Cell Segmentation Proposal Network for Microscopy Image Analysis

被引:61
作者
Akram, Saad Ullah [1 ,2 ]
Kannala, Juho [3 ]
Eklund, Lauri [2 ,4 ]
Heikkila, Janne [1 ]
机构
[1] Univ Oulu, Ctr Machine Vis & Signal Anal, SF-90100 Oulu, Finland
[2] Bioctr Oulu, Oulu, Finland
[3] Aalto Univ, Dept Comp Sci, Espoo, Finland
[4] Univ Oulu, Oulu Ctr Cell Matrix Res, Fac Biochem & Mol Med, SF-90100 Oulu, Finland
来源
DEEP LEARNING AND DATA LABELING FOR MEDICAL APPLICATIONS | 2016年 / 10008卷
关键词
Cell proposals; Cell segmentation; Cell detection; Convolutional neural network; Deep learning;
D O I
10.1007/978-3-319-46976-8_3
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Accurate cell segmentation is vital for the development of reliable microscopy image analysis methods. It is a very challenging problem due to low contrast, weak boundaries, and conjoined and overlapping cells; producing many ambiguous regions, which lower the performance of automated segmentation methods. Cell proposals provide an efficient way of exploiting both spatial and temporal context, which can be very helpful in many of these ambiguous regions. However, most proposal based microscopy image analysis methods rely on fairly simple proposal generation stage, limiting their performance. In this paper, we propose a convolutional neural network based method which provides cell segmentation proposals, which can be used for cell detection, segmentation and tracking. We evaluate our method on datasets from histology, fluorescence and phase contrast microscopy and show that it outperforms state of the art cell detection and segmentation methods.
引用
收藏
页码:21 / 29
页数:9
相关论文
共 15 条
[1]  
Akram S. U., 2016, ISBI
[2]  
Akram S. U., 2016, ICIP
[3]  
Arteta C, 2012, LECT NOTES COMPUT SC, V7510, P348, DOI 10.1007/978-3-642-33415-3_43
[4]   Cell Detection From Redundant Candidate Regions Under Nonoverlapping Constraints [J].
Bise, Ryoma ;
Sato, Yoichi .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2015, 34 (07) :1417-1427
[5]   ASP Vision: Optically Computing the First Layer of Convolutional Neural Networks using Angle Sensitive Pixels [J].
Chen, Huaijin G. ;
Jayasuriya, Suren ;
Yang, Jiyue ;
Stephen, Judy ;
Sivaramakrishnan, Sriram ;
Veeraraghavan, Ashok ;
Molnar, Alyosha .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :903-912
[6]  
Dai J., 2016, ADV NEURAL INFORM PR, V29, P379, DOI [DOI 10.1016/J.JPOWSOUR.2007.02.075, DOI 10.48550/ARXIV.1605.06409, DOI 10.1109/CVPR.2017.690]
[7]  
Girshick, 2015, P IEEE INT C COMP VI, DOI [10.1109/ICCV.2015.169, DOI 10.1109/ICCV.2015.169]
[8]   You Should Use Regression to Detect Cells [J].
Kainz, Philipp ;
Urschler, Martin ;
Schulter, Samuel ;
Wohlhart, Paul ;
Lepetit, Vincent .
MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION, PT III, 2015, 9351 :276-283
[9]  
Magnusson K. E. G., 2012, ISBI
[10]   A benchmark for comparison of cell tracking algorithms [J].
Maska, Martin ;
Ulman, Vladimir ;
Svoboda, David ;
Matula, Pavel ;
Matula, Petr ;
Ederra, Cristina ;
Urbiola, Ainhoa ;
Espana, Tomas ;
Venkatesan, Subramanian ;
Balak, Deepak M. W. ;
Karas, Pavel ;
Bolckova, Tereza ;
Streitova, Marketa ;
Carthel, Craig ;
Coraluppi, Stefano ;
Harder, Nathalie ;
Rohr, Karl ;
Magnusson, Klas E. G. ;
Jalden, Joakim ;
Blau, Helen M. ;
Dzyubachyk, Oleh ;
Krizek, Pavel ;
Hagen, Guy M. ;
Pastor-Escuredo, David ;
Jimenez-Carretero, Daniel ;
Ledesma-Carbayo, Maria J. ;
Munoz-Barrutia, Arrate ;
Meijering, Erik ;
Kozubek, Michal ;
Ortiz-de-Solorzano, Carlos .
BIOINFORMATICS, 2014, 30 (11) :1609-1617