3D Deep Convolutional Neural Networks in Lattice Light-Sheet Data Puncta Segmentation

机译：格子光片数据点分割中的3D深层卷积神经网络

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Recent advances in fluorescence imaging technology such as adaptive optics lattice light-sheet microscopy (AO-LLSM) allow live imaging of three dimensional (3D) tissues at super-resolution in 4D (xyzt) at seconds frame rates [1]. The resulting datasets contain novel biology but are higher dimensional and much larger (100s of GB / 3D movie) than datasets of traditional (e.g. confocal) microscopy. New tools and methods are necessary to segment and quantify these datasets such as the library pyLattice [2]. Here we present pyLattice_deepLearning, an image segmentation module based on 3D deep convolutional neural networks (3D U-Nets [3]). Using virtual machines on the Google Cloud Platform, the segmentation of a ~180MB 3D frame takes ~30s, which is ~7x faster than standard pyLattice v1.0. pyLattice_deepLearning is particularly well suited for very low signal-to-noise situations where retrieval of ~85% of the objects in a noisy image is possible while standard pyLattice retrieves ~17%.

机译：诸如自适应光学点阵光片显微镜（AO-LLSM）之类的荧光成像技术的最新进展允许以3D超高分辨率以秒帧速率对三维（3D）组织进行实时成像[1]。所得的数据集包含新颖的生物学信息，但具有更高的维度并且比传统（例如共聚焦）显微镜的数据集更大（100 GB的GB / 3D电影）。需要新的工具和方法来分割和量化这些数据集，例如库pyLattice [2]。在这里，我们介绍pyLattice_deepLearning，这是一种基于3D深度卷积神经网络（3D U-Nets [3]）的图像分割模块。使用Google Cloud Platform上的虚拟机，大约180MB的3D帧的分割大约需要30秒，比标准pyLattice v1.0快约7倍。 pyLattice_deepLearning特别适用于极低的信噪比情况，在这种情况下，可以在嘈杂的图像中检索到约85％的对象，而标准pyLattice可以检索到约17％的对象。

著录项

来源
《IEEE International Conference on Bioinformatics and Biomedicine》|2019年|2369-2372|共4页
会议地点
作者
Johannes Schöneberg; Gautham Raghupathi; Eric Betzig; David Drubin;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类
关键词
adaptive optics; biomedical optical imaging; cellular biophysics; convolutional neural nets; fluorescence; image resolution; image segmentation; medical image processing; optical microscopy;

机译：自适应光学;生物医学光学成像;细胞生物物理学;卷积神经网络;荧光;图像分辨率;图像分割;医学图像处理;光学显微镜;

相似文献

外文文献
中文文献
专利

1. Deep convolutional neural network-based segmentation and classification of difficult to define metastatic spinal lesions in 3D CT data [J] . Jiri Chmelik, Roman Jakubicek, Petr Walek, Medical image analysis . 2018,第期

机译：基于深度卷积神经网络的分割和难以定义3D CT数据转移脊柱病变的分类
2. 3D Prostate Segmentation in MR Image Using 3D Deeply Supervised Convolutional Neural Networks [J] . Yang X., Lei Y., Wang T., Medical Physics . 2018,第6期

机译：3D使用3D深深监督卷积神经网络的MR图像中的前列腺分段
3. 2D to 3D Evolutionary Deep Convolutional Neural Networks for Medical Image Segmentation [J] . Hassanzadeh Tahereh, Essam Daryl, Sarker Ruhul IEEE Transactions on Medical Imaging . 2021,第2期

机译：2D到3D进化深度卷积神经网络的医学图像分割
4. 3D Deep Convolutional Neural Networks in Lattice Light-Sheet Data Puncta Segmentation [C] . Johannes Sch?neberg, Gautham Raghupathi, Eric Betzig, IEEE International Conference on Bioinformatics and Biomedicine . 2019

机译：3D晶格轻型纸张数据中的深度卷积神经网络旁观分割
5. Plant Segmentation by Supervised Machine Learning Methods and Phenotypic Trait Extraction of Soybean Plants Using Deep Convolutional Neural Networks with Transfer Learning [D] . Adams, Jason R. 2020

机译：植物分割通过深度卷积神经网络与转移学习的豆豆植物的植物分割和表型特性
6. Deep convolutional neural networks for automated segmentation of brain metastases trained on clinical data [O] . Khaled Bousabarah, Maximilian Ruge, Julia-Sarita Brand, 2020

机译：深度卷积神经网络可根据临床数据对脑转移进行自动分割
7. 3D Deep Convolutional Neural Network-Based Ventilated Lung Segmentation Using Multi-nuclear Hyperpolarized Gas MRI [O] . Joshua R. Astley, Alberto M. Biancardi, Paul J. C. Hughes, 2020

机译：3D使用多核超极化气体MRI的深度卷积神经网络的通风肺分段

3D Deep Convolutional Neural Networks in Lattice Light-Sheet Data Puncta Segmentation

摘要

著录项

相似文献

相关主题

期刊订阅