Abstract
White blood cells, which have an important role in the human immune system, protect the body against various viruses, harmful bacteria and infections. If there are not enough white blood cells in the blood, it results in leukopenia. When white blood cells are examined under a microscope, their change in structure and shape indicates some diseases. When experts clinically examine these images, various problems may arise due to individual misinterpretations. In this study, a diagnostic model based on convolutional neural network (CNN), local interpretable model agnostic annotations (LIME) and minimum redundancy maximum association (mRMR) methods is proposed for the detection of four different white blood cells. For this purpose, firstly, after the deep features of the images were extracted with SqueezeNet CNN, the important regions of the images for classification purposes were determined by the LIME method and the distinctive features of the images were obtained. The features obtained with the SqueezeNet CNN model were also obtained with the mRMR feature selection algorithm. Various feature sets obtained by combining the features obtained with the LIME algorithm are classified with support vector machines. As a result, the accuracy rate of the proposed model for the diagnosis of white blood cells was 95.88%. Selecting the SqueezeNet features with the mRMR method and supporting them with LIME features positively affected the performance results.
Similar content being viewed by others
References
Kutlu, H., Avci, E., Özyurt, F.: White blood cells detection and classification based on regional convolutional neural network. Med. Hyp. 135, 109472 (2020)
Baydilli, Y.Y., Atila, Ü.: Classification of white blood cells using capsule networks. Comput. Med. Imag. Graph. 80, 101699 (2020)
Toğaçar, M., Ergen, B., Cömert, Z.: Classification of white blood cells using deep features obtained from convolutional neural network models based on the combination of feature selection methods. Appl. Soft Comput. 97, 106810 (2020)
Başaran, E., Cömert, Z., Çelik, Y.: Convolutional neural network approach for automatic tympanic membrane detection and classification. Biomed. Sig. Process. Control 56, 101734 (2020)
Toğaçar, M., Ergen, B., Cömert, Z., Özyurt, F.: A deep feature learning model for pneumonia detection applying a combination of mrmr feature selection and machine learning models. IRBM 41, 4212–4222 (2020)
Budak, Ü., Cömert, Z., Çıbuk, M., Şengür, A.: DCCMED-Net: densely connected and concatenated multi encoder-decoder CNNs for retinal vessel extraction from fundus images. Med. Hypo. 134, 109426 (2020)
Ekiz, A., Kaplan, K., Ertunç, H.M.: Classification of white blood cells using CNN and Con-SVM. In: 2021 29th Signal Processing and Communications Applications Conference (SIU), pp. 1–4 (2021) (Online)
Bani-Hani, D., Khan, N., Alsultan, F., Karanjkar, S., Nagarur, N.: Classification of leucocytes using convolutional neural network optimized through genetic algorithm. In: Proceedings of the 7th Annual World Conference of the Society for Industrial and Systems Engineering. Binghamton, NY, USA (2018)
Özyurt, F.: A fused CNN model for WBC detection with MRMR feature selection and extreme learning machine. Soft. Comput. 24(11), 8163–8172 (2020)
Liang, G., Hong, H., Xie, W., Zheng, L.: Combining convolutional neural network with recursive neural network for blood cell image classification. IEEE Access 6, 36188–36197 (2018)
Mooney, P.: Blood cell images. https://www.kaggle.com/paultimothymooney/blood-cells (2021)
Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., Keutzer, K.: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size, pp. 1–13. arXiv Preprint (2016)
Shin, J., Chang, Y.K., Heung, B., Nguyen, T., Quang, G., Price, W., Al-Mallahi, A.: A deep learning approach for RGB image-based powdery mildew disease detection on strawberry leaves. Comput. Electron. Agric. 183, 106042 (2021)
Hamid, N., Sumait, B.S., Bakri, B.I., Al-Qershi, O.: Enhancing visual quality of spatial image steganography using SqueezeNet deep learning network. Multim. Tools Appl. 80, 1–17 (2021)
Ribeiro, M.T., Singh, S., Guestrin, C.: ‘Why should i trust you?’: explaining the predictions of any classifier”. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1135–1144. New York, NY, United States (2016)
Liu, X.: A robust low data solution: dimension prediction of semiconductor nanorods. Comput. Chem. Eng. 150, 107315 (2021)
Parmar, J., Das, P., Dave, S.M.: A machine learning approach for modelling parking duration in urban land-use. Phys. A Stat. Mech. Appl. 572, 125873 (2021)
Schönhof, R., Werner, A., Elstner, J., Zopcsak, B., Awad, R., Huber, M.: Feature visualization within an automated design assessment leveraging explainable artificial intelligence methods. Procedia CIRP 100, 331–336 (2021)
Peng, H., Long, F., Ding, C.: Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans. Patt. Anal. Mach. Intell. 27(8), 1226–1238 (2005)
Ramírez-Gallego, S.: Fast-mRMR: Fast minimum redundancy maximum relevance algorithm for high-dimensional big data. Int. J. Intell. Syst. 32(2), 134–152 (2017)
Toğaçar, M., Ergen, B., Cömert, Z.: Detection of lung cancer on chest CT images using minimum redundancy maximum relevance feature selection method with convolutional neural networks. Biocybern. Biomed. Eng. 40, 23–39 (2019)
Başaran, E., Cömert, Z., Çelik, Y., Budak, Ü., Şengür, A.: Otitis Media Diagnosis Model for Tympanic Membrane Images Processed in Two-Stage Processing Blocks. IOP Publishing, Varun Bajaj , Sinha G R,14-1 (2020)
Ahmed, Y.A., Koçer, B., Huda, S., Saleh, B.A., Hassan, M.M.: A system call refinement-based enhanced minimum redundancy maximum relevance method for ransomware early detection. J. Netw. Comput. Appl. 167, 102753 (2020)
Guo, Y., Zhang, Z., Tang, F.: Feature selection with kernelized multi-class support vector machine. Pattern Recogn. 117, 107988 (2021)
Vapnik, V.: The support vector method of function estimation. In: Nonlinear Modeling, pp. 55–85. Springer (1998)
Jha, R.K., Swami, P.D.: Fault diagnosis and severity analysis of rolling bearings using vibration image texture enhancement and multiclass support vector machines
Appl. Acoust. 182, 108243 (2021)
Başaran, E., Cömert, Z., Sengur, A., Budak, Ü., Celık, Y., Toğaçar, M.: Normal ve Kronik Hastalıklı Orta Kulak İmgelerinin Evrişimsel Sinir Ağları Yöntemiyle Tespit Edilmesi. Türkiye Bilişim Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi 13(1), 1–10 (2020)
Janssens, A.C.J.W.: Martens FK. Reflection on modern methods: revisiting the area under the ROC Curve. Int. J. Epidemiol. 49(4), 1397–1403 (2020)
Türk, E., Süzek, B.E.: Taxonomic diversity-based domain interaction prediction. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi 25(2), 215–222 (2019)
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author declares that there is no conflict of interest related to this paper.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Başaran, E. Classification of white blood cells with SVM by selecting SqueezeNet and LIME properties by mRMR method. SIViP 16, 1821–1829 (2022). https://doi.org/10.1007/s11760-022-02141-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-022-02141-2