sis 18: e12

Research Article

Feature Extraction using CNN for Peripheral Blood Cells Recognition

Download41 downloads
  • @ARTICLE{10.4108/eai.20-10-2021.171548,
        author={Mohammed Ammar and Mostafa El Habib Daho and Khaled Harrar and Amel Laidi},
        title={Feature Extraction using CNN for Peripheral Blood Cells Recognition},
        journal={EAI Endorsed Transactions on Scalable Information Systems: Online First},
        keywords={Peripheral Blood Cells, CNN, Feature extraction, SVM, KNN, AdaboostM1},
  • Mohammed Ammar
    Mostafa El Habib Daho
    Khaled Harrar
    Amel Laidi
    Year: 2021
    Feature Extraction using CNN for Peripheral Blood Cells Recognition
    DOI: 10.4108/eai.20-10-2021.171548
Mohammed Ammar1,*, Mostafa El Habib Daho2, Khaled Harrar1, Amel Laidi3
  • 1: LIST Laboratory, University M’Hamed Bougara, Boumerdes, Algeria
  • 2: Biomedical Engineering Laboratory, University of Tlemcen, Tlemcen, Algeria
  • 3: LIMOSE Laboratory M’Hamed Bougara University Boumerdes, Algeria
*Contact email:


INTRODUCTION: The diagnosis of hematological diseases is based on the morphological differentiation of the peripheral blood cell types.

OBJECTIVES: In this work, a hybrid model based on CNN features extraction and machine learning classifiers were proposed to improve peripheral blood cell image classification.

METHODS: At first, a CNN model composed of four convolution layers and three fully connected layers was proposed. Second, the features from the deeper layers of the CNN classifier were extracted. Third, several models were trained and tested on the data. Moreover, a combination of CNN with traditional machine learning classifiers was carried out. This includes CNNKNN, CNNSVM (Linear), CNNSVM (RBF), and CNNAdaboostM1. The proposed methods were validated on two datasets. We have used a public dataset containing 12444 images with four types of leukocytes to find the best optimizer function(eosinophil, lymphocyte, monocyte, and neutrophil images). The second dataset contains 17,092 images divided into eight groups: lymphocytes, neutrophils, monocytes). the second public dataset was used to find the best combination of CNN and the machine learning algorithms. the dataset containing 17,092 images: lymphocytes, neutrophils, monocytes, eosinophils, basophils, immature granulocytes, erythroblasts, and platelets.

RESULTS: The results reveal that CNN combined with AdaBoost decision tree classifier provided the best performance in terms of cells recognition with an accuracy of 88.8%, demonstrating the performance of the proposed approach.

CONCLUSION: The obtained results show that the proposed system can be used in clinical practice.