Local shape descriptors for neuron segmentation.

Local shape descriptors for neuron segmentation.
复制标题

DOI:
10.1038/s41592-022-01711-z
复制
发表时间:
2023-03
期刊:
影响因子:
48
通讯作者:
Funke, Jan
Funke, Jan
中科院分区:
生物学1区
文献类型:
--
作者:
Sheridan, Arlo;Nguyen, Tri M. M.;Deb, Diptodip;Lee, Wei-Chung Allen;Saalfeld, Stephan;Turaga, Srinivas C. C.;Manor, Uri;Funke, Jan

文献摘要

参考文献

相似文献

我们针对电子显微镜体积中的神经元分割问题提出了一个辅助学习任务。辅助任务包括局部形状描述符(LSD)的预测,我们将其与传统的体素直接邻居亲和力相结合以进行神经元边界检测。形状描述符捕获有关要分割的神经元的局部统计数据,例如直径、伸长率和方向。在一项比较不同样本、成像技术和分辨率的几种现有方法的研究中,LSD 的辅助学习在一系列指标上一致提高了基于亲和力的方法的分割准确性。此外,LSD 的添加促进了基于亲和力的分割方法与神经元分割(洪水填充网络)的当前技术水平相当,同时效率提高了两个数量级——这是处理未来 PB 级数据集的关键要求。在电子显微镜数据集中的神经元分割过程中,通过预测局部形状描述符进行辅助学习可以提高效率,这对于处理不断增加的数据集非常重要。
We present an auxiliary learning task for the problem of neuron segmentation in electron microscopy volumes. The auxiliary task consists of the prediction of local shape descriptors (LSDs), which we combine with conventional voxel-wise direct neighbor affinities for neuron boundary detection. The shape descriptors capture local statistics about the neuron to be segmented, such as diameter, elongation, and direction. On a study comparing several existing methods across various specimen, imaging techniques, and resolutions, auxiliary learning of LSDs consistently increases segmentation accuracy of affinity-based methods over a range of metrics. Furthermore, the addition of LSDs promotes affinity-based segmentation methods to be on par with the current state of the art for neuron segmentation (flood-filling networks), while being two orders of magnitudes more efficient—a critical requirement for the processing of future petabyte-sized datasets. During segmentation of neurons in electron microscopy datasets, auxiliary learning via the prediction of local shape descriptors increases efficiency, which is important for the processing of datasets of ever-increasing size.
DOI: 10.3389/fncir.2018.00102
发表时间: 2018
影响因子: 3.5
作者:
Plaza SM;Funke J
通讯作者: Funke J
DOI: 10.1038/s41592-021-01183-7
发表时间: 2021-07
期刊: Nature methods
影响因子: 48
作者:
Buhmann J;Sheridan A;Malin-Mayor C;Schlegel P;Gerhard S;Kazimiers T;Krause R;Nguyen TM;Heinrich L;Lee WA;Wilson R;Saalfeld S;Jefferis GSXE;Bock DD;Turaga SC;Cook M;Funke J
通讯作者: Funke J
DOI: 10.7554/elife.73783
发表时间: 2021-12-01
期刊: eLife
影响因子: 7.7
作者:
Schneider-Mizell CM;Bodor AL;Collman F;Brittain D;Bleckert A;Dorkenwald S;Turner NL;Macrina T;Lee K;Lu R;Wu J;Zhuang J;Nandi A;Hu B;Buchanan J;Takeno MM;Torres R;Mahalingam G;Bumbarger DJ;Li Y;Chartrand T;Kemnitz N;Silversmith WM;Ih D;Zung J;Zlateski A;Tartavull I;Popovych S;Wong W;Castro M;Jordan CS;Froudarakis E;Becker L;Suckow S;Reimer J;Tolias AS;Anastassiou CA;Seung HS;Reid RC;Costa NMD
通讯作者: Costa NMD
DOI: 10.1038/s41592-018-0049-4
发表时间: 2018-08-01
期刊: NATURE METHODS
影响因子: 48
作者:
Januszewski, Michal;Kornfeld, Joergen;Jain, Viren
通讯作者: Jain, Viren
DOI: 10.7554/elife.12059
发表时间: 2016-03-18
期刊: ELIFE
影响因子: 7.7
作者:
Schneider-Mizell, Casey M.;Gerhard, Stephan;Cardona, Albert
通讯作者: Cardona, Albert