国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁> 美國衛(wèi)生研究院文獻>eLife >CEM500K a large-scale heterogeneous unlabeled cellular electron microscopy image dataset for deep learning
【2h】

CEM500K a large-scale heterogeneous unlabeled cellular electron microscopy image dataset for deep learning

機譯:CEM500K一個大型異構(gòu)未標(biāo)記的蜂窩電子顯微鏡圖像數(shù)據(jù)集用于深度學(xué)習(xí)

代理獲取
本網(wǎng)站僅為用戶提供外文OA文獻查詢和代理獲取服務(wù),本網(wǎng)站沒有原文。下單后我們將采用程序或人工為您竭誠獲取高質(zhì)量的原文,但由于OA文獻來源多樣且變更頻繁,仍可能出現(xiàn)獲取不到、文獻不完整或與標(biāo)題不符等情況,如果獲取不到我們將提供退款服務(wù)。請知悉。

摘要

Automated segmentation of cellular electron microscopy (EM) datasets remains a challenge. Supervised deep learning (DL) methods that rely on region-of-interest (ROI) annotations yield models that fail to generalize to unrelated datasets. Newer unsupervised DL algorithms require relevant pre-training images, however, pre-training on currently available EM datasets is computationally expensive and shows little value for unseen biological contexts, as these datasets are large and homogeneous. To address this issue, we present CEM500K, a nimble 25 GB dataset of 0.5 × 106 unique 2D cellular EM images curated from nearly 600 three-dimensional (3D) and 10,000 two-dimensional (2D) images from >100 unrelated imaging projects. We show that models pre-trained on CEM500K learn features that are biologically relevant and resilient to meaningful image augmentations. Critically, we evaluate transfer learning from these pre-trained models on six publicly available and one newly derived benchmark segmentation task and report state-of-the-art results on each. We release the CEM500K dataset, pre-trained models and curation pipeline for model building and further expansion by the EM community. Data and code are available at https://www.ebi.ac.uk/pdbe/emdb/empiar/entry/10592/ and https://git.io/JLLTz.
機譯:蜂窩電子顯微鏡(EM)數(shù)據(jù)集的自動分割仍然是一個挑戰(zhàn)。監(jiān)督依賴于興趣區(qū)域(ROI)注釋的方法(ROI)注釋的方法,該模型無法概括到不相關(guān)的數(shù)據(jù)集。更新的無監(jiān)督的DL算法需要相關(guān)的預(yù)訓(xùn)練圖像,然而,目前可用的EM數(shù)據(jù)集的預(yù)訓(xùn)練是計算昂貴的并且顯示出看不見的生物背景的價值很小,因為這些數(shù)據(jù)集是大而均勻的。為了解決這個問題,我們呈現(xiàn)CEM500K,一個靈活的25 GB數(shù)據(jù)集0.5×106唯一的2D蜂窩EM圖像,從> 100個無關(guān)的成像項目中從近600個三維(3D)和10,000二維(2D)圖像中靜電。我們展示了在CEM500K預(yù)培訓(xùn)的模型學(xué)習(xí)具有生物學(xué)相關(guān)和有意義的圖像增強的功能。批判性地,我們在六個公共可用的六種訓(xùn)練有素的模型和一個新導(dǎo)出的基準(zhǔn)細(xì)分任務(wù)中評估轉(zhuǎn)移學(xué)習(xí),并在每個新派生的基準(zhǔn)細(xì)分任務(wù)中報告最先進的結(jié)果。我們釋放CEM500K DataSet,預(yù)先培訓(xùn)的模型和策策管道,用于模型建設(shè)和EM社區(qū)進一步擴展。數(shù)據(jù)和代碼可在https://www.ebi.ac.uk/pdbe/emdb/empiar/entry/10592/和https://git.io/jlltz。

著錄項

  • 期刊名稱 eLife
  • 作者

    Ryan Conrad; Kedar Narayan;

  • 作者單位
  • 年(卷),期 2021(-1),-1
  • 年度 2021
  • 頁碼 -1
  • 總頁數(shù) 25
  • 原文格式 PDF
  • 正文語種
  • 中圖分類
  • 關(guān)鍵詞

    None;

    機譯:沒有任何;

相似文獻

  • 外文文獻
  • 中文文獻
  • 專利
代理獲取

客服郵箱:kefu@zhangqiaokeyan.com

京公網(wǎng)安備:11010802029741號 ICP備案號:京ICP備15016152號-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有
  • 客服微信

  • 服務(wù)號