Faster R-CNN with improved anchor box for cell recognition

Tingxi Wen; Hanxiao Wu; Yu Du; Chuanbo Huang; Tingxi Wen; Hanxiao Wu; Yu Du; Chuanbo Huang

doi:10.3934/mbe.2020395

Mathematical Biosciences and Engineering

2020, Volume 17, Issue 6: 7772-7786. doi: 10.3934/mbe.2020395

Previous Article Next Article

Research article Special Issues

Faster R-CNN with improved anchor box for cell recognition

1.
College of Engineering, Huaqiao University, Quanzhou 362021, China
2.
Postdoctoral Workstation of Linewell Software Company Limited, Quanzhou 362000, China
3.
Nlinks Technology Group Co., LTD, China

Received: 13 July 2020 Accepted: 20 October 2020 Published: 06 November 2020

As the basic units of the human body structure and function, cells have a considerable influence on maintaining the normal work of the human body. In medical diagnosis, cell examination is an important part of understanding the human function. Incorporating cell examination into medical diagnosis would greatly improve the efficiency of pathological research and patient treatment. In addition, cell segmentation and identification technology can be used to quantitatively analyze and study cellular components at the molecular level. It is conducive to the study of the pathogenesis of diseases and to the formulation of highly effective disease treatment programs. However, because cells are of diverse types, their numbers are huge, and they exist in the order of micrometers, detecting and identifying cells without using a deep learning-based computer program are extremely difficult. Therefore, the use of computers to study and analyze cells has a certain practical value. In this work, target detection theory using deep learning is applied to cell detection. A target recognition network model is built based on the faster region-based convolutional neural network (R-CNN) algorithm, and the anchor box is designed in accordance with the characteristics of the data set. Different design methods influence cell detection results. Using the object detection method based on our novel faster R-CNN framework to detect the cell image can help improve the speed and accuracy of cell detection. The method has considerable advantages in dealing with the identification of flowing cells.
- faster R-CNN,
- anchor box,
- convolutional neural network,
- object detection,
- cell recognition
Citation: Tingxi Wen, Hanxiao Wu, Yu Du, Chuanbo Huang. Faster R-CNN with improved anchor box for cell recognition[J]. Mathematical Biosciences and Engineering, 2020, 17(6): 7772-7786. doi: 10.3934/mbe.2020395

Related Papers:

Abstract

As the basic units of the human body structure and function, cells have a considerable influence on maintaining the normal work of the human body. In medical diagnosis, cell examination is an important part of understanding the human function. Incorporating cell examination into medical diagnosis would greatly improve the efficiency of pathological research and patient treatment. In addition, cell segmentation and identification technology can be used to quantitatively analyze and study cellular components at the molecular level. It is conducive to the study of the pathogenesis of diseases and to the formulation of highly effective disease treatment programs. However, because cells are of diverse types, their numbers are huge, and they exist in the order of micrometers, detecting and identifying cells without using a deep learning-based computer program are extremely difficult. Therefore, the use of computers to study and analyze cells has a certain practical value. In this work, target detection theory using deep learning is applied to cell detection. A target recognition network model is built based on the faster region-based convolutional neural network (R-CNN) algorithm, and the anchor box is designed in accordance with the characteristics of the data set. Different design methods influence cell detection results. Using the object detection method based on our novel faster R-CNN framework to detect the cell image can help improve the speed and accuracy of cell detection. The method has considerable advantages in dealing with the identification of flowing cells.

References

[1]	S. D. Olabarriaga, J. G. Snel, C. P. Botha, R. G. Belleman, Integrated support for medical image analysis methods: from development to clinical application, IEEE Trans. Inf. Technol. Biomed., 11 (2007), 47-57. doi: 10.1109/TITB.2006.874929
[2]	F. Yang, M. A. Mackey, F. Ianzini, G. M. Gallardo, M. Sonka, Segmentation and quantitative analysis of the living tumor cells using Large Scale Digital Cell Analysis System, Med. Imaging 2004: Image Proc., 5370 (2004), 1755-1763. doi: 10.1117/12.536771
[3]	A. M. Bilek, K. C. Dee, D. P. Gaver III, Mechanisms of surface-tension-induced epithelial cell damage in a model of pulmonary airway reopening, J. Appl. Physiol., 94 (2003), 770-783.
[4]	K. Luby-Phelps, Cytoarchitecture and physical properties of cytoplasm: volume, viscosity, diffusion, intracellular surface area, Int. Rev. Cytol., 192 (2000), 189-221.
[5]	A. Csiszár, B. Hoffmann, R. Merkel, Double-shell giant vesicles mimicking gram-negative cell wall behavior during dehydration, Langmuir, 25 (2009), 5753-5761. doi: 10.1021/la8041023
[6]	K. K. L. Wong, Three-dimensional discrete element method for the prediction of protoplasmic seepage through membrane in a biological cell, J. Biomech., 65 (2017), 115-124. doi: 10.1016/j.jbiomech.2017.10.023
[7]	K. K. L. Wong, G. Fortino, D. Abbott, Deep learning-based cardiovascular image diagnosis: a promising challenge, Future Gener. Comput. Syst., 110 (2020), 802-811. doi: 10.1016/j.future.2019.09.047
[8]	K. K. L Wong., J. Wu, G. Liu, W. Huang, D. N. Ghista, Coronary arteries hemodynamics: effect of arterial geometry on hemodynamic parameters causing atherosclerosis, Med. Biol. Eng. Comput., 2020.
[9]	B. S., Gardiner, K. K. L. Wong, G. R. Joldes, A. J. Rich, C. W. Tan, A. W. Burgess, et al., Discrete element framework for modelling extracellular matrix, deformable cells and subcellular components, PLoS Comput. Biol., 11 (2015), e1004544. doi: 10.1371/journal.pcbi.1004544
[10]	R. G. Joldes, K. K. L. Wong, D. W. Smith, C. W. Tan, B. S. Gardiner, Controlling seepage in discrete particle simulations of biological systems, Comput. Methods Biomech. Biomed. Eng., 19 (2016), 1160-1170. doi: 10.1080/10255842.2015.1115022
[11]	K. Metze, R. C. Ferreira, R. L. Adam, Classification of thyroid follicular lesions based on nuclear texture features-Lesion size matters, Cytometry, 77 (2010), 1101-1102.
[12]	J. Cui, J. X. Chen, Image-based clipping, U. S. Patent, 2007.
[13]	J. Liu, B. Xu, L. Shen, J. Garibaldi, G. Qiu, HEp-2 cell classification based on a deep autoencoding-classification convolutional neural network, IEEE 14^th Int. Symp. Biomed. Imaging(ISBL 2017), (2017), 1019-1023.
[14]	S. M. Kang, J. W. L. Wan, A multiscale graph cut approach to bright-field multiple cell image segmentation using a Bhattacharyya measure, Med. Imaging 2013: Image Proc., 8669 (2013).
[15]	Y. Chen, J. W. L. Wan, Bright-field cell image segmentation by principal component pursuit with an Ncut penalization, Med. Imaging 2015: Image Proc., 9413 (2015), 94133F.
[16]	O. Ronneberger, P. Fischer, T. Brox, U-Net: convolutional networks for biomedical image segmentation, in International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer, Cham, (2015), 234-241.
[17]	P. Naylor, M. Laé, F. Reyal, T. Walter, Segmentation of nuclei in histopathology images by deep regression of the distance map, IEEE Trans. Med. Imaging, 38 (2018), 448-459.
[18]	F. Kromp, L. Fischer, E. Bozsaky, I. Ambros, W. Doerr, Taschner-Mandl S, Ambros P, Hanbury A. Deep Learning architectures for generalized immunofluorescence based nuclear image segmentation, preprint, arXiv: 1907.12975.
[19]	Y. H. Huang, T. C. Yu, P. H. Tsai, Y. X. Wang, W. L. Yang, M. Ouhyoung, Scope+: a stereoscopic video see-through augmented reality microscope, in Adjunct Proceedings of the 28th Annual ACM Symposium on User Interface Software & Technology, (2015), 33-34.
[20]	P. H. C. Chen, K. Gadepalli, R. MacDonald, Y. Liu, S. Kadowaki, K. Nagpal, et al., An augmented reality microscope with real-time artificial intelligence integration for cancer diagnosis, Nat. Med., 25 (2019), 1453-1457. doi: 10.1038/s41591-019-0539-7
[21]	S. Ren, K. He, R. Girshick, J. Sun, Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks, in Advances in Neural Information Processing Systems, (2015), 91-99.
[22]	I. Guyon, M. Nikravesh, S. Gunn, L. Zadeh, Feature Extraction, Springer Berlin Heidelberg, 2006.
[23]	U. Orhan, M. Hekim, M. Ozer, EEG signals classification using the K-means clustering and a multilayer perceptron neural network model, Expert Syst. Appl., 38 (2011), 13475-13481. doi: 10.1016/j.eswa.2011.04.149
[24]	K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2016), 770-778.
[25]	X. Xie, X. Han, Q. Liao, G. Shi, Visualization and pruning of SSD with the base network VGG16, in International Conference on Deep Learning Technologies, (2017), 90-94.

Reader Comments

Your name:*

Email:*
© 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)