White Blood Cell Classification Using Convolutional Neural Network

Year 2022, , 32 - 41, 21.06.2022

Shamriz Nahzat , Ferhat Bozkurt , Mete Yağanoğlu

https://doi.org/10.53525/jster.1018213

Cited By: 4

Abstract

White blood cells (WBCs) are a key element of the immune system and demonstrate resistance to a variety of illnesses, quantitative and qualitative examination of various kinds of white blood cells is critical. Counting and categorizing the types of WBCs can help doctors detect and treat different illnesses. As a result, one of the most important steps in analyzing and testing blood samples is counting and categorizing various types of WBCs.
The main purpose of this study is to provide a CNN based model for processing of WBCs with the aim of classifying the type of these cells. Kaggle white blood cells images were used in this article, we built a CNN-based model for classifying white blood cell types and assessed the model's performance using several optimizers. We have seen that the RMSprop optimizer shows the best result in our proposed model. We have compared four pre-trained models such as MobileNetV2, DenseNet121, InceptionV3 and ResNet50 with our proposed model.
Compared to four pre-trained CNN models, and other related studies, our proposed model with the lowest number of trainable parameters and training time shows the great results with 99.5% accuracy, 99% recall, 99% precision, and 99% F1 score.

Keywords

CNN, WBC, DenseNet121

Evrişimsel Sinir Ağı Kullanarak Beyaz Kan Hücresi Sınıflandırması

Abstract

Beyaz kan hücreleri, bağışıklık sisteminin önemli bir unsurudur ve çeşitli hastalıklara karşı direnç gösterir, çeşitli türlerdeki beyaz kan hücrelerinin nicel ve nitel olarak incelenmesi çok önemlidir. BKH türlerini saymak ve sınıflandırmak, doktorların farklı hastalıkları tespit etmesine ve tedavi etmesine yardımcı olabilir. Sonuç olarak, kan numunelerini analiz etme ve test etmedeki en önemli adımlardan biri, çeşitli beyaz kan hücrelerinin sayılması ve sınıflandırılmasıdır.
Bu çalışmanın temel amacı, bu hücrelerin tipini sınıflandırmak amacıyla beyaz kan hücrelerinin işlenmesi için CNN tabanlı bir model sağlamaktır. Bu makalede Kaggle beyaz kan hücresi görüntüleri kullanıldı, beyaz kan hücresi türlerini sınıflandırmak için CNN tabanlı bir model oluşturduk ve birkaç optimize edici kullanarak modelin performansını değerlendirdik. Önerilen modelimizde en iyi sonucu RMSprop optimize edicisinin gösterdiğini gördük. MobileNetV2, DenseNet121, InceptionV3 ve ResNet50 gibi önceden eğitilmiş dört modeli önerilen modelimiz ile karşılaştırdık.
Önceden eğitilmiş dört CNN modeli ve diğer ilgili çalışmalarla karşılaştırıldığında, en düşük sayıda eğitilebilir parametreye sahip önerilen modelimiz %99,5 doğruluk, %99 kesinlik, %99 geri çağırma ve %99 F1 puanı ile mükemmel sonuç göstermektedir.

Keywords

CNN, BKH, DenseNet121

References

• Bohr A., & Memarzadeh K., “The Rise of Artificial Intelligence in Healthcare Applications.” Artificial Intelligence in Healthcare, pp. 25–60. Crossref, doi:10.1016/b978-0-12-818438-7.00002-2, (2020).
• Gurcan M., Boucheron L., Can A., Madabhushi A., Rajpoot N., & Yener B., “Histopathological Image Analysis: A Review.” IEEE Reviews in Biomedical Engineering, vol. 2, pp. 147–71. Crossref, doi:10.1109/rbme.2009.2034865, (2009).
• Mohapatra S., Patra D., & Satpathy S., “An Ensemble Classifier System for Early Diagnosis of Acute Lymphoblastic Leukemia in Blood Microscopic Images.” Neural Computing and Applications, vol. 24, no. 7–8, pp. 1887–904. Crossref, doi:10.1007/s00521-013-1438-3, (2013).
• Sadeghian F., Seman Z., Ramli A. R., Abdul Kahar B. H., & Saripan M. I., “A Framework for White Blood Cell Segmentation in Microscopic Blood Images Using Digital Image Processing.” Biological Procedures Online, vol. 11, no. 1, pp. 196–206. Crossref, doi:10.1007/s12575-009-9011-2, (2009).
• “Blood Cell Images.” Kaggle, 21 Apr. 2018, www.kaggle.com/paultimothymooney/blood-cells. Accessed (01.05.2021).
• C. Cheuque, M. Querales, R. León, R. Salas and R. Torres, "An Efficient Multi-Level Convolutional Neural Network Approach for White Blood Cells Classification", Diagnostics, vol. 12, no. 2, p. 248, 2022.
• Çınar A., & Tuncer S. A., “Classification of Lymphocytes, Monocytes, Eosinophils, and Neutrophils on White Blood Cells Using Hybrid Alexnet-GoogleNet-SVM.” SN Applied Sciences, vol. 3, no. 4. Crossref, doi:10.1007/s42452-021-04485-9, (2021).
• A. Ekiz, K. Kaplan and H. Ertunc, "Classification of White Blood Cells Using CNN and Con-SVM", 2021 29th Signal Processing and Communications Applications Conference (SIU), 2021.
• 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.” Applied Soft Computing, vol. 97, p. 106810. Crossref, doi:10.1016/j.asoc.2020.106810, (2020).
• F. Özyurt, "A fused CNN model for WBC detection with MRMR feature selection and extreme learning machine", Soft Computing, vol. 24, no. 11, pp. 8163-8172, 2019.
• Tiwari P., Qian J., Li Q., Wang B., Gupta D., Khanna A., Rodrigues J.J., de Albuquerque V. H. C., “Detection of Subtype Blood Cells Using Deep Learning.” Cognitive Systems Research, vol. 52, pp. 1036–44. Crossref, doi:10.1016/j.cogsys.2018.08.022, (2018).
• C. Lin, C. Lin and S. Wang, "Integrated Image Sensor and Light Convolutional Neural Network for Image Classification", Mathematical Problems in Engineering, vol. 2021, pp. 1-7, 2021.
• BOZKURT F., “Derin Öğrenme Tekniklerini Kullanarak Akciğer X-Ray Görüntülerinden COVID-19 Tespiti.” European Journal of Science and Technology, pp. 149–56. Crossref, doi:10.31590/ejosat.898385, (2021).
• “MobileNet-v2 Convolutional Neural Network.” MathWorks, www.mathworks.com/help/deeplearning/ref/mobilenetv2.html#mw_609c1852-ea25-4857-9b3d-cd7d1916d5ff_sep_mw_6dc28e13-2f10-44a4-9632-9b8d43b376fe. Accessed 20 May 2021.
• NAHZAT S., & YAĞANOĞLU M., “Diabetes Prediction Using Machine Learning Classification Algorithms.” European Journal of Science and Technology. Crossref, doi:10.31590/ejosat.899716, (2021).
• Karthiyayini, J., “Fused Convolutional Neural Network for White Blood Cell Detection.” International Journal for Research in Applied Science and Engineering Technology, 8(5), 2040–2043, (2020), https://doi.org/10.22214/ijraset.2020.5334
• Jeyavathana, R. B., & R.Balasubramanian, P. J. “Estimation of White Blood Cells using Convolutional Neural Network.” International Journal of Engineering and Advanced Technology, 9(1), 452–454, (2019), https://doi.org/10.35940/ijeat.a9499.109119
• Su, M. C., Cheng, C. Y., & Wang, P. C., “A Neural-Network-Based Approach to White Blood Cell Classification.” The Scientific World Journal, 2014, 1–9., (2014), https://doi.org/10.1155/2014/796371