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

Abstract

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.

Explainable CAD System for Classification of Acute Lymphoblastic Leukemia Based on a Robust White Blood Cell Segmentation

“…However, as outlined in the Related Work ( Section 1 ) and Methods ( Section 3 ) sections, we go a step further by extending our analysis beyond conventional metrics. We incorporate Explainable AI (XAI) and clustering space analysis to affirm the robustness and reliability of our model [ 5 , 7 ].…”

“…The growing spectrum of diseases and the potential of Computer Diagnosis have sparked intense research into white blood cell (WBC) segmentation and leukemia classification. Propelled by progress in computer vision and Deep Learning, considerable strides have been taken in addressing the challenges intrinsic to WBC nuclei segmentation and leukemia classification [ 6 , 7 ].…”

White blood cells segmentation and classification using a random forest and residual networks implementation

“…The great majority of methods described do not propose both segmentation and classification, and no one addressed the model degradation or the overfitting of larger models in the white blood cell count problem, which has only five classes. [30][31][32][33] Here we present a comparison between different ResNet models, ranging from 18 to 152 layers, using both standard (18 and 34 layers) and bottleneck convolutions (50, 101, and 152). We can see from the learning curves (Fig.…”

A Chronological Overview of Using Deep Learning for Leukemia Detection: A Scoping Review

“…Findings from seven articles published in 2020-2021 were included in this review [23][24][25][26][27][28][29]. Four articles published in 2020 used different models to achieve high detection accuracy [23][24][25][26] and grayscale conversion and a non-specified CNN (96.78% accuracy) [23] by (1) applying noise reduction in grayscale images with three CNN frameworks (normal group accuracy of 90% and leukemia detection accuracy of 99%) [24], (2) using a dye-sensitized solar cell (DSSCS) CNN model with novel techniques for noise suppression doing image segmentation and color normalization (reached 97.00% accuracy) [25], and (3) using color normalization only (98%) [26].…”

“…Image augmentation has proven especially helpful by allowing datasets to be augmented from only a few images, increasing the dataset pool without needing more bone marrow samples, which is useful where large patient datasets are difficult to find. This could contribute to more efficient and cost-effective research and development of neural network models [28][29][30][31][32][33][34][35][36].…”