Search

The putamen has been implicated in depressive disorders, but how its structure and function increase depression risk is not clearly understood. Here, we examined how putamen volume, neuronal density, and mood-modulated functional activity relate to family history and prospective course of depression.The study includes 115 second- and third-generation offspring at high or low risk for depression based on the presence or absence of major depressive disorder in the first generation. Offspring were followed longitudinally using semistructured clinical interviews blinded to their familial risk; putamen structure, neuronal integrity, and functional activation were indexed by structural magnetic resonance imaging (MRI), proton magnetic resonance spectroscopy (N-acetylaspartate/creatine ratio), and functional MRI activity modulated by valence and arousal components of a mood induction task, respectively.After adjusting for covariates, the high-risk individuals had lower putamen volume (standardized betas, β-Findings demonstrate abnormalities in putamen structure and function in individuals at high risk for major depressive disorder. Future studies should focus on this region as a potential biomarker for depressive illness, noting meanwhile that differences attributable to family history may peak at different ages based on which MRI modality is being used to assay them.

BACKGROUND: Obsessive-compulsive disorder (OCD) implicates alterations in cortico-striato-thalamo-cortical and fronto-limbic circuits. Building on prior structural findings, this is the largest study to date examining subcortical surface morphometry in OCD. METHODS: Structural magnetic resonance imaging data were collected from 200 participants across development (5–55 years): 28 youth and 75 adults with OCD and 27 psychiatrically healthy youth and 70 adults. General linear models were used to assess group differences and group-by-age interactions on subcortical shape (FSL FIRST). RESULTS: Compared to healthy participants, those with OCD exhibited surface expansions on the right nucleus accumbens and inward left amygdala deformations, which were associated with greater OCD symptom severity ([Children's] Yale-Brown Obsessive-Compulsive Scale). Group-by-age interactions indicated that accumbens group differences were driven by younger participants and that right pallidum shape was associated inversely with age in healthy participants, but not in participants with OCD. No differences in the shape of other subcortical regions or in volumes (FreeSurfer) were detected in supplementary analyses. CONCLUSIONS: This study is the largest to date examining subcortical shape in OCD and the first to do so across the developmental spectrum. NAcc and amygdala shape deformation builds on extant neuroimaging findings and suggests subtle, subregional alterations beyond volumetric findings. Results shed light on morphometric alterations in OCD, informing current pathophysiological models.

Purpose Our objectives were to assess the abnormalities of subcortical nuclei by combining volume and shape analyses and potential association with cognitive impairment. Patients and methods Twenty-nine patients with severe ACS of the unilateral internal carotid artery and 31 controls were enrolled between January 2017 to August 2018. All participants underwent a comprehensive neuropsychological evaluation, blood lipid biochemical measurements, and structural magnetic resonance imaging (MRI) to measure subcortical volumes and sub-regional shape deformations. Basic statistics, correction for multiple comparisons. Seventeen ACS patients underwent carotid endarterectomy (CEA) within one week after baseline measurements, cognitive assessments and MRI scans were repeated 6 months after CEA. Results The ACS patients had higher apolipoprotein B/apolipoprotein A1 (ApoB/ApoA1) ratio and worse performance in all cognitive domains than controls. Moreover, the ACS patients showed more profound thalamic atrophy assessed by shape and volume analysis, especially in the medial dorsal thalamus. No significant differences were found in other subcortical nuclei after multiple comparisons correction. At baseline, thalamic atrophy correlated with cognitive impairment and ApoB/ApoA1 ratio. Furthermore, mediation analysis at baseline showed that the association of carotid intima-media thickness with executive functioning was mediated by thalamic volume. After CEA, cognitive improvement and increase in the bilateral medial dorsal thalamic volume were observed. Conclusion Our study identified the distinct atrophy of subcortical nuclei and their association with cognition in patients with ACS. Assessments of the thalamus by volumetric and shape analysis may provide an early marker for cerebral ischemia and reperfusion after CEA.

The accurate diagnosis and prediction for individuals is crucial in computer-aided diagnosis of Alzheimer’s disease (AD). The existing structural magnetic resonance imaging based classification methods of AD diagnosis mainly focus on the voxel level, region level and patch level morphological pattern analysis. However, most of these methods extract features with high dimension which may lead to overfitting problem. Besides, the interaction of different patches is not considered in the classifier ensemble. In this article, we propose a novel anatomical landmarks and directed acyclic graph (DAG) network feature learning based classification algorithm for the diagnosis of AD individuals. First, the anatomical feature patches of gray matter image are identified by the morphological and statistical analysis. Second, a simple and efficient DAG convolutional neural network is proposed to extract the discriminative deep features of image representation. Especially, the deep features are obtained by fusing feature maps of different network levels which contain semantic high-level and high-resolution low-level features. Finally, support vector machine and deep features are utilized to construct the classification model and predict the individual of AD. Experiments on three public datasets including ADNI-1, ADNI-2 and MIRIAD demonstrate that the proposed method can effectively improve the classification performance compared with the state-of-the-art methods for AD diagnosis.

Bulimia nervosa (BN) is associated with functional abnormalities in frontostriatal and frontolimbic circuits. Although structural alterations in the frontal portions of these circuits have been observed, this is the first study of subcortical surface morphometry and the largest study of subcortical volume in BN.Anatomical magnetic resonance scans were acquired from 62 female participants with full and subthreshold BN (mean age ± SD, 18.7 ± 4.0 years) and 65 group-matched healthy control participants (mean age ± SD, 19.3 ± 5.7 years). General linear models were used to compare groups and assess the significance of group-by-age interactions on the shape and total volume of 15 subcortical structures (p.05, familywise error corrected). Associations with illness severity and duration were assessed in the BN group.Subcortical volumes did not differ across groups, but vertexwise analyses revealed inward shape deformations on the anterior surface of the pallidum in BN relative to control participants that were associated with binge-eating frequency and illness duration. Inward deformations on the ventrolateral thalamus and dorsal amygdala were more pronounced with advancing age in the BN group, and inward deformations on the caudate, putamen, and amygdala were associated with self-induced vomiting frequency.Our findings point to localized deformations on the surface of subcortical structures in areas that comprise both reward and cognitive control circuits. These deformations were more pronounced among older BN participants and among those with the most severe symptoms. Such precise localization of alterations in subcortical morphometry may ultimately aid in efforts to identify markers of risk and BN persistence.

Previous neuroimaging studies have shown that serotonin–norepinephrine reuptake inhibitor antidepressants alter functional activity in large expanses of brain regions. However, it is not clear how these regions are systemically organized on a connectome level with specific topological properties, which may be crucial to revealing neural mechanisms underlying serotonin–norepinephrine reuptake inhibitor treatment of persistent depressive disorder. To investigate the effect of serotonin–norepinephrine reuptake inhibitor antidepressants on brain functional connectome reconfiguration in persistent depressive disorder and whether this reconfiguration promotes the improvement of clinical symptoms, we combined resting-state functional magnetic resonance imaging (fMRI) scans acquired in two randomized, double-blind, placebo-controlled trial studies of serotonin–norepinephrine reuptake inhibitor antidepressant treatment of patients with persistent depressive disorder. One was a randomized, double-blind, placebo-controlled trial of 10-week duloxetine medication treatment, which included 17 patients in duloxetine group and 17 patients in placebo group (ClinicalTrials.gov Identifier: NCT00360724); the other one was a randomized, double-blind, placebo-controlled trial of 12-week desvenlafaxine medication treatment, which included 16 patients in desvenlafaxine group and 15 patients in placebo group (ClinicalTrials.gov Identifier: NCT01537068). The 24-item Hamilton Depression Rating Scale was used to measure clinical symptoms, and graph theory was employed to examine serotonin–norepinephrine reuptake inhibitor antidepressant treatment effects on the topological properties of whole-brain functional connectome of patients with persistent depressive disorder. We adopted a hierarchical strategy to examine the topological property changes caused by serotonin–norepinephrine reuptake inhibitor antidepressant treatment, calculated their small-worldness, global integration, local segregation and nodal clustering coefficient in turn. Linear regression analysis was used to test associations of treatment, graph properties changes and clinical symptom response. Symptom scores were more significantly reduced after antidepressant than placebo administration (η2 = 0.18). There was a treatment-by-time effect that optimized the functional connectome in a small-world manner, with increased global integration and increased nodal clustering coefficient in the bilateral thalamus (left thalamus η2 = 0.21; right thalamus η2 = 0.23). The nodal clustering coefficient increment of the right thalamus (ratio = 29.86; 95% confidence interval, −4.007 to −0.207) partially mediated the relationship between treatment and symptom improvement, and symptom improvement partially mediated (ratio = 21.21; 95% confidence interval, 0.0243–0.444) the relationship between treatment and nodal clustering coefficient increments of the right thalamus. Our study may indicate a putative mutually reinforcing association between nodal clustering coefficient increment of the right thalamus and symptom improvement from serotonin–norepinephrine reuptake inhibitor antidepressant treatments with duloxetine or desvenlafaxine.

The accuracy of computed tomography angiography (CTA) image interpretation depends on the radiologist. This study aims to develop a new method for automatically detecting intracranial aneurysms from CTA images using deep learning, based on a convolutional neural network (CNN) implemented on the DeepMedic platform. Ninety CTA scans of patients with intracranial aneurysms are collected and divided into two datasets: training (80 subjects) and test (10 subjects) datasets. Subsequently, a deep learning architecture with a three-dimensional (3D) CNN model is implemented on the DeepMedic platform for the automatic segmentation and detection of intracranial aneurysms from the CTA images. The samples in the training dataset are used to train the CNN model, and those in the test dataset are used to assess the performance of the established system. Sensitivity, positive predictive value (PPV), and false positives are evaluated. The overall sensitivity and PPV of this system for detecting intracranial aneurysms from CTA images are 92.3% and 100%, respectively, and the segmentation sensitivity is 92.3%. The performance of the system in the detection of intracranial aneurysms is closely related to their size. The detection sensitivity for small intracranial aneurysms (≤ 3 mm) is 66.7%, whereas the sensitivity of detection for large ( 10 mm) and medium-sized (3-10 mm) intracranial aneurysms is 100%. The deep learning architecture with a 3D CNN model on the DeepMedic platform can reliably segment and detect intracranial aneurysms from CTA images with high sensitivity.

A family history (FH+) of substance use disorder (SUD) increases an adolescent's risk for substance use initiation and progression. Greater impulsivity and reward seeking behavior is known to be associated with such risk. At the neurological level, dysfunction of cortico-striatal and cortico-limbic pathways have been proposed as contributors to the increased SUD risk in adolescents with FH+. In addition, disadvantaged environments have been associated with atypical brain connectivity and higher SUD risk. However, it remains unclear if this increased risk is manifested in structural and functional brain abnormalities prior to regular drug use. To examine this, we employed complementary imaging of structural and functional connectivity of 60 FH+ and 55 FH- minority adolescents, all from families with low socio-economic status. We acquired diffusion tensor-imaging (DTI) and resting state fMRI data across the whole brain. Structural connectivity was examined by measuring fractional anisotropy (FA) using DTI, to indicate integrity of the white matter tracts. Functional connectivity within and between resting state networks was assessed by the correlation of blood-oxygen-level-dependent (BOLD) signal between intra and inter-network nodes. Psychological measures of impulsivity and reward seeking were also obtained with standardized measures, the BIS-11 and the BIS/BAS, and their association with FA and functional connectivity was evaluated. We found no differences in white matter integrity between the groups. Compared to FH-, FH + adolescents showed significantly greater functional connectivity between posterior regions of the Default Mode Network (DMN) and the Fronto-Parietal Network (FPN). While psychological measures of reward seeking behavior did not differ between the FH+ and FH- groups, impulsivity, assessed by the BIS-11, was significantly higher for FH+. However, we did not find significant differences between the FH+ and FH- groups when comparing associations of BIS-11 scores and white matter integrity or functional connectivity measures. The stronger inter-network functional connectivity between the DMN and FPN in FH + adolescents suggests that transmitted risk for SUD may be related to large-scale brain dynamics. The lack of structural differences support the importance of early prevention efforts for FH + adolescents, before initiation of drug use, allowing for healthy brain development.

Obesity can result from excess intake in response to environmental food cues, and stress can drive greater intake and body weight. We used a novel fMRI task to explore how obesity and stress influenced appetitive responses to relatively minimal food cues (words representing food items, presented similarly to a chalkboard menu). Twenty-nine adults (16F, 13M), 17 of whom had obesity and 12 of whom were lean, completed two fMRI scans, one following a combined social and physiological stressor and the other following a control task. A food word reactivity task assessed subjective food approach (wanting) as well as food avoidant (restraint) responses, along with neural responses, to words denoting high energy-density (ED) foods, low-ED foods, and non-foods. A multi-item ad-libitum meal followed each scan. The obese and lean groups demonstrated differences as well as similarities in activation of appetitive and attention/self-regulation systems in response to food vs. non-food, and to high-ED vs. low-ED food words. Patterns of activation were largely similar across stress and non-stress conditions, with some evidence for differences between conditions within both obese and lean groups. The obese group ate more than the lean group in both conditions. Our results suggest that neural responses to minimal food cues in stressed and non-stressed states may contribute to excess consumption and adiposity.

Background Patient expectancy of therapeutic improvement is a primary mediator of placebo effects in antidepressant clinical trials, but its mechanisms are poorly understood. This study employed a novel antidepressant trial design, with integrated functional magnetic resonance imaging (fMRI), to manipulate patient outcome expectancy and examine its neural mediators. Method Twenty-three depressed outpatients, in a randomized controlled trial were assigned to either Open (high outcome expectancy) or Placebo-controlled (low outcome expectancy) treatment with citalopram for eight weeks. fMRI scans were acquired before and after the expectancy manipulation (before medication treatment), while participants performed a masked emotional face task. Focusing on an amygdala region-of-interest (ROI), we tested a model where reduction in amygdala activation mediated outcome expectancy effects on the slope of change in depressive symptoms. Results Following the manipulation, significant differences between conditions were found in neural activation changes in the amygdala, as well as in superior temporal gyrus, insula, and thalamus. Findings support the proposed mediation model according to which activation in the left amygdala ROI decreased significantly in the Open as opposed to the Placebo-controlled group following randomization (p = 0.009) for sad vs. neutral face contrast. The reduced left amygdala activation, in turn, was a significant predictor of decreased depressive symptoms during the trial (p = 0.007), and the mediation model was significant. Conclusions Results from this study, the first designed to identify the neural mechanisms of expectancy augmentation in an antidepressant randomized control trial, suggest that therapeutic modulation of amygdala activity may be an important pathway by which patient outcome expectancy influences depressive symptoms. Clinicaltrials.gov identifier NCT01919216; Trial name: Placebo Effects in the Treatment of Depression: Cognitive and Neural Mechanisms, URL: https://clinicaltrials.gov/ct2/show/NCT01919216

Objective: This study focused on the ADHD classification through functional connectivity (FC) analysis. Method: An ADHD classification method was proposed with subspace clustering and binary hypothesis testing, wherein partial information of test data was adopted for training. By hypothesizing the binary label (ADHD or control) for the test data, two feature sets of training FC data were generated during the feature selection procedure that employed both training and test data. Then, a multi-affinity subspace clustering approach was performed to obtain the corresponding subspace-projected feature sets. With the energy comparison of projected feature sets, we finally identified ADHD individuals for the test data. Results: Our method outperformed several state-of-the-art methods with the above 90% average identification accuracy. By the discriminative FC contribution analysis, it also proved the reliability of our method. Conclusion: Results demonstrate the remarkable classification performance of our method and reveal some useful brain circuits to identify ADHD.

Recently, deep learning has been an area of intense research. However, as a kind of computing-intensive task, deep learning highly relies on the scale of GPU memory, which is usually prohibitive and scarce. Although some extensive works have been proposed for dynamic GPU memory management, they are hard to apply to systems with multiple dynamic workloads, such as in-database machine learning systems. In this paper, we demonstrated TENSILE, a method of managing GPU memory in tensor granularity to reduce the GPU memory peak, considering the multiple dynamic workloads. TENSILE tackled the cold-starting and across-iteration scheduling problem existing in previous works. We implemented TENSILE on a deep learning framework built by ourselves and evaluated its performance. The experiment results show that TENSILE can save more GPU memory with less extra overhead than prior works in single and multiple dynamic workloads scenarios.

Brain gray matter is organized in a manner with interconnected brain regions, resulting in a notable covariance pattern that recapitulates either the functional coactivation or structural connectivity of brain regions, which is believed to underpin psychiatric disorders such as depression. This study aimed to investigate whether and how antidepressants took effect in treating depression and reducing symptoms by altering the gray matter covariance pattern. We combined structural magnetic resonance imaging (MRI) scans acquired in two randomized, double-blind, placebo-controlled trial (RCT) studies of the treatment using serotonin-norepinephrine reuptake inhibitor (SNRI) antidepressant medications in patients with persistent depressive disorder (PDD). One was an RCT of 10-week duloxetine medication that consisted of patients who received duloxetine (N = 21) or placebo (N = 21), and the other was an RCT of 12-week desvenlafaxine medication that consisted of 19 and 17 patients respectively who received desvenlafaxine or placebo. We examined treatment effect on gray matter volume (GMV) and topological organization of GMV covariance pattern (i.e., GMV-based network). We found a treatment-by-time effect on GMV in the angular gyrus and cuneus areas, whereas the GMV change rate of the cuneus was inversely correlated with the response rate. We observed a significant increase in the local efficiency of the GMV-based network following medication treatment compared with placebo. Our findings provide preliminary evidence for a GMV-based network-specific reconfiguration caused by antidepressants compared to placebo and the cuneus may be a possible candidate region to predict antidepressant response.

Background Major depressive disorder (MDD) is associated with aberrant limbic neural responses to emotional stimuli. We assessed how self-generated emotions modulate trial-by-trial limbic activity and whether this brain-emotion synchrony varies by familial MDD risk (regardless of personal MDD history) and neuroticism. Methods Participants (n = 74, mean age = 34 years) were later-generation family members of depressed or nondepressed probands as part of a longitudinal cohort study. Using an emotion induction task, we examined participant-specific modulation of anatomically defined limbic neurobiology. Neuroticism, mental health, and familial parenting style were assessed, and MDD assessments were routinely collected throughout the previous longitudinal assessments of the study. Results Participant-specific emotional arousal modulated amygdala and hippocampal activity. Lasso regression identified attenuated right amygdala arousal modulation as being relatively more associated with neuroticism (even though neuroticism was not associated with arousal ratings). Attenuated amygdala modulation and neuroticism were significantly more likely in offspring of parents with MDD. Parental MDD, but not personal history of MDD, predicted attenuated amygdala modulation. Conclusions Attenuated right amygdala modulation by emotional arousal was associated with neuroticism, indicating that the amygdala was less synchronous with emotional experiences in individuals higher in neuroticism. This neurophenotype was predicted by participants’ parental MDD history but not by their own MDD history; that is, it was observed in unaffected and affected offspring of parents with MDD. These data suggest that weak amygdala-emotion synchrony may be a predisposing risk factor for MDD, rather than a result of the illness, and they suggest pathways by which this risk factor for depression is passed intergenerationally.

Objective: The purpose of this study was to examine the neural substrates and mechanisms that generate memory deficits, seizures and neuropsychiatric abnormalities in encephalitis with LGI1 antibodies using a data-driven, multimodal magnetic resonance imaging (MRI) approach. Methods: Functional MRI data were acquired from 14 anti-LGI1 encephalitis patients and 14 age and gender matched normal controls. Independent component analysis with hierarchical partner matching (HPM-ICA) was used to assess the whole-brain intrinsic functional connectivity. Granger causality (GC) was applied to investigate the effective connectivity among the brain regions that identified by HPM-ICA. Diffusion tensor imaging (DTI) was utilized to investigate white matter microstructural changes of the patients. Results: Participants with LGI1 antibodies encephalitis presented reduced functional connectivity in the brain areas associated with memory, cognition and motion circuits, while increased functional connectivity in putamen and caudate in comparison to the normal controls. Moreover, the effective connectivity in patients was decreased from the frontal cortex to supplementary motor area. Finally, patients had significant reductions in fractional anisotropy (FA) for the corpus callosum, internal capsule, corona radiata and superior longitudinal fasciculus, accompanied by increases in mean diffusivity (MD) for these regions as compared to controls. Conclusion: Our findings suggest that the neural disorder and behavioral deficits of anti-LGI1 encephalitis may be associated with extensive changes in brain connectivity and microstructure. These pathological alterations affect the basal ganglia and limbic system besides the temporal and frontal lobe.

The term perceptual closure refers to the neural processes responsible for “filling-in” missing information in the visual image under highly adverse viewing conditions such as fog or camouflage. Here we used a closure task that required the participants to identify barely recognizable fragmented line-drawings of common objects. Patients with schizophrenia have been shown to perform poorly on this task. Following priming, controls and importantly patients can complete the line-drawings at greater levels of fragmentation behaviorally, suggesting an improvement in their ability to perform the task. Closure phenomena have been shown to involve a distributed network of cortical regions, notably the lateral occipital complex (LOC) of the ventral visual stream, dorsal visual stream (DS), hippocampal formation (HIPP) and the prefrontal cortex (PFC). We have previously demonstrated the failure of closure processes in schizophrenia and shown that the dysregulation in the sensory information transmitted to the prefrontal cortex plays a critical role in this failure. Here, using a multimodal imaging approach in patients, combining event related electrophysiological recordings (ERP) and functional magnetic resonance imaging (fMRI), we characterize the spatiotemporal dynamics of priming in perceptual closure. Using directed functional connectivity measures we demonstrate that priming modifies the network-level interactions between the nodes of closure processing in a manner that is functionally advantageous to patients resulting in the mitigation of their deficit in perceptual closure.

It has been suggested that intergenerational transmission of risk for substance use disorder (SUD) manifests in the brain anatomy of substance naïve adolescents. While volume and shapes of subcortical structures (SSS) have been shown to be heritable, these structures, especially the pallidum, putamen, nucleus accumbens, and hippocampus, have also been associated with substance use disorders. However, it is not clear if those anatomical differences precede substance use or are the result of that use. Therefore, we examined if volume and SSS of adolescents with a family history (FH+) of SUD differed from adolescents without such a history (FH-). Because risk for SUD is associated with anxiety and impulsivity, we also examined correlations between these psychological characteristics and volume/SSS. Using structural MRI and FSL software, we segmented subcortical structures and obtained indices of SSS and volumes of 64 FH+ and 58 FH- adolescents. We examined group differences in volume and SSS, and the correlations between volume/SSS and trait anxiety and impulsivity. FH+ adolescents had a significant inward deformation in the shape of the right anterior hippocampus compared to FH- adolescents, while the volume of this structure did not differ between groups. Neither shape nor volume of the other subcortical structures differed between groups. In the FH+ adolescents, the left hippocampus shape was positively correlated with both trait anxiety and impulsivity, while in FH- adolescents a negative correlation pattern of SSS was seen in the hippocampus. SSS appears to capture local anatomical features that traditional volumetric analysis does not. The inward shape deformation in the right anterior hippocampus in FH+ adolescents may be related to the known increased risk for behavioral dysregulation leading to SUD in FH+ offspring. Hippocampus shape also exhibits opposite patterns of correlation with anxiety and impulsivity scores across the FH+ and FH- adolescents. These novel findings may reveal neural correlates, not captured by traditional volumetric analysis, of familial transmission of increased vulnerability to SUD.

Background: Interstitial lung disease requires frequent re-examination, which directly causes excessive cumulative radiation exposure. To date, AI has not been applied to CT for enhancing clinical care; thus, we hypothesize AI may empower CT with intelligence to realize automatic and accurate pulmonary scanning, thus dramatically decrease medical radiation exposure without compromising patient care. Methods: Facial boundary detection was realized by recognizing adjacent jaw position through training and testing a region proposal network (RPN) on 76,882 human faces using a preinstalled 2-dimensional camera; the lung-fields was then segmented by V-Net on another training set with 314 subjects and calculated the moving distance of the scanning couch based on a pre-generated calibration table. A multi-cohort study, including 1,186 patients was used for validation and radiation dose quantification under three clinical scenarios. Findings: A U-HAPPY (United imaging Human Automatic Planbox for PulmonarY) scanning CT was designed. Error distance of RPN was 4·46±0·02mm with a success rate of 98·7% in training set and 2·23±0·10mm with 100% success rate in testing set. Average Dice’s coefficient was 0·99 in training set and 0·96 in testing set. A calibration table with 1,344,000 matches was generated to support the linkage between camera and scanner. This real-time automation makes an accurate plan-box to cover exact location and area needed to scan, thus reducing amounts of radiation exposures significantly (all, P

Background: The rates of the fatality and disability will be extremely high once the intracranial aneurysms rupture. Radiologic examination based on the computed tomography angiography (CTA) images is the preferred method for the diagnosis of intracranial aneurysms, where the interpretation of imaging data is usually performed by an experienced radiologist. However, the accuracy of the interpretation and the time spent highly rely on the professional competence of the radiologist, which may yield inconsistent diagnostic outcomes. In addition, with steadily increasing workloads of radiology departments for providing various imaging diagnostic services, the fatigue of radiologists due to overloading may cause them neglecting or mistaking interpretations. Therefore, the aim of this study was to develop a new method to automatically detect intracranial aneurysms from CTA images using deep learning based on convolutional neural network (CNN) implemented on the Deepmedic platform. This new method can serve as a clinical diagnostic tool or a preliminary reference for diagnosis of intracranial aneurysms. It will be of great clinical value in reducing the diagnosis time and the missed or mistaken diagnosis caused by radiologists' different professional levels or their fatigue. Methods: We implemented a deep learning architecture with a 3D CNN model on Deepmedic platform for automatic segmentation and detection of intracranial aneurysms from CTA images which were retrospectively collected. The retrospective dataset was preprocessed and then was divided into two parts: the training part (80 subjects) to train the CNN model, and the testing part (10 subjects) to evaluate the performance of the automatic detection system. Sensitivity, positive predictive value (PPV) and the false positives per case were evaluated with the reference standard of digital subtraction angiography (DSA) or intraoperative results. Findings: The overall sensitivity and PPV of this system for the detection of intracranial aneurysms from CTA images were 92.3% and 100%, and the segmentation sensitivity also was 92.3%. The performance of the system in the detection of intracranial aneurysms was related to their size: the detection sensitivity for tiny intracranial aneurysms with a diameter ≤ 3mm was 66.7%, whereas the sensitivities of detection for large and medium-sized intracranial aneurysms of > 3mm both were 100%. Interpretation: The deep learning architecture with a 3D CNN model on Deepmedic platform could reliably segment and detect intracranial aneurysms from CTA images with an extremely high sensitivity, especially for large and medium-sized intracranial aneurysms, and this required only very limited training samples. Moreover, the detection sensitivity of tiny intracranial aneurysms was higher than those reported recently in other studies. Funding Statement: This study was supported in part by National Natural Science Foundation of China (81601558), Breeding Foundation of Zhuhai People's Hospital of China (2019PY-16) and Medical Research Foundation of Zhuhai City of China (20191207A010017). Declaration of Interests: No authors report any financial relationships with commercial interests Ethics Approval Statement: This retrospective study was approved by the Ethics Committee of Zhuhai People’s Hospital and the First Affiliated Hospital of Harbin Medical University.

Electronic Health Records (EHRs) have become one of the main sources of evidence to evaluate clinical actions, improve medical quality, predict disease-risk, and optimize treatment effects. However, EHRs present several modeling challenges, including heterogeneous data types and dynamic characteristics. Reinforcement learning provides an efficient way for sequential decision-making. Powered by model-based reinforcement learning approach, we propose an EHRs-based reinforcement learning algorithm to optimize sequential treatment strategies for diseases, such as sepsis, diabetes, and their complications. We conduct our experiments with this algorithm to optimize physicians’ historical treatment strategies and achieve better glucose control for diabetic ketoacidosis (DKA) patients, which is one serious complication of diabetes. The research includes the modeling process and reinforcement learning process. During the EHRs modeling process, besides considering the necessary physiological variables, we also consider the major disease factors to enhance the interpretability of the model. In the reinforcement learning process, a deep Q network is employed to explore the optimal insulin dose for patients. Moreover, inspired by the real medical scenes, we extend the algorithm to cooperative learning environment. We use the joint policy of the two agents to simulate doctor consultations, and achieve better treatment performances in terms of policy and blood glucose control than single agent and clinicians.

Summary Background Exposure therapy is the treatment of choice for anxiety disorders but requires people to confront feared situations and can be distressing. We tested the hypothesis that exposure without conscious awareness would reduce fear in participants with specific phobia by harnessing the neural circuitry supporting the automatic extinction of fear. Methods In this single-centre, randomised controlled experiment, we recruited women aged 18–29 years from an ethnically diverse, community-based population in northeastern USA, between Sept 1, 2013, and Aug 1, 2016. Eligible participants classified as having phobia met the DSM-5 criteria for specific phobia but not for any other disorder, had scores in the top 10% of respondents to the Fear of Spiders Questionnaire, and exhibited impairing avoidance of a live tarantula. Eligible controls met no criteria for any disorder, were in the bottom 30% of questionnaire respondents, and displayed no avoidance of the tarantula. The randomisation schedule was generated with the open source Research Randomizer Tool. A research assistant randomly assigned participants to the active intervention of very brief exposure (VBE)—the repeated presentation of masked phobic stimuli (ie, spiders)—or the control intervention which used masked flowers (VBF). VBE and VBF were given code numbers to prevent staff from knowing which intervention they were administering. During a 10 min functional MRI (fMRI) task, each participant was exposed to 16 blocks of ten masked target stimuli (spiders or flowers), alternating with 16 blocks of ten masked neutral stimuli. A few minutes after fMRI, participants with spider phobia approached the tarantula again so we could measure changes in phobic behaviour. The primary outcome was real-time changes in brain activity measured by fMRI. All analyses were done by intention to treat. Results We recruited 82 women, of whom 42 had spider phobia and 40 were controls. VBE generated stronger neural activity in participants with spider phobia than in controls, particularly in regions supporting emotion, emotion regulation, and attention systems, such as the inferior frontal cortex (Cohen's d 0·95, 95% CI 0·93–0·98, Bayesian posterior probability 99·5%) and the caudate nucleus (1·16, 1·14–1·18, 100·0%). In participants with phobia, VBE also generated stronger activity in these regions than did VBF (eg, dorsal anterior cingulate cortex Cohen's d 0·80, 95% CI 0·78–0·80, Bayesian posterior probability 98·5%; caudate nucleus 1·0, 0·98–1·02, 99·5%). VBE reduced avoidance of the live tarantula in participants with phobia. Regions supporting fear extinction (including ventral medial prefrontal cortex) and emotional salience processing mediated this effect. No adverse events occurred. Interpretation VBE reduced fear non-consciously in participants with spider phobia by recruiting brain regions supporting automatic fear extinction, emotion regulation, and top-down attentional processing. Future studies should explore the use of VBE in other fear-based disorders. Funding National Institutes of Mental Health and Brain & Behavior Research Foundation.

Background Frontostriatal and frontoparietal abnormalities likely contribute to deficits in control and attentional processes in individuals with bulimia nervosa and to the persistence of dysregulated eating across development. This study assessed these processes and cortical thickness in a large sample of adolescent girls and women with bulimia nervosa compared with healthy controls. Methods We collected anatomical MRI data from adolescent girls and women (ages 12-38 yr) with full or subthreshold bulimia nervosa and age-matched healthy controls who also completed the Conners Continuous Performance Test-II (CPT-II). Groups were compared on task performance and cortical thickness. Mediation analyses explored associations among cortical thickness, CPT-II variables, bulimia nervosa symptoms and age. Results We included 60 girls and women with bulimia nervosa and 54 controls in the analyses. Compared with healthy participants, those with bulimia nervosa showed increased impulsivity and inattention on the CPT-II, along with reduced thickness of the right pars triangularis, right superior parietal and left dorsal posterior cingulate cortices. In the bulimia nervosa group, exploratory analyses revealed that binge eating frequency correlated inversely with cortical thickness of frontoparietal and insular regions and that reduced frontoparietal thickness mediated the association between age and increased symptom severity and inattention. Binge eating frequency also mediated the association between age and lower prefrontal cortical thickness. Limitations These findings are applicable to only girls and women with bulimia nervosa, and our cross-sectional design precludes understanding of whether cortical thickness alterations precede or result from bulimia nervosa symptoms. Conclusion Structural abnormalities in the frontoparietal and posterior cingulate regions comprising circuits that support control and attentional processes should be investigated as potential contributors to the maintenance of bulimia nervosa and useful targets for novel interventions.