Fig 8 - uploaded by Aiman Al-Sabaawi
Content may be subject to copyright.
![Example of WBC types that have a big nucleus that covers the whole cell such basophil cell. Neutrophil has a segmented nucleus that does not cover the cytoplasm [102].](profile/Aiman-Al-Sabaawi/publication/342430129/figure/fig8/AS:1140869512073219@1649015896296/Example-of-WBC-types-that-have-a-big-nucleus-that-covers-the-whole-cell-such-basophil.jpg)
Example of WBC types that have a big nucleus that covers the whole cell such basophil cell. Neutrophil has a segmented nucleus that does not cover the cytoplasm [102].
Source publication
Segmentation of white blood cells in digital haematology microscope images represents one of the major tools in the diagnosis and evaluation of blood disorders. Pathological examinations are being the gold standard in many haematology and histophathology, and also play a key role in the diagnosis of diseases. In clinical diagnosis, white blood cell...
Context in source publication
Context 1
... The relative sizes of the cell and nucleus differ widely; for example, some cells do not have a nucleus and other cells have a big nucleus that covers the whole cell. This introduces difficulties for techniques that segment and detect the WBC and its nucleus and cytoplasm together, as shown in Fig-8. 7) Techniques developed using a small or specific database in histopathology might not be as effective when applied to a larger, or more general, publicly available database. ...
Citations
... Recently, many technological advances have enabled us to automate procedures related to the segmentation, detec-B André Ricardo Backes arbackes@yahoo.com.br 1 Department of Computing, Federal University of São Carlos, 13565-905 São Carlos, SP, Brazil tion, and classification of microscopic images. Nowadays, these advances, especially in the field of digital pathology, have made it easy to connect and use digital cameras with microscopes, thus improving the image acquisition process, with results in image quality and resolution, attributes that are essential for a proper image analysis [1]. ...
... Recently, many technological advances have enabled us to automate procedures related to the segmentation, detec-B André Ricardo Backes arbackes@yahoo.com.br 1 Department of Computing, Federal University of São Carlos, 13565-905 São Carlos, SP, Brazil tion, and classification of microscopic images. Nowadays, these advances, especially in the field of digital pathology, have made it easy to connect and use digital cameras with microscopes, thus improving the image acquisition process, with results in image quality and resolution, attributes that are essential for a proper image analysis [1]. As a result, the development of computer vision methods for these tasks has seen a surge in recent years. ...
In this paper, we address the problem of detecting and segmenting leukocytes in images. These cells are an important part of the immune system and some of their attributes (such as the amount and aspect of the cells) are used to detect several diseases (e.g., leukemia). To accomplish this task, we used the deep learning architecture named U-Net, a commonly used segmentation network originally developed aiming at biomedical image segmentation. Since the background in leukocytes images is more constant than in other segmentation tasks, i.e., there is little variety of undesired objects, we opted to use a personalized version of the network and we evaluated the impact of using different combinations of convolutional blocks and filters to build the network model. We compared our network model with different approaches found in the literature using an image dataset containing images of leukocytes and other blood structures. Results demonstrated the superiority of our approach in terms of Jaccard index and, for this given problem, the number of blocks is more important than the total number of trainable parameters of the U-Net.
... The results of analysis in the form of images are helpful for the doctors in identifying the defected region of the body, disease detection, spread of disease into nearby tissues and treatment strategy of the patients. Image processing techniques and algorithms can also be applied for the detection of nucleus and cytoplasm of the white blood cells to help out the doctors finding blood diseases [31]. ...
In computer vision, the goal of object identification in images is to track or count objects such as persons, cars, animals and so on. Object detection techniques have found their place in several applications such as video analytic, contact-less checkouts, inventory management, defect detection, productivity improvements, landscape patch analysis and many more. One of the most common techniques for image object identification in binary images is the four and eight-neighbourhood of pixels in the images. The image object identification mechanisms based on the four or eight-neighbour (more generally, the k nearest neighbours) techniques identify objects in the images but there is no guarantee of their correctness. To the best of our knowledge, there is no image object identification tool with proof that it correctly identifies the image objects. To meet the high standards of correctness, we provide a formal foundation to allow proof of correctness of the image object identification tools based on the four or eight-neighbourhood techniques. As a proof of concept, an object identification tool is developed by implementing both, the four neighbour and eight neighbour, techniques in a general purpose programming language. The tool is experimentally evaluated by running it against real life RGB images. Furthermore, the correctness of the objects identified by the prototype tool are checked by encoding the objects in the formal notations and running the predicate over each object using the type checker program of the Coq tool.
... PBF is made of fresh drop of blood that is acquired from a syringe or a finger stick puncture and then it is smeared on one glass side to make a blood film of length 2.5 cm [1]- [4]. WBCs have five different kinds (i.e., neutrophils, basophils, monocytes, eosinophils, and lymphocytes) and each kind is responsible for a specific job in the immune system [5]- [7]. Additionally, each kind has its own characteristics in terms of morphology, cell size, and nucleus size. ...
Peripheral blood film (PBF) plays a vital role in diagnosis of various blood diseases based on white blood cells (WBCs) count. However, such a diagnosis method is highly dependent on the hematologist’s skill, therefore, it has a varying accuracy. Thus, the aim of the proposed study is to provide a new optical technique to visualize and discriminate between normal and cancer lymphocyte WBCs. The proposed study is based on computer generated holography which is capable of providing precise representation of WBCs floating in the air. To accomplish the purpose of the study, the lymphocyte WBCs have been segmented from microscopic images using in house-developed MATLAB code. The accuracy of the obtained segmentation results is evaluated by comparing the results of our segmentation method and the results of the open-source CellProfiler software. Then, the associated computer-generated holograms (CGHS) for the segmented lymphocytes are calculated and optimized based on the iterative Fourier transform algorithm (IFTA). To solve the problem of speckle noise which is generated in the optical reconstruction of the phase-only holograms, a speckle noise reduction algorithm based on temporal multiplexing of spatial frequencies is utilized. Lastly, the calculated CGHs are displayed on a phase-only spatial light modulator (SLM) for the optical reconstruction. The results revealed that the difference between normal and cancer WBCs can easily visualized using holographic projection technique. Thus, the proposed technique can be used as a helpful tool for visualization and interpretation of lymphocytes WBCs.
... The final type is platelets which contribute to the process of blood clotting in case of bleeding [3]. Experts can segment and classify the blood cells based on the morphological features, including shape, size and nucleus position whether or not they contain a nucleus, color of nucleus and cytoplasm [4]. Previously, the pathologists used the traditional procedure of blood cell analysis-based microscope properties. ...
Blood cell detection considers a gold standard key in diagnosing blood disease and producing automatic reports to hematologists and doctors. Blood cell detection is a challenging task due to non-illumination level, high number of overlapped cells per image, variations in cell densities among platelets, white blood cells and red blood cells, and the variety of staining process. Traditional procedure of blood cell detection requires pathologist effort and time. In computer aided diagnosis, machine learning and deep learning techniques become the practical way to automate the procedure of diagnosing, classify microscopic blood cells, and increase the accuracy and speed of the procedure. This paper provides a review of the detection and classification of blood cell, including red blood cells, white blood cells and platelets and their characteristics using machine learning techniques. We also have detailed the dataset of microscope blood cell. We have divided the previous works into four categories based on the output of the models, including pre-processing, segmentation, feature extraction and classification. Then, we discuss the challenges that face these methods and suggest the potential future techniques.
... The diameter of platelets ranges between 2-4 µ, and blood cell may contain 150,000-350,000 platelets. The platelet contributes to the process of blood clotting in case of bleeding [1], [2]. ...
... This procedure has required tedious work, time-consuming and human effort. Recently, pathologists have used an automatic blood cell analysis, including image enhancement, segmentation, feature extraction, and classification, as shown in Fig-2 in red rectangle [2]. ...
Segmentation of blood cell considers a gold standard key in diagnosing blood disease and producing automatic reports to haematologists and doctors. Blood cell segmentation is a challenging task due to non-illumination level, high number of overlapped cells per image, variations in cell densities among platelets, white blood cells and red blood cells, and the variety of staining process. In this paper, level set-based force in normal direction via edge detection of cell shape is proposed , for first time, to segment and split overlapped blood cell components, and deal with these challenges. The proposed method has used publicly CellaVision DM96 dataset. It is also compared with other existing segmentation methods. The result has shown that this method reduces the effects non-illumination and variation in cell densities. An extensive experiment demonstrates that the proposed method performance is superior against all other existing blood cell segmentation approaches except for one method in this paper which is slightly better than our proposed method. The result also shows that the proposed method can support pathology practice in the diagnosis of autoimmune diseases in the future.
... Thus, it is vital for doctors to know the count of WBCs amongst the different categories to diagnose any specific disease or underlying health condition. Recently, due to the advent of machine learning and deep learning, there have been a plethora of methods developed for automated detection and classification of WBCs [13]. Most of these methods are based on blood smear microscopic images which provide rich features about the morphology and structure of the cells but require manual preparation and staining of the slides by trained personnel [16]. ...
Recently, a lot of automated white blood cells (WBC) or leukocyte classification techniques have been developed. However, all of these methods only utilize a single modality microscopic image i.e. either blood smear or fluorescence based, thus missing the potential of a better learning from multimodal images. In this work, we develop an efficient multimodal architecture based on a first of its kind multimodal WBC dataset for the task of WBC classification. Specifically, our proposed idea is developed in two steps - 1) First, we learn modality specific independent subnetworks inside a single network only; 2) We further enhance the learning capability of the independent subnetworks by distilling knowledge from high complexity independent teacher networks. With this, our proposed framework can achieve a high performance while maintaining low complexity for a multimodal dataset. Our unique contribution is two-fold - 1) We present a first of its kind multimodal WBC dataset for WBC classification; 2) We develop a high performing multimodal architecture which is also efficient and low in complexity at the same time.KeywordsWBCs classificationMultimodalKnowledge distillation
... Thus, it is vital for doctors to know the count of WBCs amongst the different categories to diagnose any specific disease or underlying health condition. Recently, due to the advent of machine learning and deep learning, there have been a plethora of methods developed for automated detection and classification of WBCs [13]. Most of these methods are based on blood smear microscopic images which provide rich features about the morphology and structure of the cells but require manual preparation and staining of the slides by trained personnel [16]. ...
Recently, a lot of automated white blood cells (WBC) or leukocyte classification techniques have been developed. However, all of these methods only utilize a single modality microscopic image i.e. either blood smear or fluorescence based, thus missing the potential of a better learning from multimodal images. In this work, we develop an efficient multimodal architecture based on a first of its kind multimodal WBC dataset for the task of WBC classification. Specifically, our proposed idea is developed in two steps - 1) First, we learn modality specific independent subnetworks inside a single network only; 2) We further enhance the learning capability of the independent subnetworks by distilling knowledge from high complexity independent teacher networks. With this, our proposed framework can achieve a high performance while maintaining low complexity for a multimodal dataset. Our unique contribution is two-fold - 1) We present a first of its kind multimodal WBC dataset for WBC classification; 2) We develop a high performing multimodal architecture which is also efficient and low in complexity at the same time.
... Deep learning-based models have revolutionized the imaging and clinical sector with their accurate and efficient frameworks [13]- [15]. In [16] WBC segmentation is performed using CNN with transfer learning. For small cell cases, this method could not show good performance [16]. ...
... In [16] WBC segmentation is performed using CNN with transfer learning. For small cell cases, this method could not show good performance [16]. Another study [17] performed deep learning-based WBC segmentation. ...
White blood cells (WBCs), also known as leukocytes, are one of the valuable parts of the blood and immune system. Typically, pathologists use microscope for the manual inspection of blood smears which is a time-consuming, error-prone, and labor-intensive procedure. To address these issues, we present two novel shallow networks: a leukocyte deep segmentation network (LDS-Net) and leukocyte deep aggregation segmentation network (LDAS-Net) for the joint segmentation of cytoplasm and nuclei in WBC images. LDS-Net is a shallow architecture with three downsampling stages and seven convolution layers. LDAS-Net is an extended version of LDS-Net that utilizes a novel pool-less low-level information transfer bridge to transfer low-level information to the deep layers of the network. This information is aggregated with deep features in a dense feature concatenation block to achieve accurate cytoplasm and nuclei joint segmentation. We evaluated our developed architectures on four WBC publicly available datasets. For cytoplasmic segmentation in WBCs, the proposed method achieved the dice coefficients of 98.97%, 99.0%, 96.05%, and 98.79% on Datasets 1, 2, 3, and 4, respectively. For nuclei segmentation, the dice coefficients of 96.35% and 98.09% are achieved for Datasets 1 and 2, respectively. Proposed method outperforms state-of-the-art methods with superior computational efficiency and requires only 6.5 million trainable parameters.
... Hematologists have shown that, in addition to the nucleus' features, the color of the cells can reveal important details regarding the type of WBCs [37]. For instance, the nucleus of lymphocytes has more chromatin and is stained a deeper bluish-purple. ...
White blood cell (WBC) identification performance is highly correlated with the quality of the extracted features, with radiomic features having greater resolution and more detailed information and deep features having more robust semantic information. This research integrates these two aspects to creatively suggest a WBC classification model based on radiomics and deep learning. This research suggests a brand-new method for extracting radiomic features from WBC images as well as a dual-branch feature fusion network RCTNet based on CNN and Transformer for extracting deep features. The radiomic feature extraction method not only has a simple segmentation algorithm but also solves the problem of cell adhesion, can obtain higher quality shape, color, and texture features without segmenting intact cells, and is more generalizable. RCTNet can extract more critical and recognizable deep features from WBC nuclei, avoiding the influence of too much redundant and invalid information on the results, and has better performance than several existing CNN models. We compared the classification outcomes based on radiomics with those based on deep learning to confirm the efficacy of the WBC classification model based on radiomics and deep learning suggested in this research. The experimental results demonstrated that combining radiomic features and deep features significantly improved the classification accuracy, with an AUC exceeding 0.9995, accuracy, sensitivity, specificity, precision, and F1-score reaching 0.9880, 0.9823, 0.9883, 0.9968, and 0.9881, respectively. The model has significant research significance in clinical applications and aids physicians in improving diagnosis and screening for diseases of WBCs.
... The second group of features is color characteristics. According to the experience of hematologists, in addition to the shape features of the nucleus, the color features of the nucleus and the cytoplasm can also provide us with useful information about the type of WBC 11 . In this research, four novel color features by means of nucleus region, convex hull region, and ROC region are designed as follows: ...
... Then, the ground truths for these images were identified by an expert with the help of Easy-GT software 35 . Also, since very dark purple granules cover the basophil's surface, it is almost impossible to distinguish the nucleus 11 . Therefore, the whole basophil cell was considered as the ground truth. ...
... In order to evaluate the classification accuracy, four metrics are used: Precision, Sensitivity, F1-score (F1), and Accuracy (Acc). If we face a two-class classification problem such the first class is called Positive and the second class is called Negative, the confusion matrix can be assumed as Table 4, and the mentioned criteria are obtained through relations (8), (9), (10), and (11). ...
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 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.
