A robust approach to 3D neuron shape representation for quantification and classification.

We consider the problem of finding an accurate representation of neuron shapes, extracting sub-cellular features, and classifying neurons based on neuron shapes. In neuroscience research, the skeleton representation is often used as a compact and abstract representation of neuron shapes. However, existing methods are limited to getting and analyzing "curve" skeletons which can only be applied for tubular shapes. This paper presents a 3D neuron morphology analysis method for more general and complex neuron shapes. First, we introduce the concept of skeleton mesh to represent general neuron shapes and propose a novel method for computing mesh representations from 3D surface point clouds. A skeleton graph is then obtained from skeleton mesh and is used to extract sub-cellular features. Finally, an unsupervised learning method is used to embed the skeleton graph for neuron classification. Extensive experiment results are provided and demonstrate the robustness of our method to analyze neuron morphology.

Geographic Atrophy Segmentation Using Multimodal Deep Learning.

Purpose: To examine deep learning (DL)-based methods for accurate segmentation of geographic atrophy (GA) lesions using fundus autofluorescence (FAF) and near-infrared (NIR) images. Methods: This retrospective analysis utilized imaging data from study eyes of patients enrolled in Proxima A and B (NCT02479386; NCT02399072) natural history studies of GA. Two multimodal DL networks (UNet and YNet) were used to automatically segment GA lesions on FAF; segmentation accuracy was compared with annotations by experienced graders. The training data set comprised 940 image pairs (FAF and NIR) from 183 patients in Proxima B; the test data set comprised 497 image pairs from 154 patients in Proxima A. Dice coefficient scores, Bland-Altman plots, and Pearson correlation coefficient (r) were used to assess performance. Results: On the test set, Dice scores for the DL network to grader comparison ranged from 0.89 to 0.92 for screening visit; Dice score between graders was 0.94. GA lesion area correlations (r) for YNet versus grader, UNet versus grader, and between graders were 0.981, 0.959, and 0.995, respectively. Longitudinal GA lesion area enlargement correlations (r) for screening to 12 months (n = 53) were lower (0.741, 0.622, and 0.890, respectively) compared with the cross-sectional results at screening. Longitudinal correlations (r) from screening to 6 months (n = 77) were even lower (0.294, 0.248, and 0.686, respectively). Conclusions: Multimodal DL networks to segment GA lesions can produce accurate results comparable with expert graders. Translational Relevance: DL-based tools may support efficient and individualized assessment of patients with GA in clinical research and practice.

3D Neuron Morphology Analysis.

We consider the problem of finding an accurate representation of neuron shapes, extracting sub-cellular features, and classifying neurons based on neuron shapes. In neuroscience research, the skeleton representation is often used as a compact and abstract representation of neuron shapes. However, existing methods are limited to getting and analyzing"curve"skeletons which can only be applied for tubular shapes. This paper presents a 3D neuron morphology analysis method for more general and complex neuron shapes. First, we introduce the concept of skeleton mesh to represent general neuron shapes and propose a novel method for computing mesh representations from 3D surface point clouds. A skeleton graph is then obtained from skeleton mesh and is used to extract sub-cellular features. Finally, an unsupervised learning method is used to embed the skeleton graph for neuron classification. Extensive experiment results are provided and demonstrate the robustness of our method to analyze neuron morphology.

Segmentation, tracking, and sub-cellular feature extraction in 3D time-lapse images.

This paper presents a method for time-lapse 3D cell analysis. Specifically, we consider the problem of accurately localizing and quantitatively analyzing sub-cellular features, and for tracking individual cells from time-lapse 3D confocal cell image stacks. The heterogeneity of cells and the volume of multi-dimensional images presents a major challenge for fully automated analysis of morphogenesis and development of cells. This paper is motivated by the pavement cell growth process, and building a quantitative morphogenesis model. We propose a deep feature based segmentation method to accurately detect and label each cell region. An adjacency graph based method is used to extract sub-cellular features of the segmented cells. Finally, the robust graph based tracking algorithm using multiple cell features is proposed for associating cells at different time instances. We also demonstrate the generality of our tracking method on C. elegans fluorescent nuclei imagery. Extensive experiment results are provided and demonstrate the robustness of the proposed method. The code is available on GitHub and the method is available as a service through the BisQue portal.

Corrigendum: Improving Patch-Based Convolutional Neural Networks for MRI Brain Tumor Segmentation by Leveraging Location Information.

[This corrects the article DOI: 10.3389/fnins.2019.01449.].

The lncRNA PVT1 regulates nasopharyngeal carcinoma cell proliferation via activating the KAT2A acetyltransferase and stabilizing HIF-1α.

Long noncoding RNAs (lncRNAs) play important roles in regulating the development and progression of many cancers. However, the clinical significance of specific lncRNAs in the context of nasopharyngeal carcinoma (NPC) and the molecular mechanisms by which they regulate this form of cancer remain largely unclear. In this study we found that the lncRNA PVT1 was upregulated in NPC, and that in patients this upregulation was associated with reduced survival. RNA sequencing revealed that PVT1 was responsible for regulating NPC cell proliferation and for controlling a hypoxia-related phenotype in these cells. PVT1 knockdown reduced NPC cell proliferation, colony formation, and tumorigenesis in a subcutaneous mouse xenograft model systems. We further found that PVT1 serves as a scaffold for the chromatin modification factor KAT2A, which mediates histone 3 lysine 9 acetylation (H3K9), recruiting the nuclear receptor binding protein TIF1ß to activate NF90 transcription, thereby increasing HIF-1α stability and promoting a malignant phenotype in NPC cells. Overexpression of NF90 or HIF-1α restored the proliferation in cells that had ceased proliferating due to PVT1 or KAT2A depletion. Conversely, overexpression of active KAT2A or TIF1ß, but not of KAT2A acetyltransferase activity-deficient mutants or TIF1ß isoforms lacking H3K9ac binding sites, promoted a PVT1-mediated increase in NF90 transcription, as well as increased HIF-1α stability and cell proliferation. PVT1 knockdown enhanced the radiosensitization effect in NPC cells via inhibiting binding between H3K9ac and TIF1ß in a manner. Taken together, our results demonstrate that PVT1 serves an oncogenic role and plays an important role in radiosensitivity in malignant NPC via activating the KAT2A acetyltransferase and stabilizing HIF-1α.

Improving Patch-Based Convolutional Neural Networks for MRI Brain Tumor Segmentation by Leveraging Location Information.

The manual brain tumor annotation process is time consuming and resource consuming, therefore, an automated and accurate brain tumor segmentation tool is greatly in demand. In this paper, we introduce a novel method to integrate location information with the state-of-the-art patch-based neural networks for brain tumor segmentation. This is motivated by the observation that lesions are not uniformly distributed across different brain parcellation regions and that a locality-sensitive segmentation is likely to obtain better segmentation accuracy. Toward this, we use an existing brain parcellation atlas in the Montreal Neurological Institute (MNI) space and map this atlas to the individual subject data. This mapped atlas in the subject data space is integrated with structural Magnetic Resonance (MR) imaging data, and patch-based neural networks, including 3D U-Net and DeepMedic, are trained to classify the different brain lesions. Multiple state-of-the-art neural networks are trained and integrated with XGBoost fusion in the proposed two-level ensemble method. The first level reduces the uncertainty of the same type of models with different seed initializations, and the second level leverages the advantages of different types of neural network models. The proposed location information fusion method improves the segmentation performance of state-of-the-art networks including 3D U-Net and DeepMedic. Our proposed ensemble also achieves better segmentation performance compared to the state-of-the-art networks in BraTS 2017 and rivals state-of-the-art networks in BraTS 2018. Detailed results are provided on the public multimodal brain tumor segmentation (BraTS) benchmarks.

LncRNA DANCR upregulates PI3K/AKT signaling through activating serine phosphorylation of RXRA.

ABSTACT: Conventional therapies and novel molecular targeted therapies against breast cancer have gained great advances over the past two decades. However, poor prognosis and low survival rate are far from expectation for improvement, particularly in patients with triple negative breast cancer (TNBC). Here, we found that lncRNA DANCR was significantly overregulated in TNBC tissues and cell lines compared with normal breast tissues or other type of breast cancer. Knockdown of DANCR suppressed TNBC proliferation both in vitro and in vivo. Further study of underlying mechanisms demonstrated that DANCR bound with RXRA and increased its serine 49/78 phosphorylation via GSK3ß, resulting in activating PIK3CA transcription, and subsequently enhanced PI3K/AKT signaling and TNBC tumorigenesis. Taken together, Our findings identified DANCR as an pro-oncogene and uncoverd a new working pattern of lncRNA to mediate TNBC tumorigenesis, which may be a potential therapeutic target for improving treatment of TNBC.

Enhanced Electrochemical Performances of Bi2O3/rGO Nanocomposite via Chemical Bonding as Anode Materials for Lithium Ion Batteries.

Bismuth oxide/reduced graphene oxide (termed Bi2O3@rGO) nanocomposite has been facilely prepared by a solvothermal method via introducing chemical bonding that has been demonstrated by Raman and X-ray photoelectron spectroscopy spectra. Tremendous single-crystal Bi2O3 nanoparticles with an average size of ∼5 nm are anchored and uniformly dispersed on rGO sheets. Such a nanostructure results in enhanced electrochemical reversibility and cycling stability of Bi2O3@rGO composite materials as anodes for lithium ion batteries in comparison with agglomerated bare Bi2O3 nanoparticles. The Bi2O3@rGO anode material can deliver a high initial capacity of ∼900 mAh/g at 0.1C and shows excellent rate capability of ∼270 mAh/g at 10C rates (1C = 600 mA/g). After 100 electrochemical cycles at 1C, the Bi2O3@rGO anode material retains a capacity of 347.3 mAh/g with corresponding capacity retention of 79%, which is significantly better than that of bare Bi2O3 material. The lithium ion diffusion coefficient during lithiation-delithiation of Bi2O3@rGO nanocomposite has been evaluated to be around ∼10-15-10-16 cm2/S. This work demonstrates the effects of chemical bonding between Bi2O3 nanoparticles and rGO substrate on enhanced electrochemical performances of Bi2O3@rGO nanocomposite, which can be used as a promising anode alterative for superior lithium ion batteries.