scholarly journals On the Effectiveness of Leukocytes Classification Methods in a Real Application Scenario

AI ◽  
2021 ◽  
Vol 2 (3) ◽  
pp. 394-412
Author(s):  
Andrea Loddo ◽  
Lorenzo Putzu

Automating the analysis of digital microscopic images to identify the cell sub-types or the presence of illness has assumed a great importance since it aids the laborious manual process of review and diagnosis. In this paper, we have focused on the analysis of white blood cells. They are the body’s main defence against infections and diseases and, therefore, their reliable classification is very important. Current systems for leukocyte analysis are mainly dedicated to: counting, sub-types classification, disease detection or classification. Although these tasks seem very different, they share many steps in the analysis process, especially those dedicated to the detection of cells in blood smears. A very accurate detection step gives accurate results in the classification of white blood cells. Conversely, when detection is not accurate, it can adversely affect classification performance. However, it is very common in real-world applications that work on inaccurate or non-accurate regions. Many problems can affect detection results. They can be related to the quality of the blood smear images, e.g., colour and lighting conditions, absence of standards, or even density and presence of overlapping cells. To this end, we performed an in-depth investigation of the above scenario, simulating the regions produced by detection-based systems. We exploit various image descriptors combined with different classifiers, including CNNs, in order to evaluate which is the most suitable in such a scenario, when performing two different tasks: Classification of WBC subtypes and Leukaemia detection. Experimental results have shown that Convolutional Neural Networks are very robust in such a scenario, outperforming common machine learning techniques combined with hand-crafted descriptors. However, when exploiting appropriate images for model training, even simpler approaches can lead to accurate results in both tasks.

White blood cell (Leukocytes) is made up of bone marrow located in the blood and lymph tissue. They are portion of the human body’s immune system, thereby helping the body system to fight against infection and other related diseases. The number of leukocytes in the blood is usually part of a complete blood cell (CBC) test, which may be used to check for conditions such as infection, inflammation, allergies, and leukemia. Automation of variance count of leukocytes offers valuable information to medical pathologist to diagnose and treat of many blood based diseases. Early characterization and classification of blood sample is a major lacuna in the medical field, giving rise to lots of challenges for pathologist to adequately predict blood based disease. Several successful efforts have been made to address the aforementioned challenges with the use of machine learning generally and Convolution Neural Network in particular. However the processor configuration which can result in real time, and accurate classification of the high dimensional pattern is imminent, and a vast number of researchers are not explicit on the system configuration used to obtain the result in their report, which is the crux of this research. In this research,12,500 augment images of blood cells was obtained from the Kaggle Repository online. The leukocytes are contained in the blood smear image and categorized into five major types of their types: Neutrophil, Eosinophil, Basophil, Lymphocyte and Monocyte. The color, geometric and texture features are used by the pathologists to differentiate the leukocytes. The Simulation was done using python programing language and python libraries including Keras, pandas, sklearn, numpy, scipy and matplot for potting of graphs of results. The simulation was done on both CPU and GPU processor to compare the performance of the processors on CNNs based classification of the data. While CPU has faster clock speed GPU has more cores. Hence the evaluation metrics used which are precision, specificity, sensitivity, training accuracy and validation accuracy revealed that GPU processor outperforms CPU in terms of the stated metrics of comparison. Therefore a high configuration processor (GPU), which handles graphics better is recommended for processing image data that involves the use of machine learning techniques


Author(s):  
Padmavathi .S ◽  
M. Chidambaram

Text classification has grown into more significant in managing and organizing the text data due to tremendous growth of online information. It does classification of documents in to fixed number of predefined categories. Rule based approach and Machine learning approach are the two ways of text classification. In rule based approach, classification of documents is done based on manually defined rules. In Machine learning based approach, classification rules or classifier are defined automatically using example documents. It has higher recall and quick process. This paper shows an investigation on text classification utilizing different machine learning techniques.


2021 ◽  
Vol 23 ◽  
pp. 100545
Author(s):  
Israel Elujide ◽  
Stephen G. Fashoto ◽  
Bunmi Fashoto ◽  
Elliot Mbunge ◽  
Sakinat O. Folorunso ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document