Automatic classification of cells in microscopic fecal images using convolutional neural networks

Abstract The analysis of fecal-type components for clinical diagnosis is important. The main examination involves the counting of red blood cells (RBCs), white blood cells (WBCs), and molds under the microscopic. With the development of machine vision, some vision-based detection schemes have been proposed. However, these methods have a single target for detection, with low detection efficiency and low accuracy. We proposed an algorithm to identify the visible image of fecal composition based on intelligent deep learning. The algorithm mainly includes region proposal and candidate recognition. In the process of segmentation, we proposed a morphology extraction algorithm in a complex background. As for the candidate recognition, we proposed a new convolutional neural network (CNN) architecture based on Inception-v3 and principal component analysis (PCA). This method achieves high-average Precision of 90.7%, which is better than the other mainstream CNN models. Finally, the images within the rectangle marks were obtained. The total time for detection of an image was roughly 1200 ms. The algorithm proposed in the present paper can be integrated into an automatic fecal detection system.

Download Full-text

New Segmentation and Feature Extraction Algorithm for the Classification of White Blood Cells in Peripheral Smear Images

10.1101/2021.04.29.441751 ◽

2021 ◽

Author(s):

Eslam Tavakoli ◽

Ali Ghaffari ◽

Seyedeh-Zahra Mousavi Kouzehkanan ◽

Reshad Hosseini

Keyword(s):

Blood Cells ◽

White Blood Cells ◽

Dice Similarity Coefficient ◽

Second Phase ◽

Peripheral Smear ◽

Color Features ◽

Svm Model ◽

Extraction Algorithm ◽

Processing Techniques

This article addresses a new method for classification of white blood cells (WBCs) using image processing techniques and machine learning methods. The proposed method consists of three steps: detecting the nucleus and cytoplasm, extracting features, and classification. At first, a new algorithm is designed to segment the nucleus. For the cytoplasm to be detected, only a part of it which is located inside the convex hull of the nucleus is involved in the process. This attitude helps us overcome the difficulties of segmenting the cytoplasm. In the second phase, three shape and four novel color features are devised and extracted. Finally, by using an SVM model, the WBCs are classified. The segmentation algorithm can detect the nucleus with a dice similarity coefficient of 0.9675. The proposed method can categorize WBCs in Raabin-WBC, LISC, and BCCD datasets with accuracies of 94.47 %, 92.21 %, and 94.20 %, respectively. It is worth mentioning that the hyperparameters of the classifier are fixed only with Raabin-WBC dataset, and these parameters are not readjusted for LISC and BCCD datasets. The obtained results demonstrate that the proposed method is robust, fast, and accurate.

Download Full-text

ROBUST DISCRIMINATION OF LEUKOCYTES PROTUBERANT TYPES FOR EARLY DIAGNOSIS OF LEUKEMIA

Journal of Mechanics in Medicine and Biology ◽

10.1142/s0219519419500556 ◽

2019 ◽

Vol 19 (06) ◽

pp. 1950055 ◽

Cited By ~ 7

Author(s):

ISRA NAZ ◽

NAZEER MUHAMMAD ◽

MUSSARAT YASMIN ◽

MUHAMMAD SHARIF ◽

JAMAL HUSSAIN SHAH ◽

...

Keyword(s):

Early Diagnosis ◽

Blood Cells ◽

Detection System ◽

White Blood Cells ◽

Training Model ◽

Blood Film ◽

Morphological Examination ◽

Efficient Detection ◽

Laboratory Technology

Advanced laboratory technology has made blood testing more automated, robust and tests are being implemented comprehensively. Leukocytes type differentiation is a critical hematological images analysis step of blood film as it delivers valuable information in diagnosis of several diseases. At present, the morphological examination of leukocytes is done manually and this process is very tedious, inefficient and slow. Although many white blood cells detection or classification techniques are presented by different researchers, there is still a need of fully automated and an efficient detection system of blood cells with its particular types for an early diagnosis of leukemia. This paper presents a technique for the classification of protuberant types of leukocytes and early diagnosis of leukemia. The work is divided into the following main stages: (a) image augmentation, (b) wavelet composition and decomposition for attaining high and low frequency bands of the cell image, (c) convolutional neural network (CNN) training model for the classification of leukocytes categories and (d) prediction of leukemia. The main intention behind this study is to develop an automated, robust and efficient classification and detection system of leukocytes for microscopic blood images.

Download Full-text

New Segmentation and Feature Extraction Algorithm for Classification of White Blood Cells in Peripheral Smear Images

10.21203/rs.3.rs-496778/v1 ◽

2021 ◽

Author(s):

Eslam Tavakoli ◽

Ali Ghaffari ◽

Zahra Mousavi Kouzehkanan ◽

Reshad Hosseini

Keyword(s):

Blood Cells ◽

White Blood Cells ◽

Dice Similarity Coefficient ◽

Second Phase ◽

Peripheral Smear ◽

Color Features ◽

Svm Model ◽

Extraction Algorithm ◽

Processing Techniques

Abstract This article addresses a new method for the classification of white blood cells (WBCs) using image processing techniques and machine learning methods. The proposed method consists of three steps: detecting the nucleus and cytoplasm, extracting features, and classification. At first, a new algorithm is designed to segment the nucleus. For the cytoplasm to be detected, only a part of it which is located inside the convex hull of the nucleus is involved in the process. This attitude helps us overcome the difficulties of segmenting the cytoplasm. In the second phase, three shapes and four novel color features are devised and extracted. Finally, by using an SVM model, the WBCs are classified. The segmentation algorithm can detect the nucleus with a dice similarity coefficient of 0.9675. The proposed method can categorize WBCs in Raabin-WBC, LISC, and BCCD datasets with accuracies of 94.65 %, 92.21 %, and 94.20 %, respectively. It is worth mentioning that the hyperparameters of the classifier are fixed only with the Raabin-WBC dataset, and these parameters are not readjusted for LISC and BCCD datasets. The obtained results demonstrate that the proposed method is robust, fast, and accurate.

Download Full-text

New segmentation and feature extraction algorithm for classification of white blood cells in peripheral smear images

Scientific Reports ◽

10.1038/s41598-021-98599-0 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Sajad Tavakoli ◽

Ali Ghaffari ◽

Zahra Mousavi Kouzehkanan ◽

Reshad Hosseini

Keyword(s):

Blood Cells ◽

White Blood Cells ◽

Dice Similarity Coefficient ◽

Second Phase ◽

Peripheral Smear ◽

Color Features ◽

Svm Model ◽

Extraction Algorithm ◽

Processing Techniques

AbstractThis article addresses a new method for the classification of white blood cells (WBCs) using image processing techniques and machine learning methods. The proposed method consists of three steps: detecting the nucleus and cytoplasm, extracting features, and classification. At first, a new algorithm is designed to segment the nucleus. For the cytoplasm to be detected, only a part of it located inside the convex hull of the nucleus is involved in the process. This attitude helps us overcome the difficulties of segmenting the cytoplasm. In the second phase, three shapes and four novel color features are devised and extracted. Finally, by using an SVM model, the WBCs are classified. The segmentation algorithm can detect the nucleus with a dice similarity coefficient of 0.9675. The proposed method can categorize WBCs in Raabin-WBC, LISC, and BCCD datasets with accuracies of 94.65%, 92.21%, and 94.20%, respectively. Besides, we show that the proposed method possesses more generalization power than pre-trained CNN models. It is worth mentioning that the hyperparameters of the classifier are fixed only with the Raabin-WBC dataset, and these parameters are not readjusted for LISC and BCCD datasets.

Download Full-text

Classification of White Blood Cells Based on Cell Diameter, Specific Membrane Capacitance and Cytoplasmic Conductivity Leveraging Microfluidic Constriction Channel

2021 21st International Conference on Solid-State Sensors, Actuators and Microsystems (Transducers) ◽

10.1109/transducers50396.2021.9495490 ◽

2021 ◽

Author(s):

Huiwen Tan ◽

Minruihong Wang ◽

Yi Zhang ◽

Xukun Huang ◽

Deyong Chen ◽

...

Keyword(s):

Blood Cells ◽

White Blood Cells ◽

Membrane Capacitance ◽

Cell Diameter ◽

Specific Membrane

Download Full-text

Association between serum markers of the humoral immune system and inflammation in the Swedish AMORIS study

10.21203/rs.3.rs-27195/v2 ◽

2020 ◽

Author(s):

aida santaolalla ◽

Sam Sollie ◽

Ali Rislan ◽

Debra H. Josephs ◽

Niklas Hammar ◽

...

Keyword(s):

Immune System ◽

Educational Level ◽

Blood Cells ◽

Adaptive Response ◽

White Blood Cells ◽

Principal Component ◽

Serum Markers ◽

Response Component ◽

Humoral Immune ◽

Humoral Immune System

Abstract Background: Although the onset of inflammatory cascades may profoundly influence the nature of antibody responses, the interplay between inflammatory and humoral (antibody) immune markers remains unclear. Thus, we explored the reciprocity between the humoral immune system and inflammation and assessed how external socio-demographic factors may influence these interactions.Methods: From the AMORIS cohort, 5,513 individuals were identified with baseline measurements of serum humoral immune (immunoglobulin G, A & M (IgG, IgA, IgM)) and inflammation (C-reactive protein (CRP), albumin, haptoglobin, white blood cells (WBC), iron and total iron-binding capacity) markers measured on the same day. Correlation analysis, principal component analysis and hierarchical clustering were used to evaluate biomarkers correlation, variation and associations. Multivariate analysis of variance was used to assess associations between biomarkers and educational level, socio-economic status, sex and age.Results: Frequently used serum markers for inflammation, CRP, haptoglobin and white blood cells, correlated together. Hierarchical clustering and principal component analysis confirmed the interaction between these main biological responses, showing an acute response component (CRP, Haptoglobin, WBC, IgM) and adaptive response component (Albumin, Iron, TIBC, IgA, IgG). A socioeconomic gradient associated with worse health outcomes was observed, specifically low educational level, older age and male sex were associated with serum levels that indicated infection and inflammation.Conclusions: These findings indicate that serum markers of the humoral immune system and inflammation closely interact in response to infection or inflammation. Clustering analysis presented two main immune response components: an acute and an adaptive response, comprising markers of both biological pathways. Future studies should shift from single internal marker assessment to multiple humoral and inflammation serum markers combined, when assessing risk of clinical outcomes such as cancer.

Download Full-text

Klasifikasi Sel Darah Putih dan Sel Limfoblas Menggunakan Metode Multilayer Perceptron Backpropagation

IJEIS (Indonesian Journal of Electronics and Instrumentation Systems) ◽

10.22146/ijeis.49943 ◽

2019 ◽

Vol 9 (2) ◽

pp. 173

Author(s):

Apri Nur Liyantoko ◽

Ika Candradewi ◽

Agus Harjoko

Keyword(s):

Blood Cell ◽

White Blood Cell ◽

Blood Cells ◽

White Blood Cells ◽

Back Propagation ◽

Cell Types ◽

Diagnostic Process ◽

The Subject ◽

Processing Techniques

Leukemia is a type of cancer that is on white blood cell. This disease are characterized by abundance of abnormal white blood cell called lymphoblast in the bone marrow. Classification of blood cell types, calculation of the ratio of cell types and comparison with normal blood cells can be the subject of diagnosing this disease. The diagnostic process is carried out manually by hematologists through microscopic image. This method is likely to provide a subjective result and time-consuming.The application of digital image processing techniques and machine learning in the process of classifying white blood cells can provide more objective results. This research used thresholding method as segmentation and multilayer method of back propagation perceptron with variations in the extraction of textural features, geometry, and colors. The results of segmentation testing in this study amounted to 68.70%. Whereas the classification test shows that the combination of feature extraction of GLCM features, geometry features, and color features gives the best results. This test produces an accuration value 91.43%, precision value of 50.63%, sensitivity 56.67%, F1Score 51.95%, and specitifity 94.16%.

Download Full-text

Segmentation of Leukemia Cells Using Clustering

International Journal of Synthetic Emotions ◽

10.4018/ijse.2019070103 ◽

2019 ◽

Vol 10 (2) ◽

pp. 39-48

Author(s):

Eman Mostafa ◽

Heba A. Tag El-Dien

Keyword(s):

Bone Marrow ◽

Cancer Cells ◽

Blood Cells ◽

White Blood Cells ◽

Leukemia Cells ◽

Microscopic Image ◽

Processing Stage ◽

Clustering Techniques ◽

Fuzzy C Means

Leukemia is a blood cancer which is defined as an irregular augment of undeveloped white blood cells called “blasts.” It develops in the bone marrow, which is responsible for blood cell generation including leukocytes and white blood cells. The early diagnosis of leukemia greatly helps in the treatment. Accordingly, researchers are interested in developing advanced and accurate automated techniques for localizing such abnormal blood cells. Subsequently, image segmentation becomes an important image processing stage for successful feature extraction and classification of leukemia in further stages. It aims to separate cancer cells by segmenting the microscopic image into background and cancer cells that are known as the region of interested (ROI). In this article, the cancer blood cells were segmented using two separated clustering techniques, namely the K-means and Fuzzy-c-means techniques. Then, the results of these techniques were compared to in terms of different segmentation metrics, such as the Dice, Jac, specificity, sensitivity, and accuracy. The results proved that the k-means provided better performance in leukemia blood cells segmentation as it achieved an accuracy of 99.8% compared to 99.6% with the fuzzy c-means.

Download Full-text