Faster Mean-shift: GPU-accelerated clustering for cosine embedding-based cell segmentation and tracking.
Faster Mean-shift: GPU-accelerated clustering for cosine embedding-based cell segmentation and tracking.
复制标题
DOI：
10.1016/j.media.2021.102048
复制
发表时间：
2021-07
影响因子：
10.9
通讯作者：
Huo, Yuankai
中科院分区：
文献类型：
--
作者：
Zhao, Mengyang;Jha, Aadarsh;Liu, Quan;Millis, Bryan A.;Mahadevan-Jansen, Anita;Lu, Le;Landman, Bennett A.;Tyska, Matthew J.;Huo, Yuankai
近年来，基于单步嵌入的深度学习算法在细胞分割和跟踪中受到越来越多的关注。与传统的先分割后关联的两阶段算法相比，单阶段算法不仅同时实现了一致的实例细胞分割和跟踪，而且在区分边界和重叠上的模糊像素时获得了优越的性能。然而，基于嵌入的算法的部署受到推理速度慢(例如，每帧≈1-2分钟)的限制。在这项研究中，我们提出了一种新的更快的Mean-Shift算法，解决了基于嵌入的细胞分割和跟踪的计算瓶颈。与以往GPU加速的快速均值漂移算法不同，引入了一种新的在线种子优化策略(OSOP)，以自适应地确定最小种子数，加快计算速度，节省GPU内存。通过对ISBI细胞跟踪挑战中的四个队列进行嵌入仿真和经验验证，与现有的基于嵌入的细胞实例分割和跟踪算法相比，提出的Mean-Shift算法获得了7-10倍的加速比。与其他GPU基准测试相比，我们更快的Mean-Shift算法也获得了最高的计算速度，并优化了内存消耗。较快的Mean-Shift是一种即插即用的模型，可以应用于其他基于像素嵌入的聚类推理，用于医学图像分析。(即插即用模式已公开提供：https://github.com/masqm/Faster-Mean-Shift)
Recently, single-stage embedding based deep learning algorithms gain increasing attention in cell segmentation and tracking. Compared with the traditional ”segment-then-associate” two-stage approach, a single-stage algorithm not only simultaneously achieves consistent instance cell segmentation and tracking but also gains superior performance when distinguishing ambiguous pixels on boundaries and overlaps. However, the deployment of an embedding based algorithm is restricted by slow inference speed (e.g., ≈1-2 mins per frame). In this study, we propose a novel Faster Mean-shift algorithm, which tackles the computational bottleneck of embedding based cell segmentation and tracking. Different from previous GPU-accelerated fast mean-shift algorithms, a new online seed optimization policy (OSOP) is introduced to adaptively determine the minimal number of seeds, accelerate computation, and save GPU memory. With both embedding simulation and empirical validation via the four cohorts from the ISBI cell tracking challenge, the proposed Faster Mean-shift algorithm achieved 7-10 times speedup compared to the state-of-the-art embedding based cell instance segmentation and tracking algorithm. Our Faster Mean-shift algorithm also achieved the highest computational speed compared to other GPU benchmarks with optimized memory consumption. The Faster Mean-shift is a plug-and-play model, which can be employed on other pixel embedding based clustering inference for medical image analysis. (Plug-and-play model is publicly available: https://github.com/masqm/Faster-Mean-Shift)
登录
查看更多内容
细胞跟踪精度测量基于无环的图表的比较。
影响因子：
3.7
作者：
Matula P;Maška M;Sorokin DV;Matula P;Ortiz-de-Solórzano C;Kozubek M
通讯作者：
Kozubek M
影响因子：
10.6
作者：
Debeir, O;Van Ham, P;Decaestecker, C
通讯作者：
Decaestecker, C
影响因子：
5.1
作者：
Huang, Fang;Chen, Yinjie;Fan, Guangsong
通讯作者：
Fan, Guangsong
DOI：
10.1083/jcb.141.5.1147
发表时间：
1998-06-01
期刊：
The Journal of cell biology
影响因子：
--
作者：
Allen WE;Zicha D;Ridley AJ;Jones GE
通讯作者：
Jones GE
影响因子：
8.6
作者：
Czirok, A;Schlett, K;Vicsek, T
通讯作者：
Vicsek, T