Publications by authors named "Weidong Cai"

112 Publications

Latent brain state dynamics distinguish behavioral variability, impaired decision-making, and inattention.

Mol Psychiatry 2021 Feb 15. Epub 2021 Feb 15.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA.

Children with Attention Deficit Hyperactivity Disorder (ADHD) have prominent deficits in sustained attention that manifest as elevated intra-individual response variability and poor decision-making. Influential neurocognitive models have linked attentional fluctuations to aberrant brain dynamics, but these models have not been tested with computationally rigorous procedures. Here we use a Research Domain Criteria approach, drift-diffusion modeling of behavior, and a novel Bayesian Switching Dynamic System unsupervised learning algorithm, with ultrafast temporal resolution (490 ms) whole-brain task-fMRI data, to investigate latent brain state dynamics of salience, frontoparietal, and default mode networks and their relation to response variability, latent decision-making processes, and inattention. Our analyses revealed that occurrence of a task-optimal latent brain state predicted decreased intra-individual response variability and increased evidence accumulation related to decision-making. In contrast, occurrence and dwell time of a non-optimal latent brain state predicted inattention symptoms and furthermore, in a categorical analysis, distinguished children with ADHD from controls. Importantly, functional connectivity between salience and frontoparietal networks predicted rate of evidence accumulation to a decision threshold, whereas functional connectivity between salience and default mode networks predicted inattention. Taken together, our computational modeling reveals dissociable latent brain state features underlying response variability, impaired decision-making, and inattentional symptoms common to ADHD. Our findings provide novel insights into the neurobiology of attention deficits in children.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41380-021-01022-3DOI Listing
February 2021

Panoptic Feature Fusion Net: A Novel Instance Segmentation Paradigm for Biomedical and Biological Images.

IEEE Trans Image Process 2021 21;30:2045-2059. Epub 2021 Jan 21.

Instance segmentation is an important task for biomedical and biological image analysis. Due to the complicated background components, the high variability of object appearances, numerous overlapping objects, and ambiguous object boundaries, this task still remains challenging. Recently, deep learning based methods have been widely employed to solve these problems and can be categorized into proposal-free and proposal-based methods. However, both proposal-free and proposal-based methods suffer from information loss, as they focus on either global-level semantic or local-level instance features. To tackle this issue, we present a Panoptic Feature Fusion Net (PFFNet) that unifies the semantic and instance features in this work. Specifically, our proposed PFFNet contains a residual attention feature fusion mechanism to incorporate the instance prediction with the semantic features, in order to facilitate the semantic contextual information learning in the instance branch. Then, a mask quality sub-branch is designed to align the confidence score of each object with the quality of the mask prediction. Furthermore, a consistency regularization mechanism is designed between the semantic segmentation tasks in the semantic and instance branches, for the robust learning of both tasks. Extensive experiments demonstrate the effectiveness of our proposed PFFNet, which outperforms several state-of-the-art methods on various biomedical and biological datasets.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2021.3050668DOI Listing
January 2021

Combination Therapy Using Kartogenin-Based Chondrogenesis and Complex Polymer Scaffold for Cartilage Defect Regeneration.

ACS Biomater Sci Eng 2020 Nov 13;6(11):6276-6284. Epub 2020 Oct 13.

Institute of Nano Biomedicine and Engineering, Shanghai Engineering Research Centre for Intelligent Diagnosis and Treatment Instrument, Department of Instrument Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, P. R. China.

Articular cartilage has a highly organized structure, responsible for supporting tremendous mechanical loads. How to repair defected articular cartilage has become a great challenge as the avascular nature of cartilage limits its regenerative ability. Aiming to facilitate chondrogenic differentiation and cartilage regeneration, we recently explored a novel combination therapy using soluble poly-l-lysine/Kartogenin (L-K) nanoparticles and a poly(lactic--glycolic acid) PLGA/methacrylated hyaluronic acid (PLHA) complex scaffold. The potential use for joint cartilage reconstruction was investigated through L-K nanoparticles stimulating adipose-derived stem cells (ADSCs) on PLHA scaffolding, which ultimately differentiated into cartilage . In this study, on one hand, an effective method was established for obtaining uniform L-K nanoparticles by self-assembly. They were further proved to be biocompatible to ADSCs cytotoxicity assays and to accelerate ADSCs secreting type 2 collagen in a dose-dependent manner by immunofluorescence. On the other hand, the porous PLHA scaffold was manufactured by the combination of coprecipitation and ultraviolet (UV) cross-linking. Nanoindentation technology-verified PLHA had an appropriate stiffness close to actual cartilage tissue. Additional microscopic observation confirmed that the PLHA platform supported proliferation and chondrogenesis for ADSCs . In the presence of ADSCs, a 12-week osteochondral defect regeneration by the combination therapy showed that smooth and intact cartilage tissue successfully regenerated. Furthermore, the results of combination therapy were superior to those of phosphate-buffered saline (PBS) only, KGN, or KGN/PLHA treatment. The results of magnetic resonance imaging (MRI) and histological assessment indicated that the renascent tissue gradually regenerated while the PLHA scaffold degraded. In conclusion, we have developed a novel multidimensional combination therapy of cartilage defect repair that facilitated cartilage regeneration. This strategy has a great clinical translational potential for articular cartilage repair in the near future.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acsbiomaterials.0c00724DOI Listing
November 2020

Deep Learning Methodology for Differentiating Glioma Recurrence From Radiation Necrosis Using Multimodal Magnetic Resonance Imaging: Algorithm Development and Validation.

JMIR Med Inform 2020 Nov 17;8(11):e19805. Epub 2020 Nov 17.

Centre for Health Informatics, Australian Institute of Health Innovation, Macquarie University, Sydney, Australia.

Background: The radiological differential diagnosis between tumor recurrence and radiation-induced necrosis (ie, pseudoprogression) is of paramount importance in the management of glioma patients.

Objective: This research aims to develop a deep learning methodology for automated differentiation of tumor recurrence from radiation necrosis based on routine magnetic resonance imaging (MRI) scans.

Methods: In this retrospective study, 146 patients who underwent radiation therapy after glioma resection and presented with suspected recurrent lesions at the follow-up MRI examination were selected for analysis. Routine MRI scans were acquired from each patient, including T1, T2, and gadolinium-contrast-enhanced T1 sequences. Of those cases, 96 (65.8%) were confirmed as glioma recurrence on postsurgical pathological examination, while 50 (34.2%) were diagnosed as necrosis. A light-weighted deep neural network (DNN) (ie, efficient radionecrosis neural network [ERN-Net]) was proposed to learn radiological features of gliomas and necrosis from MRI scans. Sensitivity, specificity, accuracy, and area under the curve (AUC) were used to evaluate performance of the model in both image-wise and subject-wise classifications. Preoperative diagnostic performance of the model was also compared to that of the state-of-the-art DNN models and five experienced neurosurgeons.

Results: DNN models based on multimodal MRI outperformed single-modal models. ERN-Net achieved the highest AUC in both image-wise (0.915) and subject-wise (0.958) classification tasks. The evaluated DNN models achieved an average sensitivity of 0.947 (SD 0.033), specificity of 0.817 (SD 0.075), and accuracy of 0.903 (SD 0.026), which were significantly better than the tested neurosurgeons (P=.02 in sensitivity and P<.001 in specificity and accuracy).

Conclusions: Deep learning offers a useful computational tool for the differential diagnosis between recurrent gliomas and necrosis. The proposed ERN-Net model, a simple and effective DNN model, achieved excellent performance on routine MRI scans and showed a high clinical applicability.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.2196/19805DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7708085PMC
November 2020

PDAM: A Panoptic-Level Feature Alignment Framework for Unsupervised Domain Adaptive Instance Segmentation in Microscopy Images.

IEEE Trans Med Imaging 2021 Jan 29;40(1):154-165. Epub 2020 Dec 29.

In this work, we present an unsupervised domain adaptation (UDA) method, named Panoptic Domain Adaptive Mask R-CNN (PDAM), for unsupervised instance segmentation in microscopy images. Since there currently lack methods particularly for UDA instance segmentation, we first design a Domain Adaptive Mask R-CNN (DAM) as the baseline, with cross-domain feature alignment at the image and instance levels. In addition to the image- and instance-level domain discrepancy, there also exists domain bias at the semantic level in the contextual information. Next, we, therefore, design a semantic segmentation branch with a domain discriminator to bridge the domain gap at the contextual level. By integrating the semantic- and instance-level feature adaptation, our method aligns the cross-domain features at the panoptic level. Third, we propose a task re-weighting mechanism to assign trade-off weights for the detection and segmentation loss functions. The task re-weighting mechanism solves the domain bias issue by alleviating the task learning for some iterations when the features contain source-specific factors. Furthermore, we design a feature similarity maximization mechanism to facilitate instance-level feature adaptation from the perspective of representational learning. Different from the typical feature alignment methods, our feature similarity maximization mechanism separates the domain-invariant and domain-specific features by enlarging their feature distribution dependency. Experimental results on three UDA instance segmentation scenarios with five datasets demonstrate the effectiveness of our proposed PDAM method, which outperforms state-of-the-art UDA methods by a large margin.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2020.3023466DOI Listing
January 2021

DeepAntigen: a novel method for neoantigen prioritization via 3D genome and deep sparse learning.

Bioinformatics 2020 Dec;36(19):4894-4901

Key Laboratory of Systems Biomedicine (Ministry of Education), Shanghai Centre for Systems Biomedicine, Shanghai Jiao Tong University, Shanghai 200240, China.

Motivation: The mutations of cancers can encode the seeds of their own destruction, in the form of T-cell recognizable immunogenic peptides, also known as neoantigens. It is computationally challenging, however, to accurately prioritize the potential neoantigen candidates according to their ability of activating the T-cell immunoresponse, especially when the somatic mutations are abundant. Although a few neoantigen prioritization methods have been proposed to address this issue, advanced machine learning model that is specifically designed to tackle this problem is still lacking. Moreover, none of the existing methods considers the original DNA loci of the neoantigens in the perspective of 3D genome which may provide key information for inferring neoantigens' immunogenicity.

Results: In this study, we discovered that DNA loci of the immunopositive and immunonegative MHC-I neoantigens have distinct spatial distribution patterns across the genome. We therefore used the 3D genome information along with an ensemble pMHC-I coding strategy, and developed a group feature selection-based deep sparse neural network model (DNN-GFS) that is optimized for neoantigen prioritization. DNN-GFS demonstrated increased neoantigen prioritization power comparing to existing sequence-based approaches. We also developed a webserver named deepAntigen (http://yishi.sjtu.edu.cn/deepAntigen) that implements the DNN-GFS as well as other machine learning methods. We believe that this work provides a new perspective toward more accurate neoantigen prediction which eventually contribute to personalized cancer immunotherapy.

Availability And Implementation: Data and implementation are available on webserver: http://yishi.sjtu.edu.cn/deepAntigen.

Supplementary Information: Supplementary data are available at Bioinformatics online.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/bioinformatics/btaa596DOI Listing
December 2020

Microstructural organization of human insula is linked to its macrofunctional circuitry and predicts cognitive control.

Elife 2020 06 4;9. Epub 2020 Jun 4.

Parietal, Inria Saclay Île-de-France, CEA Université Paris Sud, Palaiseau, France.

The human insular cortex is a heterogeneous brain structure which plays an integrative role in guiding behavior. The cytoarchitectonic organization of the human insula has been investigated over the last century using postmortem brains but there has been little progress in noninvasive in vivo mapping of its microstructure and large-scale functional circuitry. Quantitative modeling of multi-shell diffusion MRI data from 413 participants revealed that human insula microstructure differs significantly across subdivisions that serve distinct cognitive and affective functions. Insular microstructural organization was mirrored in its functionally interconnected circuits with the anterior cingulate cortex that anchors the salience network, a system important for adaptive switching of cognitive control systems. Furthermore, insular microstructural features, confirmed in Macaca mulatta, were linked to behavior and predicted individual differences in cognitive control ability. Our findings open new possibilities for probing psychiatric and neurological disorders impacted by insular cortex dysfunction, including autism, schizophrenia, and fronto-temporal dementia.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.7554/eLife.53470DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7308087PMC
June 2020

Chitosan modified FeO/KGN self-assembled nanoprobes for osteochondral MR diagnose and regeneration.

Theranostics 2020 15;10(12):5565-5577. Epub 2020 Apr 15.

Institute of Nano Biomedicine and Engineering, Shanghai Engineering Research Centre for Intelligent Diagnosis and Treatment Instrument, Department of Instrument Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, 800 Dongchuan RD, Shanghai 200240, PR China.

Chondral and osteochondral defects caused by trauma or pathological changes, commonly progress into total joint degradation, even resulting in disability. The cartilage restoration is a great challenge because of its avascularity and limited proliferative ability. Additionally, precise diagnosis using non-invasive detection techniques is challenging, which increases problems associated with chondral disease treatment. To achieve a theranostic goal, we used an integrated strategy that relies on exploiting a multifunctional nanoprobe based on chitosan-modified Fe3O4 nanoparticles, which spontaneously self-assemble with the oppositely charged small molecule growth factor, kartogenin (KGN). This nanoprobe was used to obtain distinctively brighter T-weighted magnetic resonance (MR) imaging, allowing its use as a positive contrast agent, and could be applied to obtain accurate diagnosis and osteochondral regeneration therapy. This nanoprobe was first investigated using adipose tissue-derived stem cells (ADSCs), and was found to be a novel positive contrast agent that also plays a significant role in stimulating ADSCs differentiation into chondrocytes. This self-assembled probe was not only biocompatible both and , contributing to cellular internalization, but was also used to successfully make distinction of normal/damaged tissue in T-weighted MR imaging. This novel combination was systematically shown to be biosafe via the decrement of apparent MR signals and elimination of ferroferric oxide over a 12-week regeneration period. Here, we established a novel method for osteochondral disease diagnosis and reconstruction. Using the FeO-CS/KGN nanoprobe, it is easy to distinguish the defect position, and it could act as a tool for dynamic observation as well as a stem cell-based therapy for directionally chondral differentiation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.7150/thno.43569DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7196312PMC
April 2020

Anxiety and Stress Alter Decision-Making Dynamics and Causal Amygdala-Dorsolateral Prefrontal Cortex Circuits During Emotion Regulation in Children.

Biol Psychiatry 2020 Oct 21;88(7):576-586. Epub 2020 Feb 21.

Department of Psychiatry and Behavioral Sciences, Stanford School of Medicine, Stanford, California. Electronic address:

Background: Anxiety and stress reactivity are risk factors for the development of affective disorders. However, the behavioral and neurocircuit mechanisms that potentiate maladaptive emotion regulation are poorly understood. Neuroimaging studies have implicated the amygdala and dorsolateral prefrontal cortex (DLPFC) in emotion regulation, but how anxiety and stress alter their context-specific causal circuit interactions is not known. Here, we use computational modeling to inform affective pathophysiology, etiology, and neurocircuit targets for early intervention.

Methods: Forty-five children (10-11 years of age; 25 boys) reappraised aversive stimuli during functional magnetic resonance imaging scanning. Clinical measures of anxiety and stress were acquired for each child. Drift-diffusion modeling of behavioral data and causal circuit analysis of functional magnetic resonance imaging data, with a National Institute of Mental Health Research Domain Criteria approach, were used to characterize latent behavioral and neurocircuit decision-making dynamics driving emotion regulation.

Results: Children successfully reappraised negative responses to aversive stimuli. Drift-diffusion modeling revealed that emotion regulation was characterized by increased initial bias toward positive reactivity during viewing of aversive stimuli and increased drift rate, which captured evidence accumulation during emotion evaluation. Crucially, anxiety and stress reactivity impaired latent behavioral dynamics associated with reappraisal and decision making. Anxiety and stress increased dynamic casual influences from the right amygdala to DLPFC. In contrast, DLPFC, but not amygdala, reactivity was correlated with evidence accumulation and decision making during emotion reappraisal.

Conclusions: Our findings provide new insights into how anxiety and stress in children impact decision making and amygdala-DLPFC signaling during emotion regulation, and uncover latent behavioral and neurocircuit mechanisms of early risk for psychopathology.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.biopsych.2020.02.011DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7442664PMC
October 2020

NFN+: A novel network followed network for retinal vessel segmentation.

Neural Netw 2020 Jun 4;126:153-162. Epub 2020 Mar 4.

Biomedical and Multimedia Information Technology (BMIT) Research Group, School of Computer Science, University of Sydney, Sydney, NSW 2006, Australia.

In the early diagnosis of diabetic retinopathy, the morphological attributes of blood vessels play an essential role to construct a retinal computer-aided diagnosis system. However, due to the challenges including limited densely annotated data, inter-vessel differences and structured prediction problem, it remains challenging to segment accurately the retinal vessels, particularly the capillaries on color fundus images. To address these issues, in this paper, we propose a novel deep learning-based model called NFN+ to effectively extract multi-scale information and make full use of deep feature maps. In NFN+, the front network converts an image patch into a probabilistic retinal vessel map, and the followed network further refines the map to achieve a better post-processing module, which helps represent the vessel structures implicitly. We employ the inter-network skip connections to unite two identical multi-scale backbones, which enables the useful multi-scale features to be directly transferred from shallow layers to deeper layers. The refined probabilistic retinal vessel maps produced from the augmented images are then averaged to construct the segmentation results. We evaluated this model on the digital retinal images for vessel extraction (DRIVE), structured analysis of the retina (STARE), and the child heart and health study (CHASE) databases. Our results indicate that the elaborated cascaded designs can produce performance gain and the proposed NFN+ model, to our best knowledge, achieved the state-of-the-art retinal vessel segmentation accuracy on color fundus images (AUC: 98.30%, 98.75% and 98.94%, respectively).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2020.02.018DOI Listing
June 2020

m6A Methylation Analysis of Differentially Expressed Genes in Skin Tissues of Coarse and Fine Type Liaoning Cashmere Goats.

Front Genet 2019 22;10:1318. Epub 2020 Jan 22.

College of Animal Science & Veterinary Medicine, Shenyang Agricultural University, Shenyang, China.

N6-methyladenosine (m6A) is the most common internal modification in mRNAs of all higher eukaryotes. Here we perform two high-throughput sequencing methods, m6A-modified RNA immunoprecipitation sequence (MeRIP-seq) and RNA sequence (RNA-seq) to identify key genes with m6A modification in cashmere fiber growth. A total of 9,085 m6A sites were differentially RNA m6A methylated as reported from by MeRIP-seq, including 7,170 upregulated and 1,915 downregulated. In addition, by comparing m6A-modified genes between the fine-type Liaoning cashmere goat (FT-LCG) and coarse-type Liaoning Cashmere Goat (CT-LCG) skin samples, we obtain 1,170 differentially expressed genes. In order to identify the differently methylated genes related to cashmere fiber growth, 19 genes were selected to validate by performing qRT-PCR in FT-LCG and CT-LCG. In addition, GO enrichment analysis shows that differently methylated genes are mainly involved in keratin filament and intermediate filament. These findings provide a theoretical basis for future research on the function of m6A modification during the growth of cashmere fiber.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fgene.2019.01318DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6987416PMC
January 2020

Inhibition-related modulation of salience and frontoparietal networks predicts cognitive control ability and inattention symptoms in children with ADHD.

Mol Psychiatry 2019 Oct 29. Epub 2019 Oct 29.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, 94305, USA.

Attention-deficit hyperactivity disorder (ADHD) is associated with pervasive impairments in attention and cognitive control. Although brain circuits underlying these impairments have been extensively investigated with resting-state fMRI, little is known about task-evoked functional brain circuits and their relation to cognitive control deficits and inattention symptoms in children with ADHD. Children with ADHD and age, gender and head motion matched typically developing (TD) children completed a Go/NoGo fMRI task. We used multivariate and dimensional analyses to investigate impairments in two core cognitive control systems: (i) cingulo-opercular "salience" network (SN) anchored in the right anterior insula, dorsal anterior cingulate cortex (rdACC), and ventrolateral prefrontal cortex (rVLPFC) and (ii) dorsal frontoparietal "central executive" (FPN) network anchored in right dorsolateral prefrontal cortex (rDLPFC) and posterior parietal cortex (rPPC). We found that multivariate patterns of task-evoked effective connectivity between brain regions in SN and FPN distinguished the ADHD and TD groups, with rDLPFC-rPPC connectivity emerging as the most distinguishing link. Task-evoked rdACC-rVLPFC connectivity was positively correlated with NoGo accuracy, and negatively correlated with severity of inattention symptoms. Brain-behavior relationships were robust against potential age, gender, and head motion confounds. Our findings highlight aberrancies in task-evoked modulation of SN and FPN connectivity in children with ADHD. Crucially, cingulo-frontal connectivity was a common locus of deficits in cognitive control and clinical measures of inattention symptoms. Our study provides insights into a parsimonious systems neuroscience model of cognitive control deficits in ADHD, and suggests specific circuit biomarkers for predicting treatment outcomes in childhood ADHD.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41380-019-0564-4DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7188596PMC
October 2019

Hyperdirect insula-basal-ganglia pathway and adult-like maturity of global brain responses predict inhibitory control in children.

Nat Commun 2019 10 22;10(1):4798. Epub 2019 Oct 22.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, 94305, USA.

Inhibitory control is fundamental to children's self-regulation and cognitive development. Here we investigate cortical-basal ganglia pathways underlying inhibitory control in children and their adult-like maturity. We first conduct a comprehensive meta-analysis of extant neurodevelopmental studies of inhibitory control and highlight important gaps in the literature. Second, we examine cortical-basal ganglia activation during inhibitory control in children ages 9-12 and demonstrate the formation of an adult-like inhibitory control network by late childhood. Third, we develop a neural maturation index (NMI), which assesses the similarity of brain activation patterns between children and adults, and demonstrate that higher NMI in children predicts better inhibitory control. Fourth, we show that activity in the subthalamic nucleus and its effective connectivity with the right anterior insula predicts children's inhibitory control. Fifth, we replicate our findings across multiple cohorts. Our findings provide insights into cortical-basal ganglia circuits and global brain organization underlying the development of inhibitory control.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41467-019-12756-8DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6805945PMC
October 2019

Biological impact of nanodiamond particles - label free, high-resolution methods for nanotoxicity assessment.

Nanotoxicology 2019 11 14;13(9):1210-1226. Epub 2019 Sep 14.

The University of Sydney, Sydney Nano Institute, Faculty of Medicine and Health, Sydney Pharmacy School, Sydney , Australia.

Current methods for the assessment of nanoparticle safety that are based on 2D cell culture models and fluorescence-based assays show limited sensitivity and they lack biomimicry. Consequently, the health risks associated with the use of many nanoparticles have not yet been established. There is a need to develop models that mimic physiology more accurately and enable high throughput assessment. There is also a need to set up new assays that offer high sensitivity and are label-free. Here we developed 'mini-liver' models using scaffold-free bioprinting and used these models together with label-free nanoscale techniques for the assessment of toxicity of nanodiamond produced by laser-assisted technology. Results showed that NDs induced cytotoxicity in a concentration and exposure-time dependent manner. The loss of cell function was confirmed by increased cell stiffness, decreased cell membrane barrier integrity and reduced cells mobility. We further showed that NDs elevated the production of reactive oxygen species and reduced cell viability. Our approach that combined mini-liver models with label-free high-resolution techniques showed improved sensitivity in toxicity assessment. Notably, this approach allowed for label-free semi-high throughput measurements of nanoparticle-cell interactions, thus could be considered as a complementary approach to currently used methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1080/17435390.2019.1650970DOI Listing
November 2019

3D APA-Net: 3D Adversarial Pyramid Anisotropic Convolutional Network for Prostate Segmentation in MR Images.

IEEE Trans Med Imaging 2020 02 11;39(2):447-457. Epub 2019 Jul 11.

Accurate and reliable segmentation of the prostate gland using magnetic resonance (MR) imaging has critical importance for the diagnosis and treatment of prostate diseases, especially prostate cancer. Although many automated segmentation approaches, including those based on deep learning have been proposed, the segmentation performance still has room for improvement due to the large variability in image appearance, imaging interference, and anisotropic spatial resolution. In this paper, we propose the 3D adversarial pyramid anisotropic convolutional deep neural network (3D APA-Net) for prostate segmentation in MR images. This model is composed of a generator (i.e., 3D PA-Net) that performs image segmentation and a discriminator (i.e., a six-layer convolutional neural network) that differentiates between a segmentation result and its corresponding ground truth. The 3D PA-Net has an encoder-decoder architecture, which consists of a 3D ResNet encoder, an anisotropic convolutional decoder, and multi-level pyramid convolutional skip connections. The anisotropic convolutional blocks can exploit the 3D context information of the MR images with anisotropic resolution, the pyramid convolutional blocks address both voxel classification and gland localization issues, and the adversarial training regularizes 3D PA-Net and thus enables it to generate spatially consistent and continuous segmentation results. We evaluated the proposed 3D APA-Net against several state-of-the-art deep learning-based segmentation approaches on two public databases and the hybrid of the two. Our results suggest that the proposed model outperforms the compared approaches on three databases and could be used in a routine clinical workflow.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2928056DOI Listing
February 2020

Temporal Correlation Structure Learning for MCI Conversion Prediction.

Med Image Comput Comput Assist Interv 2018 Sep 13;11072:446-454. Epub 2018 Sep 13.

Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, USA.

In Alzheimer's research, Mild Cognitive Impairment (MCI) is an important intermediate stage between normal aging and Alzheimer's. How to distinguish MCI samples that finally convert to AD from those do not is an essential problem in the prevention and diagnosis of Alzheimer's. Traditional methods use various classification models to distinguish MCI converters from non-converters, while the performance is usually limited by the small number of available data. Moreover, previous methods only use the data at baseline time for training but ignore the longitudinal information at other time points along the disease progression. To tackle with these problems, we propose a novel deep learning framework that uncovers the temporal correlation structure between adjacent time points in the disease progression. We also construct a generative framework to learn the inherent data distribution so as to produce more reliable data to strengthen the training process. Extensive experiments on the ADNI cohort validate the superiority of our model.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/978-3-030-00931-1_51DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6519075PMC
September 2018

Computation Methods for Biomedical Information Analysis.

J Healthc Eng 2018 1;2018:8683601. Epub 2018 Nov 1.

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1155/2018/8683601DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6236979PMC
November 2019

Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks.

BMC Genomics 2018 Aug 13;19(Suppl 6):565. Epub 2018 Aug 13.

Key Laboratory of Systems Biomedicine, Shanghai Center for Systems Biomedicine, Shanghai Jiaotong University, Shanghai, 200240, China.

Background: With the developments of DNA sequencing technology, large amounts of sequencing data have been produced that provides unprecedented opportunities for advanced association studies between somatic mutations and cancer types/subtypes which further contributes to more accurate somatic mutation based cancer typing (SMCT). In existing SMCT methods however, the absence of high-level feature extraction is a major obstacle in improving the classification performance.

Results: We propose DeepCNA, an advanced convolutional neural network (CNN) based classifier, which utilizes copy number aberrations (CNAs) and HiC data, to address this issue. DeepCNA first pre-process the CNA data by clipping, zero padding and reshaping. Then, the processed data is fed into a CNN classifier, which extracts high-level features for accurate classification. Experimental results on the COSMIC CNA dataset indicate that 2D CNN with both cell lines of HiC data lead to the best performance. We further compare DeepCNA with three widely adopted classifiers, and demonstrate that DeepCNA has at least 78% improvement of performance.

Conclusions: This paper demonstrates the advantages and potential of the proposed DeepCNA model for processing of somatic point mutation based gene data, and proposes that its usage may be extended to other complex genotype-phenotype association studies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12864-018-4919-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6101087PMC
August 2018

Copper nanoparticles with near-unity, omnidirectional, and broadband optical absorption for highly efficient solar steam generation.

Nanotechnology 2019 Jan 26;30(1):015402. Epub 2018 Oct 26.

Key Laboratory of Optoelectronic Technology of Jiangsu Province, Center for Quantum Transport and Thermal Energy Science, School of Physics and Technology, Nanjing Normal University, Nanjing 210023, People's Republic of China.

Solar steam generation provides a renewable and environmentally friendly approach to solve the water shortage issue. The pursuit of efficient, stable, and cheap photothermal agents is thus of great significance. In this work, Cu nanoparticles (NPs) fabricated simply by a substitution reaction, exhibit a near-unity (∼97.7%) light absorption, covering a broad incident angle and wavelength range (200-1300 nm). Thereby, a high photothermal conversion efficiency of 93% is achieved. The excellent photothermal performance offers a unique opportunity for the development of solar steam generation. By coating the Cu NPs on a cellulose membrane, a solar steam generation efficiency up to 73% is acquired at a low irradiation power density of 2 kW m (1 kW m = 1 sun). Moreover, the Cu NPs are recyclable with the high stability being resistant to heat, photoirradiation and corrosion of brine.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1088/1361-6528/aae678DOI Listing
January 2019

Knowledge-based Collaborative Deep Learning for Benign-Malignant Lung Nodule Classification on Chest CT.

IEEE Trans Med Imaging 2019 04 17;38(4):991-1004. Epub 2018 Oct 17.

The accurate identification of malignant lung nodules on chest CT is critical for the early detection of lung cancer, which also offers patients the best chance of cure. Deep learning methods have recently been successfully introduced to computer vision problems, although substantial challenges remain in the detection of malignant nodules due to the lack of large training data sets. In this paper, we propose a multi-view knowledge-based collaborative (MV-KBC) deep model to separate malignant from benign nodules using limited chest CT data. Our model learns 3-D lung nodule characteristics by decomposing a 3-D nodule into nine fixed views. For each view, we construct a knowledge-based collaborative (KBC) submodel, where three types of image patches are designed to fine-tune three pre-trained ResNet-50 networks that characterize the nodules' overall appearance, voxel, and shape heterogeneity, respectively. We jointly use the nine KBC submodels to classify lung nodules with an adaptive weighting scheme learned during the error back propagation, which enables the MV-KBC model to be trained in an end-to-end manner. The penalty loss function is used for better reduction of the false negative rate with a minimal effect on the overall performance of the MV-KBC model. We tested our method on the benchmark LIDC-IDRI data set and compared it to the five state-of-the-art classification approaches. Our results show that the MV-KBC model achieved an accuracy of 91.60% for lung nodule classification with an AUC of 95.70%. These results are markedly superior to the state-of-the-art approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2018.2876510DOI Listing
April 2019

Dysregulated Brain Dynamics in a Triple-Network Saliency Model of Schizophrenia and Its Relation to Psychosis.

Biol Psychiatry 2019 01 1;85(1):60-69. Epub 2018 Aug 1.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, California; Department of Neurology & Neurological Sciences, Stanford University School of Medicine, Stanford, California; Stanford Neurosciences Institute, Stanford University School of Medicine, Stanford, California. Electronic address:

Background: Schizophrenia is a highly disabling psychiatric disorder characterized by a range of positive "psychosis" symptoms. However, the neurobiology of psychosis and associated systems-level disruptions in the brain remain poorly understood. Here, we test an aberrant saliency model of psychosis, which posits that dysregulated dynamic cross-network interactions among the salience network (SN), central executive network, and default mode network contribute to positive symptoms in patients with schizophrenia.

Methods: Using task-free functional magnetic resonance imaging data from two independent cohorts, we examined 1) dynamic time-varying cross-network interactions among the SN, central executive network, and default mode network in 130 patients with schizophrenia versus well-matched control subjects; 2) accuracy of a saliency model-based classifier for distinguishing dynamic brain network interactions in patients versus control subjects; and 3) the relation between SN-centered network dynamics and clinical symptoms.

Results: In both cohorts, we found that dynamic SN-centered cross-network interactions were significantly reduced, less persistent, and more variable in patients with schizophrenia compared with control subjects. Multivariate classification analysis identified dynamic SN-centered cross-network interaction patterns as factors that distinguish patients from control subjects, with accuracies of 78% and 80% in the two cohorts, respectively. Crucially, in both cohorts, dynamic time-varying measures of SN-centered cross-network interactions were correlated with positive, but not negative, symptoms.

Conclusions: Aberrations in time-varying engagement of the SN with the central executive network and default mode network is a clinically relevant neurobiological signature of psychosis in schizophrenia. Our findings provide strong evidence for dysregulated brain dynamics in a triple-network saliency model of schizophrenia and inform theoretically motivated systems neuroscience approaches for characterizing aberrant brain dynamics associated with psychosis.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.biopsych.2018.07.020DOI Listing
January 2019

Dopamine-related dissociation of cortical and subcortical brain activations in cognitively unimpaired Parkinson's disease patients OFF and ON medications.

Neuropsychologia 2018 10 21;119:24-33. Epub 2018 Jul 21.

Stanford University Medical Center, Department of Neurology & Neurological Sciences, Stanford, CA 94305, USA; Stanford University Medical Center, Department of Neurosurgery, Stanford, CA 94305, USA. Electronic address:

Background: Despite dopaminergic depletion that is severe enough to cause the motor symptoms of Parkinson's disease (PD), many patients remain cognitively unimpaired. Little is known about brain mechanisms underlying such preserved cognitive abilities and their alteration by dopaminergic medications.

Objectives: We investigated brain activations underlying dopamine-related differences in cognitive function using a unique experimental design with PD patients off and on dopaminergic medications. We tested the dopamine overdose hypothesis, which posits that the excess of exogenous dopamine in the frontal cortical regions can impair cognition.

Methods: We used a two-choice forced response Choice Reaction Time (CRT) task to probe cognitive processes underlying response selection and execution. Functional magnetic resonance imaging data were acquired from 16 cognitively unimpaired (Level-II) PD participants and 15 well-matched healthy controls (HC). We compared task performance (i.e. reaction time and accuracy) and brain activation of PD participants off dopaminergic medications (PD_OFF) in comparison with HC, and PD_OFF participants with those on dopaminergic medications (PD_ON).

Results: PD_OFF and PD_ON groups did not differ from each other, or from the HC group, in reaction time or accuracy. Compared to HC, PD_OFF activated the bilateral putamen less, and this was compensated by higher activation of the anterior insula. No such differences were observed in the PD_ON group, compared to HC. Compared to both HC and PD_OFF, PD_ON participants showed dopamine-related hyperactivation in the frontal cortical regions and hypoactivation in the amygdala.

Conclusion: Our data provide further evidence that PD_OFF and PD_ON participants engage different cortical and subcortical systems to achieve similar levels of cognitive performance as HC. Crucially, our findings demonstrate dopamine-related dissociation in brain activation between cortical and subcortical regions, and provide novel support for the dopamine overdose hypothesis.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neuropsychologia.2018.07.024DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6191343PMC
October 2018

Automated 3-D Neuron Tracing With Precise Branch Erasing and Confidence Controlled Back Tracking.

IEEE Trans Med Imaging 2018 11 4;37(11):2441-2452. Epub 2018 May 4.

The automatic reconstruction of single neurons from microscopic images is essential to enable large-scale data-driven investigations in neuron morphology research. However, few previous methods were able to generate satisfactory results automatically from 3-D microscopic images without human intervention. In this paper, we developed a new algorithm for automatic 3-D neuron reconstruction. The main idea of the proposed algorithm is to iteratively track backward from the potential neuronal termini to the soma centre. An online confidence score is computed to decide if a tracing iteration should be stopped and discarded from the final reconstruction. The performance improvements comparing with the previous methods are mainly introduced by a more accurate estimation of the traced area and the confidence controlled back-tracking algorithm. The proposed algorithm supports large-scale batch-processing by requiring only one user specified parameter for background segmentation. We bench tested the proposed algorithm on the images obtained from both the DIADEM challenge and the BigNeuron challenge. Our proposed algorithm achieved the state-of-the-art results.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2018.2833420DOI Listing
November 2018

Multi-Pass Fast Watershed for Accurate Segmentation of Overlapping Cervical Cells.

IEEE Trans Med Imaging 2018 09 12;37(9):2044-2059. Epub 2018 Mar 12.

The task of segmenting cell nuclei and cytoplasm in pap smear images is one of the most challenging tasks in automated cervix cytological analysis due to specifically the presence of overlapping cells. This paper introduces a multi-pass fast watershed-based method (MPFW) to segment both nucleus and cytoplasm from large cell masses of overlapping cervical cells in three watershed passes. The first pass locates the nuclei with barrier-based watershed on the gradient-based edge map of a pre-processed image. The next pass segments the isolated, touching, and partially overlapping cells with a watershed transform adapted to the cell shape and location. The final pass introduces mutual iterative watersheds separately applied to each nucleus in the largely overlapping clusters to estimate the cell shape. In MPFW, the line-shaped contours of the watershed cells are deformed with ellipse fitting and contour adjustment to give a better representation of cell shapes. The performance of the proposed method has been evaluated using synthetic, real extended depth-of-field, and multi-layers cervical cytology images provided by the first and second overlapping cervical cytology image segmentation challenges in ISBI 2014 and ISBI 2015. The experimental results demonstrate superior performance of the proposed MPFW in terms of segmentation accuracy, detection rate, and time complexity, compared with recent peer methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2018.2815013DOI Listing
September 2018

Uncovering hidden brain state dynamics that regulate performance and decision-making during cognition.

Nat Commun 2018 06 27;9(1):2505. Epub 2018 Jun 27.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, 94305, USA.

Human cognition is influenced not only by external task demands but also latent mental processes and brain states that change over time. Here, we use novel Bayesian switching dynamical systems algorithm to identify hidden brain states and determine that these states are only weakly aligned with external task conditions. We compute state transition probabilities and demonstrate how dynamic transitions between hidden states allow flexible reconfiguration of functional brain circuits. Crucially, we identify latent transient brain states and dynamic functional circuits that are optimal for cognition and show that failure to engage these states in a timely manner is associated with poorer task performance and weaker decision-making dynamics. We replicate findings in a large sample (N = 122) and reveal a robust link between cognition and flexible latent brain state dynamics. Our study demonstrates the power of switching dynamical systems models for investigating hidden dynamic brain states and functional interactions underlying human cognition.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41467-018-04723-6DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6021386PMC
June 2018

Regularized Modal Regression with Applications in Cognitive Impairment Prediction.

Adv Neural Inf Process Syst 2017 Dec;30:1448-1458

Department of Electrical and Computer Engineering, University of Pittsburgh, USA.

Linear regression models have been successfully used to function estimation and model selection in high-dimensional data analysis. However, most existing methods are built on least squares with the mean square error (MSE) criterion, which are sensitive to outliers and their performance may be degraded for heavy-tailed noise. In this paper, we go beyond this criterion by investigating the regularized modal regression from a statistical learning viewpoint. A new regularized modal regression model is proposed for estimation and variable selection, which is robust to outliers, heavy-tailed noise, and skewed noise. On the theoretical side, we establish the approximation estimate for learning the conditional mode function, the sparsity analysis for variable selection, and the robustness characterization. On the application side, we applied our model to successfully improve the cognitive impairment prediction using the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort data.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5895184PMC
December 2017

Aberrant Time-Varying Cross-Network Interactions in Children With Attention-Deficit/Hyperactivity Disorder and the Relation to Attention Deficits.

Biol Psychiatry Cogn Neurosci Neuroimaging 2018 03 7;3(3):263-273. Epub 2017 Nov 7.

Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, California; Department of Neurology & Neurological Sciences, Stanford University School of Medicine, Stanford, California; Stanford Neurosciences Institute, Stanford University School of Medicine, Stanford, California.

Background: Attention-deficit/hyperactivity disorder (ADHD) is thought to stem from aberrancies in large-scale cognitive control networks. However, the exact nature of aberrant brain circuit dynamics involving these control networks is poorly understood. Using a saliency-based triple-network model of cognitive control, we tested the hypothesis that dynamic cross-network interactions among the salience, central executive, and default mode networks are dysregulated in children with ADHD, and we investigated how these dysregulations contribute to inattention.

Methods: Using functional magnetic resonance imaging data from 140 children with ADHD and typically developing children from two cohorts (primary cohort = 80 children, replication cohort = 60 children) in a case-control design, we examined both time-averaged and dynamic time-varying cross-network interactions in each cohort separately.

Results: Time-averaged measures of salience network-centered cross-network interactions were significantly lower in children with ADHD compared with typically developing children and were correlated with severity of inattention symptoms. Children with ADHD displayed more variable dynamic cross-network interaction patterns, including less persistent brain states, significantly shorter mean lifetimes of brain states, and intermittently weaker cross-network interactions. Importantly, dynamic time-varying measures of cross-network interactions were more strongly correlated with inattention symptoms than with time-averaged measures of functional connectivity. Crucially, we replicated these findings in the two independent cohorts of children with ADHD and typically developing children.

Conclusions: Aberrancies in time-varying engagement of the salience network with the central executive network and default mode network are a robust and clinically relevant neurobiological signature of childhood ADHD symptoms. The triple-network neurocognitive model provides a novel, replicable, and parsimonious dynamical systems neuroscience framework for characterizing childhood ADHD and inattention.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.bpsc.2017.10.005DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5833018PMC
March 2018

Automated 3D Soma Segmentation with Morphological Surface Evolution for Neuron Reconstruction.

Neuroinformatics 2018 04;16(2):153-166

School of Information Technologies, University of Sydney, Sydney, NSW, Australia.

The automatic neuron reconstruction is important since it accelerates the collection of 3D neuron models for the neuronal morphological studies. The majority of the previous neuron reconstruction methods only focused on tracing neuron fibres without considering the somatic surface. Thus, topological errors often present around the soma area in the results obtained by these tracing methods. Segmentation of the soma structures can be embedded in the existing neuron tracing methods to reduce such topological errors. In this paper, we present a novel method to segment the soma structures with complex geometry. It can be applied along with the existing methods in a fully automated pipeline. An approximate bounding block is firstly estimated based on a geodesic distance transform. Then the soma segmentation is obtained by evolving the surface with a set of morphological operators inside the initial bounding region. By evaluating the methods against the challenging images released by the BigNeuron project, we showed that the proposed method can outperform the existing soma segmentation methods regarding the accuracy. We also showed that the soma segmentation can be used for enhancing the results of existing neuron tracing methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s12021-017-9353-xDOI Listing
April 2018

Suprathreshold fiber cluster statistics: Leveraging white matter geometry to enhance tractography statistical analysis.

Neuroimage 2018 05 11;171:341-354. Epub 2018 Jan 11.

Brigham and Women's Hospital, Harvard Medical School, Boston, USA.

This work presents a suprathreshold fiber cluster (STFC) method that leverages the whole brain fiber geometry to enhance statistical group difference analyses. The proposed method consists of 1) a well-established study-specific data-driven tractography parcellation to obtain white matter tract parcels and 2) a newly proposed nonparametric, permutation-test-based STFC method to identify significant differences between study populations. The basic idea of our method is that a white matter parcel's neighborhood (nearby parcels with similar white matter anatomy) can support the parcel's statistical significance when correcting for multiple comparisons. We propose an adaptive parcel neighborhood strategy to allow suprathreshold fiber cluster formation that is robust to anatomically varying inter-parcel distances. The method is demonstrated by application to a multi-shell diffusion MRI dataset from 59 individuals, including 30 attention deficit hyperactivity disorder patients and 29 healthy controls. Evaluations are conducted using both synthetic and in-vivo data. The results indicate that the STFC method gives greater sensitivity in finding group differences in white matter tract parcels compared to several traditional multiple comparison correction methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neuroimage.2018.01.006DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5857470PMC
May 2018

A Deep Learning-Based Algorithm Identifies Glaucomatous Discs Using Monoscopic Fundus Photographs.

Ophthalmol Glaucoma 2018 Jul - Aug;1(1):15-22. Epub 2018 Jun 5.

Save Sight Institute, Sydney Medical School, The University of Sydney, Sydney, Australia; Department of Clinical Medicine, Faculty of Medicine and Health Sciences, Macquarie University, Sydney, Australia. Electronic address:

Purpose: To develop and test the performance of a deep learning-based algorithm for glaucomatous disc identification using monoscopic fundus photographs.

Design: Fundus photograph database study.

Participants: Four thousand three hundred ninety-four fundus photographs, including 3768 images from previous Sydney-based clinical studies and 626 images from publicly available online RIM-ONE and High-Resolution Fundus (HRF) databases with definitive diagnoses.

Methods: We merged all databases except the HRF database, and then partitioned the dataset into a training set (80% of all cases) and a testing set (20% of all cases). We used the HRF images as an additional testing set. We compared the performance of the artificial intelligence (AI) system against a panel of practicing ophthalmologists including glaucoma subspecialists from Australia, New Zealand, Canada, and the United Kingdom.

Main Outcome Measures: The sensitivity and specificity of the AI system in detecting glaucomatous optic discs.

Results: By using monoscopic fundus photographs, the AI system demonstrated a high accuracy rate in glaucomatous disc identification (92.7%; 95% confidence interval [CI], 91.2%-94.2%), achieving 89.3% sensitivity (95% CI, 86.8%-91.7%) and 97.1% specificity (95% CI, 96.1%-98.1%), with an area under the receiver operating characteristic curve of 0.97 (95% CI, 0.96-0.98). Using the independent online HRF database (30 images), the AI system again accomplished high accuracy, with 86.7% in both sensitivity and specificity (for ophthalmologists, 75.6% sensitivity and 77.8% specificity) and an area under the receiver operating characteristic curve of 0.89 (95% CI, 0.76-1.00).

Conclusions: This study demonstrated that a deep learning-based algorithm can identify glaucomatous discs at high accuracy level using monoscopic fundus images. Given that it is far easier to obtain monoscopic disc images than high-quality stereoscopic images, this study highlights the algorithm's potential application in large population-based disease screening or telemedicine programs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ogla.2018.04.002DOI Listing
June 2018