A method to classify bone marrow cells with rejected option
-
Liang Guo
and
Jie Li
Abstract
Bone marrow cell morphology has always been an important tool for the diagnosis of blood diseases. Still, it requires years of experience from a suitable person. Furthermore, the outcomes of their recognition are subjective and there is no objective quantitative standard. As a result, developing a deep learning automatic classification system for bone marrow cells is extremely important. However, typical classification machine learning systems only produce classification answers, and will not refuse to generate predictions when the prediction reliability is low. It will pose a big problem in some high-risk systems such as bone marrow cell recognition. This paper proposes a bone marrow cell classification method with rejected option (CMWRO) to classify 11 bone marrow cells. CMWRO is based on convolutional neural networks, ICP and SoftMax (CNN-ICP-SoftMax), containing a classifier with rejected option. When the rejected rate (RR) of tested samples is 0.3143, it can ensure that the precision, sensitivity, accuracy of the accepted samples reach 0.9921, 0.9917 and 0.9944 respectively. And the rejected samples will be handled by other ways, such as identified by doctors. Besides, the method has a good filtering effect on cell types that the classifier is not trained, such as abnormal cells and cells with less sample distribution. It can reach more than 82% in filtering efficiency. CMWRO improves the doctors’ trust in the results of accepted samples to a certain extent. They only need to carefully identify the samples that CMWRO refuses to recognize, and finally combines the two results. It can greatly improve the efficiency and accuracy of bone marrow cell recognition.
Funding source: Science and Technology Program of Guangzhou
Award Identifier / Grant number: 202002030165
Award Identifier / Grant number: 2019050001
Funding source: Featured Innovation Project of Guangdong Education Department
Award Identifier / Grant number: 2019KTSCX034
Funding source: Young Innovative Talents Project in Universities of Guangdong Province
Award Identifier / Grant number: 2018KQNCX057
Funding source: Special Funds for the Cultivation of Guangdong College Students’ Scientific and Technological Innovation (“Climbing Program” Special Funds)
Award Identifier / Grant number: pdjh2021b0134
Funding source: Department of Science and Technology of Guangdong Province
Award Identifier / Grant number: 2018B030323017
Funding source: Young Scholar Foundation of South China Normal University
Award Identifier / Grant number: 19KJ13
Funding source: National Key Research and Development Program of China
Award Identifier / Grant number: 2017YFB1104500
Funding source: National Natural Science Foundation of China
Award Identifier / Grant number: 62005081
Funding source: Key-Area Research and Development Program of Guangdong Province
Award Identifier / Grant number: 2020B090922006
-
Research funding: This work was supported by Department of Science and Technology of Guangdong Province (2018B030323017); Key-Area Research and Development Program of Guangdong Province (No. 2020B090922006); National Key Research and Development Program of China (2017YFB1104500); National Natural Science Foundation of China (62005081); Special Funds for the Cultivation of Guangdong College Students’ Scientific and Technological Innovation (“Climbing Program” Special Funds) (No. pdjh2021b0134); Science and Technology Program of Guangzhou (No. 202002030165, 2019050001); Featured Innovation Project of Guangdong Education Department (No. 2019KTSCX034); Young Innovative Talents Project in Universities of Guangdong Province (No. 2018KQNCX057); Young Scholar Foundation of South China Normal University (No. 19KJ13).
-
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: Authors state no conflict of interest.
-
Informed consent: Informed consent was obtained from all individuals included in this study.
-
Ethical approval: The conducted research is not related to either human or animals use.
References
1. Focosi, D. Bone marrow aspiration and biopsy. N Engl J Med 2010;362:182–3. https://doi.org/10.1056/NEJMc0910593.Search in Google Scholar PubMed
2. Wu, YY, Huang, TC, Ye, RH, Fang, WH, Lai, SW, Chang, PY, et al.. A Hematologist - level deep learning algorithm (BMSNet) for assessing the morphologies of single nuclear balls in bone marrow smears: algorithm development. JMIR Med Inform 2020;8. https://doi.org/10.2196/15963.Search in Google Scholar PubMed PubMed Central
3. Lee, SH, Erber, WN, Porwit, A, Tomonaga, M, Peterson, LC. ICSH guidelines for the standardization of bone marrow specimens and reports. Int J Lab Hematol 2008;30:349–64. https://doi.org/10.1111/j.1751-553x.2008.01100.x.Search in Google Scholar PubMed
4. Shi, T. Research on Key Techniques of Automatic Examination of Bone Marrow Cell Morphology for Acute Leukemia [Master thesis]. South China University of Technology; 2018.Search in Google Scholar
5. Deng, L. Artificial intelligence in the rising wave of deep learning: the historical path and future outlook. IEEE Signal Process Mag 2018;35:180–77 https://doi.org/10.1109/msp.2017.2762725.Search in Google Scholar
6. Lu, H, Li, Y, Chen, M, Kim, H, Serikawa, S. Brain intelligence: go beyond artificial intelligence. Mobile Netw Appl 2018;23:368–75. https://doi.org/10.1007/s11036-017-0932-8.Search in Google Scholar
7. Gurkaynak, G, Yilmaz, I, Haksever, G. Stifling artificial intelligence: human perils. Comput Law Secur Rev 2016;32:749–58. https://doi.org/10.1016/j.clsr.2016.05.003.Search in Google Scholar
8. Zang, Y, Zhang, F, Di, C, Zhu, D. Advances of flexible pressure sensors toward artificial intelligence and health care applications. Mater Horiz 2015;2:140–56. https://doi.org/10.1039/c4mh00147h.Search in Google Scholar
9. Chakrabarty, S. Context-aware song recommendation system. In: 3rd National conference on trends in communication, cloud, and big data. CCB, Majitar, India, 2018.10.1007/978-981-15-1624-5_16Search in Google Scholar
10. Kumar, S. Movie recommendation system using sentiment analysis from microblogging data. IEEE Trans Comput Soc Syst 2020;7:915–23. https://doi.org/10.1109/tcss.2020.2993585.Search in Google Scholar
11. Sharma, N. Movie recommendation systems: a brief overview. In: ICCCM’20: Proceedings of the 8th international conference on computer and communications management; 2020.10.1145/3411174.3411194Search in Google Scholar
12. Baghel, N. Automatic diagnosis of multiple cardiac diseases from PCG signals using convolutional neural network. Comput Methods Progr Biomed 2020;197. https://doi.org/10.1016/j.cmpb.2020.105750.Search in Google Scholar PubMed
13. Al-Zinati, M. Enabling multiple health security threats detection using mobile edge computing. Simulat Model Pract Theor 2020;101. https://doi.org/10.1016/j.simpat.2019.101957.Search in Google Scholar
14. Asif, A. Generalized Learning with Rejection for Classification and Regression Problems. arXiv 2019.Search in Google Scholar
15. Acevedo, A, Alferez, S, Merino, A, Puigvi, L,Rodellar, J. Recognition of peripheral blood cell images using convolutional neural networks. Comput Methods Progr Biomed 2019; 180. https://doi.org/10.1016/j.cmpb.2019.105020.Search in Google Scholar PubMed
16. Zhang, CH, Wu, SS, Lu, ZM, Shen, YJ, Wang, J, Huang, P, et al.. Hybrid adversarial-discriminative network for leukocyte classification in leukemia. Med Phys 2020;47:3732–44. https://doi.org/10.1002/mp.14144.Search in Google Scholar PubMed
17. Rezatofighi, SH, Soltanian-Zadeh, H. Automatic recognition of five types of white blood cells in peripheral blood. Comput Med Imag Graph 2011;35:333–43. https://doi.org/10.1016/j.compmedimag.2011.01.003.Search in Google Scholar PubMed
18. Wang, YP, Cao, Y. A computer-assisted human peripheral blood leukocyte image classification method based on Siamese network. Med Biol Eng Comput 2020;58:1575–82. https://doi.org/10.1007/s11517-020-02180-2.Search in Google Scholar PubMed
19. Chuanyi, Z. Data-driven Meta-set Based Fine-Grained Visual Classification. arXiv 2020.Search in Google Scholar
20. Hanczar, BD, Edward, R. Classification with reject option in gene expression data. Bioinformatics 2008;24:1889–95. https://doi.org/10.1093/bioinformatics/btn349.Search in Google Scholar PubMed
21. Vovk, V. Transductive conformal predictors, In: 9th IFIP WG 12.5 International conference on artificial intelligence applications and innovations (AIAI) 2013. https://doi.org/10.1007/978-3-642-41142-7_36.Search in Google Scholar
22. Linusson, H. Reliable confidence predictions using conformal prediction. In: 20th Pacific-asia conference on knowledge discovery and data mining (PAKDD), Univ Auckland, Auckland, New Zealand; 2016.10.1007/978-3-319-31753-3_7Search in Google Scholar
23. Vovk, V, Gammerman, A, Shafer, G. Algorithmic Learning in a Random World. Springer-Verlag 2006.Search in Google Scholar
24. Papadopoulos, H, Proedrou, K, Vovk, V, Gammerman, A. Inductive confidence machines for regression. Proceedings of the machine learning: ECML, 2002;2430:345–56. https://doi.org/10.1007/3-540-36755-1_29.Search in Google Scholar
25. Linusson, H. Efficiency comparison of unstable transductive and inductive conformal classifiers. In: IFIP International conference on artificial intelligence applications and innovations; 2014.10.1007/978-3-662-44722-2_28Search in Google Scholar
26. Khan, S, Islam, N, Jan, Z, Din, IU, Rodrigues, J. A novel deep learning based framework for the detection and classification of breast cancer using transfer learning. Pattern Recogn Lett 2019;125:1–6. https://doi.org/10.1016/j.patrec.2019.03.022.Search in Google Scholar
27. Yang, L, Hanneke, S, Carbonell, J. A theory of transfer learning with applications to active learning. Mach Learn 2013;90:161–89. https://doi.org/10.1007/s10994-012-5310-y.Search in Google Scholar
28. Gao, H. Densely connected convolutional networks. In: 2017 IEEE Conference on computer vision and pattern recognition; 2017.Search in Google Scholar
29. Breiman, L. Random forests. Mach Learn 2001;45:5–32. https://doi.org/10.1023/a:1010933404324.10.1023/A:1010933404324Search in Google Scholar
30. Chang, CC, Lin, CJ. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol 2011:2. https://doi.org/10.1145/1961189.1961199.Search in Google Scholar
31. https://pytorch.org.Search in Google Scholar
32. Szegedy, C. Going deeper with convolutions. In: 2015 IEEE Conference on computer vision and pattern recognition; 2015.10.1109/CVPR.2015.7298594Search in Google Scholar
33. Simonyan, K, Zisserman, A. Very deep convolutional networks for large-Scale image recognition. In: Computer vision and pattern recognition; 2014.Search in Google Scholar
34. He, K, Zhang, X, Ren, S, Sun, J. Deep residual learning for image recognition. In: 2016 IEEE Conference on computer vision and pattern recognition; 2016.10.1109/CVPR.2016.90Search in Google Scholar
© 2022 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Research Articles
- Towards fully automated detection of epileptic disorders: a novel CNSVM approach with Clough–Tocher interpolation
- Frontal-occipital network alterations while viewing 2D & 3D movies: a source-level EEG and graph theory approach
- Real-time recognition of different imagined actions on the same side of a single limb based on the fNIRS correlation coefficient
- Simultaneous validation of wearable motion capture system for lower body applications: over single plane range of motion (ROM) and gait activities
- Embedded system design for classification of COPD and pneumonia patients by lung sound analysis
- Investigation of the impact of electromagnetic fields emitted close to the head by smart glasses
- A method to classify bone marrow cells with rejected option
Articles in the same Issue
- Frontmatter
- Research Articles
- Towards fully automated detection of epileptic disorders: a novel CNSVM approach with Clough–Tocher interpolation
- Frontal-occipital network alterations while viewing 2D & 3D movies: a source-level EEG and graph theory approach
- Real-time recognition of different imagined actions on the same side of a single limb based on the fNIRS correlation coefficient
- Simultaneous validation of wearable motion capture system for lower body applications: over single plane range of motion (ROM) and gait activities
- Embedded system design for classification of COPD and pneumonia patients by lung sound analysis
- Investigation of the impact of electromagnetic fields emitted close to the head by smart glasses
- A method to classify bone marrow cells with rejected option
