TY - GEN
T1 - 3-D density kernel estimation for counting in microscopy image volumes using 3-D image filters and random decision trees
AU - Waithe, Dominic
AU - Hailstone, Martin
AU - Lalwani, Mukesh Kumar
AU - Parton, Richard
AU - Yang, Lu
AU - Patient, Roger
AU - Eggeling, Christian
AU - Davis, Ilan
N1 - Funding Information:
We acknowledge the WIMM, The Dunn School of Pathology and the Biochemistry Department for infrastructure support. Authors are grateful to the staff of the Biomedical Services Unit at the John Racliffe Hospital site for aquatic support. We thank the Wolfson Imaging Centre Oxford and to MICRON Oxford ( http://micronoxford.com , supported by the Wellcome Trust Strategic Award 091911) for access to equipment and assistance with data acquisition and analysis. MKL and RP acknowledge funding from the BHF-Centre for Regenerative Medicine, Oxford-UK (grant ref RM/13/3/30159). The work was supported by the Wolfson Foundation, the Medical Research Council (MRC, grant number MC_UU_12010/unit programmes G0902418 and MC_UU_12025), MRC/BBSRC/ EPSRC (grant number MR/K01577X/1), and Wellcome Trust (grant ref 104924/ 14/Z/14). MH was supported through the ONBI DPhil programme in biomedical imaging technology development funded by the MRC and Engineering and Physical Sciences Research Council (EPSRC) (grant number EP/L016052/1). I.D. and R.M.P. were supported by a Wellcome Trust Senior Research Fellowship (081858) to I.D. LY was supported by a Clarendon Fund Scholarship in Humanities and by a Goodger fund Scholarship. DW was supported by funding from the MRC and EPSRC (grant number EP/L016052/1). None of the funding organisations have had any role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Publisher Copyright:
© Springer International Publishing Switzerland 2016.
PY - 2016/9/18
Y1 - 2016/9/18
N2 - We describe a means through which cells can be accurately counted in 3-D microscopy image data, using only weakly annotated images as input training material. We update an existing 2-D density kernel estimation approach into 3-D and we introduce novel 3-D features which encapsulate the 3-D neighbourhood surrounding each voxel. The proposed 3-D density kernel estimation (DKE-3-D) method, which utilises an ensemble of random decision trees, is computationally efficient and achieves state-of-the-art performance. DKE-3-D avoids the problem of discrete object identification and segmentation, common to many existing 3-D counting techniques, and we show that it outperforms other methods when quantification of densely packed and heterogeneous objects is desired. In this article we successfully apply the technique to two simulated and to two experimentally derived datasets and show that DKE-3-D has great potential in the biomedical sciences and any field where volumetric datasets are used.
AB - We describe a means through which cells can be accurately counted in 3-D microscopy image data, using only weakly annotated images as input training material. We update an existing 2-D density kernel estimation approach into 3-D and we introduce novel 3-D features which encapsulate the 3-D neighbourhood surrounding each voxel. The proposed 3-D density kernel estimation (DKE-3-D) method, which utilises an ensemble of random decision trees, is computationally efficient and achieves state-of-the-art performance. DKE-3-D avoids the problem of discrete object identification and segmentation, common to many existing 3-D counting techniques, and we show that it outperforms other methods when quantification of densely packed and heterogeneous objects is desired. In this article we successfully apply the technique to two simulated and to two experimentally derived datasets and show that DKE-3-D has great potential in the biomedical sciences and any field where volumetric datasets are used.
KW - 3-D
KW - Counting
KW - Density kernel estimation
KW - Microscopy
KW - Random decision trees
UR - http://www.scopus.com/inward/record.url?scp=84989934353&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-46604-0_18
DO - 10.1007/978-3-319-46604-0_18
M3 - Conference contribution
AN - SCOPUS:84989934353
SN - 9783319466033
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 244
EP - 255
BT - Computer Vision - ECCV 2016 Workshops, Proceedings
A2 - Hua, Gang
A2 - Jégou, Hervé
PB - Springer
T2 - Computer Vision - ECCV 2016 Workshops, Proceedings
Y2 - 8 October 2016 through 16 October 2016
ER -