability corresponding to the test images. Seed points on the original test image were automatically selected and further used in the random walk (RW) algorithm to achieve comparable results to previous segmentation methods.

Efficacy, safety, predictability, aberrations and corneal biomechnical parameters after SMILE and FLEx: Meta-analysis.

AIM: To identify possible differences of efficacy, safety, predictability, higher-order aberrations and corneal biomechnical parameters after small-incision lenticule extraction (SMILE) and femtosecond lenticule extraction (FLEx). METHODS: A systematic literature retrieval was conducted in Medline, Embase and the Cochrane Library, up to October, 2015. The included studies were subject to a Meta-analysis. Comparison between SMILE and FLEx was measured as pooled odds ratio (OR) or weighted mean differences (WMD). Of 95% confidence intervals (CI) were used to analyze data. RESULTS: A total of seven studies were included. Firstly, there were no differences in uncorrected distance visual acuity (UDVA) 20/20 or better (OR, 1.37; 95% CI, 0.69 to 2.69; P=0.37) and logMAR UDVA (WMD, -0.02; 95% CI, -0.05 to 0.01; P=0.17) after SMILE versus FLEx. We found no differences in corrected distance visual acuity (CDVA) unchanged (OR, 0.98; 95% CI, 0.46 to 2.11; P=0.97) and logMAR CDVA (WMD, -0.00; 95% CI, -0.01 to 0.01; P=0.90) either. Secondly, we found no differences in refraction within ±1.00 D (OR, 0.98; 95% CI, 0.13 to 7.28; P=0.99) and ±0.50 D (OR, 1.62; 95% CI, 0.62 to 4.28; P=0.33) of target postoperatively. Thirdly, for higher-order aberrations, we found no differences in the total higher-order aberrations (WMD, -0.04; 95% CI, -0.09 to 0.01; P=0.14), coma (WMD, -0.04; 95% CI, -0.09 to 0.01; P=0.11), spherical (WMD, 0.01; 95% CI, -0.02 to 0.03; P=0.60) and trefoil (WMD, -0.00; 95% CI, -0.04 to 0.03; P=0.76). Furthermore, for corneal biomechanical parameters, we also found no differences (WMD, 0.08; 95% CI, -0.17 to 0.33; P=0.54) after SMILE versus FLEx. CONCLUSION: There are no statistically differences in efficacy, safety, predictability, higher-order aberrations and corneal biomechnical parameters postoperative between SMILE and FLEx.